# UtilitiesLabs — Full Content Export > Independent guidance on utility ERP, CIS, and billing systems: SAP IS-U and S/4HANA Utilities, Oracle CC&B, and Cayenta CIS. Backed by AvanSaber. > Machine-readable full text of 92 published pages, generated at build time from https://www.utilitieslabs.com. Canonical source is the live site; each entry links its URL. --- ## How to Choose Utility CIS and Billing Software: A Selection Guide URL: https://www.utilitieslabs.com/a-comprehensive-guide-to-choosing-the-right-utilities-software/ # How to Choose Utility CIS and Billing Software: A Selection Guide AvanSaber Research/June 8, 2023/Updated September 24, 2026/12 min read Choosing utility CIS and billing software comes down to fit, not feature count. SAP, Oracle, and Cayenta can all run meter-to-cash for a utility. The right choice depends on your size, your rate complexity, the systems you already run, and who will actually implement it. The fastest way to make that decision defensible is a weighted RFP scorecard that every vendor is measured against, using your own tariffs and bills rather than a vendor demo script. This guide gives you that scorecard, a set of weights you can adapt, fit notes on how each vendor typically positions, the questions to put in your RFP, and the red flags that should slow a selection down. Vendor facts are sourced and dated as of September 2026. ## Who this guide is for This is written for the team running a CIS and billing replacement: the customer service, billing, finance, and IT leads, plus whoever owns procurement. It applies to: - Investor-owned utilities with large customer bases, complex tariffs, and regulatory reporting to a state commission. - Municipal and public power utilities, often multi-service (electric, water, wastewater, refuse) and tied to city finance and budgeting processes. - Electric cooperatives that need member billing, capital credits, and a lean IT team. - Water and wastewater districts with tiered and seasonal rates and growing AMI programs. The criteria are the same for all of them. The weights are what change. ## The evaluation criteria Eight criteria cover the decisions that make or break a CIS program. For each one, the question is what the vendor can prove with your data, not what the brochure says. - Billing complexity and rate engine. Can the system bill your hardest tariffs as configured, not custom coded? Test time-of-use, tiered, seasonal, demand, net metering, budget billing, and any bundled multi-service bill you issue today. - Meter-to-cash and AMI/MDM integration. How do interval reads get from the head-end system through validation, estimation, and editing (VEE) into billing? Is meter data management part of the same platform, a sibling product from the same vendor, or a third-party integration? - Customer self-service. Portal, mobile, payments, start-stop-transfer, outage and usage information, and increasingly AI agents that answer account questions. - Deployment model and cloud path. SaaS, single-tenant hosted, private cloud, or on-premises, and what the upgrade model looks like five years after go-live. - Implementation timeline and partner ecosystem. Who does the work, how many of their people have delivered this product at a utility like yours, and what a realistic plan looks like. - Total cost of ownership drivers. Subscription or license, implementation services, integrations, internal staff backfill, testing, hosting, and ongoing support over the full term. - Regulatory and reporting fit. Commission reporting, disconnection and payment arrangement rules, low-income programs, and, for public power, fund accounting and budget reporting. - Data migration. How much history converts, how many mock conversions are planned, and how bills from the new system are reconciled against the old one before cutover. ## A weighting you can adapt Weights should be agreed by the steering group before any vendor presents. A baseline that works for most utilities, out of 100 points: | Criterion | Baseline weight (points) | Raise it when | Lower it when | Billing complexity and rate engine | 20 | You have many rate classes, demand or TOU rates, or frequent rate cases | Your tariffs are simple and stable | Meter-to-cash and AMI/MDM integration | 15 | AMI is live or funded, or MDM is also being replaced | Meters are read manually and AMI is years away | Customer self-service | 10 | Call volumes are high or digital adoption is a board goal | A separate portal contract is staying in place | Deployment model and cloud path | 10 | Your IT team is small or your data center is aging | Policy requires on-premises hosting | Implementation timeline and partner ecosystem | 15 | You have a hard deadline, such as end of vendor support | You have time and strong internal staff | Total cost of ownership drivers | 15 | Rates are politically sensitive or budgets are fixed | Capital is available and scope is the priority | Regulatory and reporting fit | 5 | You face heavy commission reporting or complex market rules | Reporting is mostly internal | Data migration | 10 | Your legacy system is heavily customized or history is required | You are converting from a clean, recent system A large investor-owned utility in a competitive retail market might move points into billing and regulatory fit. A multi-service municipal utility might add a criterion for integrated financials and work management and fund it from self-service. What matters is that the weights are written down first, so the scores cannot be tuned afterward to fit a favorite. ## The RFP scorecard template Score each criterion from 1 to 5, multiply by its weight, and total the result. Define the scale in the RFP so every evaluator scores the same way: - 5: Demonstrated with your data in a scripted demo or sandbox, and confirmed by a reference. - 4: Demonstrated with your data, with minor configuration gaps. - 3: Demonstrated with sample data, or delivered through a standard integration. - 2: Possible only with custom development or a third-party product not in the bid. - 1: Not supported, or the vendor could not show it. | Criterion | Weight | SAP score (1 to 5) | Oracle score (1 to 5) | Cayenta score (1 to 5) | Evidence required | Billing complexity and rate engine | 20 | | | | Your top ten hardest bills recalculated live | Meter-to-cash and AMI/MDM integration | 15 | | | | Interval reads from your head-end through VEE to a bill | Customer self-service | 10 | | | | Start-stop-transfer and payment on the proposed portal | Deployment model and cloud path | 10 | | | | Written upgrade and release policy | Implementation timeline and partner ecosystem | 15 | | | | Named team, resumes, and a comparable reference | Total cost of ownership drivers | 15 | | | | Ten-year cost model on a common template | Regulatory and reporting fit | 5 | | | | Your commission reports produced from test data | Data migration | 10 | | | | Conversion plan with mock cycles and reconciliation | Weighted total | 100 | | | | Keep the evidence column. It turns the scorecard from an opinion poll into an audit trail your board, council, or commission can follow. ## How each vendor typically positions These are fit notes, not scores. Your scorecard should produce the scores. ### SAP: S/4HANA Utilities (and IS-U on ECC) SAP’s utilities functionality runs inside the ERP itself. SAP’s own training material describes S/4HANA Utilities as covering the chain from metering and consumption determination to invoice creation and customer payment, with components for meter reading, device management, energy data management, billing, contract accounting (the receivables subledger usually called FI-CA), and customer engagement. That tight coupling to SAP finance and logistics is the main reason SAP-estate utilities start here. For the full module map, see our SAP IS-U guide. Timing is the other driver. SAP committed in February 2020 to mainstream maintenance for SAP Business Suite 7 core applications until the end of 2027, with optional extended maintenance until the end of 2030. In August 2025 SAP added a time-bound SAP ERP, private edition, transition option covering 2031 to 2033, aimed at its largest and most complex ERP customers, with conditions that include moving to SAP ERP, private edition on SAP HANA before the end of 2030 and a minimum system size of 2 TB. A utility on IS-U today is therefore choosing a platform on a clock, whichever vendor it picks. On the cloud path, SAP’s newer utility capabilities point to its private cloud: the Utilities Customer Self-Service Agent in SAP’s Q4 2025 Business AI release highlights draws on contracts, tariffs, and consumption data through its integration with SAP S/4HANA Cloud Private Edition. SAP also documents S/4HANA Utilities under its on-premise edition, which the utility or its hosting partner installs and runs. ### Oracle: Customer Cloud Service and Customer Care and Billing Oracle offers its utility customer platform in two deployment forms. Its CIS product page describes Oracle Utilities Customer Cloud Service as the SaaS option and Oracle Utilities Customer to Meter as the on-premises option, both combining meter data management, customer care, and billing on a shared database. Oracle also highlights point-and-click rate configuration with no coding required, self-service for start-stop-transfer and payments, and updates delivered in small releases throughout the year. The same page says its accelerated implementation package delivers the CIS “in months, not years,” which is a vendor claim to test against references. Oracle also documents Customer Care and Billing Cloud Service as a billing system for utilities of all sizes that supports one to many service types, with releases listed as 25.4, 25.10, 26.4, and 26.10 as of September 2026. That twice-yearly cadence is good news for staying current and a real workload for regression testing, so price it in. For utilities with AMI underway but a CIS replacement further out, Oracle’s 2020 Customer Cloud Service data sheet describes starting with its meter solution cloud service and adding the full customer platform later. For a deeper product-by-product comparison with SAP, read our Oracle vs SAP for utilities comparison. ### Cayenta: CIS plus integrated utility ERP Cayenta, a division of N. Harris Computer Corporation, was founded in 1983 as Solutions for Government to serve local governments and utilities, and that public-sector heritage still shows in its product line. Its solutions page lists a customer information system with a flexible rates engine alongside mobile workforce management, a customer engagement portal, meter data management, an outage management system, and work, financial, and human capital management designed for utilities and municipalities. It also includes Cayla, an AI layer with a customer service agent and an in-app assistant. Deployment is either CayentaCloud, which the same page describes as single-tenant SaaS in dedicated environments backed by SOC 2 certification, or on-premises. On implementation, Cayenta’s services page says its own team drives the program from planning through go-live, which is a different model from a systems integrator bid. For a full assessment of where it fits, see our Cayenta CIS review. ### Fit at a glance | Criterion | SAP (S/4HANA Utilities) | Oracle (CCS / CC&B) | Cayenta | Where the CIS lives | Inside the SAP ERP, with FI-CA receivables | Utility customer platform with shared meter and customer database | CIS within a utility and municipal ERP suite | Deployment options | SAP S/4HANA Cloud Private Edition or on-premises | SaaS (CCS) or on-premises (Customer to Meter) | Single-tenant SaaS (CayentaCloud) or on-premises | MDM | Energy data management within S/4HANA Utilities | Meter data management built into the platform | Meter data management integrated with the CIS | Strongest fit signal | Existing SAP finance and logistics estate | Oracle stack, or a SaaS-first CIS and MDM together | Municipal, public power, and multi-service utilities wanting one suite | Timing pressure to check | IS-U on ECC support dates | Release cadence and testing load | Scope of native modules versus integrations AI is moving fast across all three. Our SAP vs Oracle vs Cayenta AI comparison covers what each has shipped, but treat AI as a tiebreaker in the self-service score rather than a reason to pick a platform. ## RFP questions to ask each vendor Ask every vendor the same core questions, then add the vendor-specific ones. Questions for all three: - Recalculate these ten bills from our current tariff book live in the demo. Which rates required configuration, and which required code? - Who, by name, will lead our implementation, and which comparable utility projects have they delivered on this exact product? - What is included in the subscription or license, and what is priced separately (MDM, portal, mobile, analytics, AI features, test environments)? - How many mock data conversions are in the plan, and who owns bill reconciliation? - Can we speak with two reference utilities of our size and service mix, chosen by us from your list? For SAP: - Which deployment are you proposing, S/4HANA Cloud Private Edition or on-premises, and what does that mean for upgrades? - If we run IS-U on ECC today, which path do you recommend relative to the 2027 and 2030 maintenance dates, and does the 2031 to 2033 transition option apply to us? - Which systems integrator is bidding, and how is responsibility split between SAP and that partner? For Oracle: - Are you proposing Customer Cloud Service or Customer to Meter, and why? - With releases twice a year, what regression testing do we own, and what tooling is included? - If our financials are not Oracle, what integrations are prebuilt and what must be built? For Cayenta: - In CayentaCloud, what does Cayenta manage and what stays with our staff? Can we review the SOC 2 report? - Which modules in your proposal are native to the platform and which are integrated products? - How is your implementation team staffed for a project of our size, and what do you need from us? ## Red flags Slow the process down if you see any of these: - The vendor will not run your bills. A demo built on the vendor’s sample tariffs tells you nothing about your rate book. - “Configuration” that turns out to be code. Ask for a written list of every custom object in the proposal. - An unnamed implementation team. A bid that lists roles but not people can change after signature. - Pricing without scope. A per-customer price with no stated modules, integrations, or environments is not comparable. - No conversion plan. If the vendor cannot say how many mock conversions are planned, data migration has not been thought through. - Only hand-picked references. Ask for the full reference list and choose your own calls. - A vague upgrade story. You should know how the system stays current five years after go-live. ## A step-by-step selection process - Document the current state. Your rate book, bill types, integrations, and the pain points you want solved. Pull your hardest bills now; they become the demo scripts. - Set weights before you talk to vendors. Agree the scorecard with the steering group and record it. - Run an RFI to build a shortlist. Two or three vendors is enough for a serious evaluation. - Issue the RFP with scripted scenarios. Include your tariffs, sample meter data, and the questions above. - Hold scripted demos and a sandbox trial. Score in the room, against the evidence column. - Check references you chose. Ask what the vendor underestimated and what the utility would do differently. - Build a ten-year cost model. Put every bid on the same template so the totals compare. - Negotiate the statement of work, not just the price. Lock the team, the conversion plan, and acceptance criteria into the contract. Data migration deserves its own plan from day one. Our guide to utility CIS modernization and migration covers what real moves off legacy platforms look like. ## Sources and disclosure Vendor facts on this page come from the vendors’ own material, checked in September 2026: SAP’s S/4HANA Utilities training content and SAP News Center announcements, Oracle’s CIS product page and product documentation, and Cayenta’s solutions and about pages. None of these vendors publishes list pricing, so this guide gives no prices. Disclosure: UtilitiesLabs is backed by AvanSaber, which operates a SAP utilities consulting practice. That is a commercial relationship that could bias us toward SAP. We manage it by naming it, citing vendor sources for product claims, and holding SAP, Oracle, and Cayenta to the same scorecard. See our editorial policy. ## Frequently asked questions How do I choose between SAP, Oracle, and Cayenta for a utility? Agree your evaluation criteria and weights before you see a demo, then score each vendor against your own rate book, meter data, and customer scenarios. In practice, utilities already standardized on SAP start with S/4HANA Utilities, utilities on an Oracle stack start with Oracle Customer Cloud Service or Customer Care and Billing, and municipal and mid-market utilities should put Cayenta on the shortlist. The scorecard, not the vendor's positioning, should make the final call. What is the most common selection mistake? Scoring on feature checklists instead of evidence. All three platforms can bill a utility customer. Programs go wrong on rate complexity nobody tested, data migration nobody sized, and an implementation team nobody interviewed. Weight those areas heavily and make vendors prove them with your own data. --- ## Oracle Utilities Implementation: A Practical Guide to CC&B Projects URL: https://www.utilitieslabs.com/a-comprehensive-guide-to-oracle-utilities-implementation-best-practices-and-tips/ # Oracle Utilities Implementation: A Practical Guide to CC&B Projects AvanSaber Research/July 4, 2023/Updated June 1, 2026/1 min read Oracle Customer Care and Billing (CC&B) is one of the two enterprise standards for utility customer information and billing. The hard part is rarely the software, it is the implementation. This guide covers what an Oracle Utilities implementation involves and where programs go wrong. ## What the program actually includes A CC&B implementation is more than installing the billing engine: - Rating and billing configuration. Encoding every rate, rider, and tax the utility charges. - Data migration. Moving accounts, meters, and billing history off the legacy CIS cleanly. - Integration. Wiring CC&B to meter data management (Oracle MDM), outage management (Oracle Network Management), payment processors, and the general ledger. - Testing. Running real billing scenarios, including estimated reads, prorated bills, and deregulated-market cases, before go-live. ## Common pitfalls - Underestimating rate complexity. Time-of-use, tiered, and seasonal rates multiply test cases. Scope them early. - Dirty legacy data. Bad meter and account data in the old CIS surfaces as billing errors after cutover. Cleanse before you migrate, not after. - Big-bang cutovers. Phasing by service type or region reduces risk on high-volume billing. - Thin reconciliation. Plan parallel billing runs so you can prove the new system bills the same as the old one before you switch off the legacy system. ## Where this fits If you are still selecting between platforms rather than implementing, start with the Oracle vs SAP for utilities comparison. For utilities weighing a lighter-weight option, the Cayenta CIS review covers the mid-market alternative. ## Frequently asked questions What is involved in an Oracle CC&B implementation? An Oracle Customer Care and Billing implementation covers rate and rating configuration, billing and invoicing, payments and collections, data migration from the legacy CIS, integration with MDM and OMS, and testing against real billing scenarios. It is typically a multi-phase program. How long does an Oracle Utilities implementation take? For a mid-to-large utility, CC&B programs commonly run one to three years depending on service types, customer count, and how much custom rate logic and integration is in scope. Phased rollouts by service or region are common. --- ## Agentic AI for the Utility Back Office: Collections and Dunning URL: https://www.utilitieslabs.com/agentic-ai-utility-back-office-collections-dunning/ # Agentic AI for the Utility Back Office: Collections and Dunning AvanSaber Research/May 17, 2026/Updated June 2, 2026/5 min read Utility back-office teams carry a workload that most ERP vendors did not design for. Millions of contract accounts, dunning runs that must respect state and local regulation, billing exceptions generated by AMI data quality issues, and payment allocations across complex rate structures are all routine. Agentic AI is starting to address parts of this workload, but the places where it genuinely helps are narrower than the marketing suggests, and the places where human control is non-negotiable are wider. ## The utility back-office workload The billing-to-cash process in a utility runs through two main systems. On the SAP side, billing flows through SAP IS-U and the resulting receivables post to FI-CA, the Contract Accounts Receivable and Payable subledger that handles the high-volume, per-contract AR that standard accounts receivable cannot manage at utility scale. On the Oracle side, Customer Care and Billing (CC&B) handles the equivalent scope. The volume is the defining characteristic. A large municipal utility might have hundreds of thousands of active accounts. An investor-owned utility serving a major metro area can have millions. Dunning schedules, payment plans, disconnection notices, and collections steps run at that same volume, with regulatory requirements governing when each step can fire and what notice a customer must receive. Billing exceptions, meaning records that failed automated processing, arrive continuously as AMI data quality fluctuates, rate changes hit edge cases, and customer account configurations shift. That combination of scale, regulatory constraint, and exception volume is exactly the kind of problem agentic AI was built for in theory. In practice, the constraints shape what it can actually do. ## Where agentic AI genuinely helps Exception triage and routing. A model trained on historical exception data can rank the exception queue by root cause, likely resolution time, and revenue impact. Analysts stop working the queue in arrival order and start with the items that matter most. The model recommends a resolution path; the analyst confirms before anything posts to FI-CA or CC&B. This is one of the clearest wins because the task is data-rich, bounded, and the human confirmation step is easy to enforce. Payment pattern anomaly detection. Accounts that deviate from their historical payment pattern (long-paying accounts suddenly going silent, installment-plan accounts slipping) can be flagged before they reach the dunning queue. Earlier intervention often means better recovery rates and fewer costly late-stage collections steps. Dunning communication drafting and timing. AI can generate dunning notice text tailored to account history and channel preference, and it can propose the timing of each communication within the parameters set by regulatory rules. A workflow layer confirms before any notice is issued. AP reconciliation and matching. On the payables side, matching invoices to purchase orders at volume is a well-established AI use case that applies to utility AP as much as any other industry. Oracle shipped AI call summarization for utility customer service in May 2025, which is adjacent: the goal is reducing manual reconciliation and documentation work so back-office staff focus on exceptions rather than routine capture. For the broader integration picture, see how AI layers onto SAP IS-U and Oracle CC&B in practice. ## Where the model must stay human-in-the-loop The back-office tasks where AI cannot operate autonomously are not edge cases. They are the core of regulated utility billing. Disconnection and collections progression. Most jurisdictions regulate when a utility can issue a disconnection notice and what steps must precede it. An AI agent that autonomously progresses a collections action against a customer account is operating in a regulated domain. The risk is not just operational error; it is regulatory exposure. Every step that advances a customer toward disconnection requires a controlled workflow with an auditable human decision point. Regulated rate adjustments and corrections. Billing corrections on regulated tariffs often require documented justification. An AI recommendation to credit or rebill an account needs a human to own the decision, especially where the amount exceeds defined thresholds or touches a low-income or medical-baseline customer. Payment plan approvals. Payment plans often carry regulatory protections for customers who qualify. AI can identify candidates and draft plan terms, but approval should remain with a staff member who can verify customer circumstances. The principle is consistent: AI reads data, produces a ranked recommendation, and hands off to a person or a controlled workflow for any action that posts to the CIS, advances a collections step, or generates a customer-facing communication with regulatory implications. ## The AI-native ERP pattern A separate trend worth watching is the emergence of AI-native ERP systems in the general accounting market, products designed from the start with conversational interfaces and autonomous finance workflows rather than AI layered onto a decades-old transactional core. ERPClaw, an AvanSaber AI-ERP product, is one example of this pattern in general accounting and ERP. It is not a utility CIS and does not handle utility-specific billing, rates, or regulatory compliance. The relevance is architectural: these systems show what it looks like when the AI model is the primary interface rather than an add-on to a legacy core. Utility-specific back-office AI has not yet reached that pattern. The regulatory complexity, rate-structure depth, and FI-CA or CC&B integration requirements mean that utility CIS vendors are still in the phase of layering AI alongside the existing core rather than replacing it. That is likely the right approach for the regulated utility context, even as the general accounting market moves faster. ## Honest limits Agentic AI for utility back-office work is real and producing measurable value in exception triage, anomaly detection, and communication timing. It is not ready to autonomously run collections, issue dunning notices, or post billing corrections without human confirmation. The utilities getting value from it in 2026 are the ones that drew that line clearly in their implementation design and held it. The back-office finance layer is tightly connected to the broader meter-to-cash architecture. Any AI deployment touching FI-CA or CC&B should be evaluated in the context of the full platform strategy, not as a standalone AI project. ## Frequently asked questions Can AI automate utility dunning and collections? AI can automate the classification, prioritization, and routing of collections work, and it can draft and time dunning communications. The actual decision to progress a collections action against a customer on a regulated service must stay in a controlled workflow with human review, because utility disconnection and collections rules are regulated in most jurisdictions. What is FI-CA and why does it matter for AI? FI-CA is SAP's Contract Accounts Receivable and Payable subledger, the high-volume AR engine in SAP IS-U utilities. It handles millions of customer contract accounts, dunning schedules, payment allocations, and collections steps. AI that reads FI-CA data for prioritization or anomaly detection has access to the most complete view of a utility's receivable position. What is the AI-native ERP pattern for back-office finance? AI-native ERP refers to systems designed from the start with conversational interfaces and autonomous finance workflows, rather than AI bolted onto an existing transactional core. In general accounting, products like ERPClaw (an AvanSaber AI-ERP product) represent this pattern. Utility-specific back-office AI still largely means AI layered onto a legacy CIS or FI-CA core, because the regulatory and billing complexity of utilities has not yet been replicated in AI-native form. How does AI help with billing exceptions in a utility? Billing exceptions are records that failed automated processing and need manual review. A model trained on historical exception data can rank exceptions by likely root cause, estimated revenue impact, and urgency, so analysts work the highest-value items first. The model recommends; a person or controlled workflow confirms the resolution before it posts to the AR subledger. --- ## AI for Utility Operations: The Use Cases That Pay Off URL: https://www.utilitieslabs.com/ai-for-optimizing-utility-operations-enhancing-efficiency-and-performance/ # AI for Utility Operations: The Use Cases That Pay Off AvanSaber Research/July 21, 2024/Updated June 2, 2026/1 min read “AI for utility operations” only becomes useful when you name the specific jobs. Across a utility, the use cases that consistently pay off share a shape: a real data source, a bounded decision, and a human or controlled workflow that acts on the AI’s output. Here are the ones worth funding. ## Predictive maintenance Sensor, SCADA, and asset-history data feed models that flag transformers, pumps, and other assets likely to fail, so crews intervene before an outage. The output lands as work orders in the asset and work-management system. This is among the most mature utility AI categories. ## Outage prediction and faster restoration Weather, load, and historical-fault data predict where outages are likely, and during events AI helps prioritize crews and estimate restoration. This connects the grid platforms (ADMS) to the outage management system and customer communication. ## Load and demand forecasting Consumption, weather, and increasingly distributed-energy data drive forecasts that improve procurement, dispatch, and demand-response targeting. The value grows every year as electrification and DERs make demand harder to predict. ## Billing-exception triage High-volume utilities generate thousands of billing exceptions per cycle. A model ranks and routes them so analysts clear the highest-impact ones first, against data in SAP IS-U, Oracle CC&B, or Cayenta CIS. The detail is in AI for utility operations on SAP IS-U and Oracle CC&B. ## How to adopt it Pick one use case with a clear data source and a measurable outcome, keep the system of record in control, and prove value before scaling. For the customer-data side, see AI for energy consumption analysis; for the grid side, smart grid optimization; and for the vendor landscape, top AI software for utility companies. ## Frequently asked questions What are the best AI use cases for utility operations? The ones with clear data and a measurable outcome: predictive maintenance on assets, outage prediction and faster restoration, load and demand forecasting, and billing-exception triage. Each pairs an existing data source with a decision a person or controlled workflow confirms. Does AI run the utility, or assist it? Assist. The systems of record (SAP IS-U, Oracle CC&B, the ADMS for the grid) stay in control. AI produces predictions and recommendations that operators or controlled workflows act on; it does not autonomously run billing or the grid. --- ## AI in Utilities: Where It Actually Lands in 2026 URL: https://www.utilitieslabs.com/ai-in-utilities-where-it-actually-lands/ # AI in Utilities: Where It Actually Lands in 2026 AvanSaber Research/July 10, 2026/9 min read Search for “AI in utilities” and you get two kinds of content: vendor promises and academic surveys. This page is neither. It is a map of where AI has concretely landed in utility operations as of July 2026, organized by function, with named deployments and sources, plus an honest flag on what is still a pilot. Each section links to a deeper article on this site, so treat this as the hub. ## Key Takeaways - Customer engagement and billing are the most scaled AI surface in utilities today. Oracle reported approximately 45 million households in Opower AI-driven programs as of April 2026, and SAP’s Utilities Customer Self-Service Agent reached general availability in Q4 2025. - Machine-learning outage prediction is operational. The UConn and Eversource outage prediction model forecasts storm damage from weather and infrastructure data before crews are staged. - Meter data is the quiet giant: about 119 million AMI meters are installed in the United States per EIA’s most recent count, and AI is now moving into the meter itself. - Field-work AI is proven for inspection imagery and wildfire monitoring (PG&E), much thinner for the scheduling-optimization claims on vendor slides. - The clearest failure mode in the whole space is a chatbot arguing with a customer about a bill. Route disputes to people. - No utility CIS is AI-native yet. Everything real ships as a layer on the system of record, a distinction we unpack in AI-native ERP for utilities. “The pattern I trust is AI next to the system of record, never in place of it. When a vendor demo cannot show you the audit trail for a bill adjustment, you are not looking at a production capability. You are looking at a roadmap slide.” Nikhil Jathar, founder of AvanSaber ## Billing and the CIS: Exception Handling and Bill-Shock Prediction Billing is where AI meets the utility’s system of record, and it is where the most scaled deployments live. The pattern across every major vendor is the same: the CIS stays authoritative, and the AI works in front of it or beside it. On the SAP side, the Utilities Customer Self-Service Agent went generally available in Q4 2025, handling inbound account and billing inquiries against IS-U and FI-CA data. We cover the wider agent roadmap in our review of SAP Joule for utilities. At the mid-market end, Cayenta’s Cayla agent went live at the City of Lakeland, Florida in 2026, which matters because a municipal utility running this in production is stronger evidence than any enterprise pilot. Bill-shock prediction is further along than most buyers realize. Oracle’s Opower Proactive Alerts identify customers trending toward an unusually high bill and message them before it arrives; Oracle’s product documentation credits high bill alerts with 0.3% incremental energy savings on top of the 1.5% from Home Energy Reports, and a 9% reduction in high-bill calls. Fewer shocked customers means fewer angry calls, which is why this single capability pays for itself in contact-center load before anyone counts the energy savings. Exception handling is the less glamorous half of billing AI and arguably the more valuable one. Every billing run produces exceptions, and the win is triaging them by revenue impact and clearing the routine ones automatically, with a person approving anything that changes a bill. Where that is heading, including collections and dunning, is the subject of our piece on agentic AI in the utility back office, and the broader billing picture is in AI in utility billing. ## Customer Contact: Summarization Shipped, Dispute Handling Did Not The production win in the contact center is agent-facing, not customer-facing. Oracle shipped AI call summarization and tagging for utility customer service in May 2025, cutting after-call work rather than replacing the agent. That is the right shape for the first deployment: the AI drafts, the human owns the interaction. Customer-facing bots are a mixed record. A Verint survey of 1,500 consumers in 2024 found more than two-thirds reported a bad chatbot experience, with failure to understand the issue as the leading complaint. Bots do fine on outage status, balance checks, and payment arrangements with clear rules. They do badly on disputes, which are exactly the interactions that decide how a customer talks about you at a public utility commission hearing. Our deeper look at AI in utility customer service covers the agent-assist and chatbot architecture in detail. ## Field Service: Inspection AI Is Real, Scheduling Claims Need References The proven field-work AI is computer vision on inspection imagery. PG&E built its Sherlock application to let inspectors label defects in drone and helicopter photos, then trained computer-vision models on those labels to pre-screen imagery for equipment defects. Ahead of the 2025 fire season, PG&E also reported more than 630 high-definition cameras monitoring 90% of its high fire-risk territory, with AI detecting smoke and automating alerts. Those are named, dated, production programs. AI-optimized crew scheduling and dispatch is a different story. The capability exists inside field-service suites, and the logic (skills, drive time, job urgency, weather) is sound, but public evidence of named utility deployments with measured results is thin as of July 2026. Treat scheduling optimization as a claim to verify with a reference customer of your size, not a fact to assume. The adjacent and better-evidenced area is failure prediction on assets, covered in AI predictive maintenance for utilities. ## Grid Operations and Load Forecasting Load forecasting is the oldest machine-learning problem in the industry, and it is getting formal infrastructure. In March 2025, EPRI, NVIDIA, and Articul8 launched the Open Power AI Consortium to build domain-specific AI models for the power sector, with an advisory committee drawn from more than 20 energy companies including Duke Energy, Exelon, and PG&E. That consortium exists largely because data-center load growth broke the old forecasting assumptions and utilities need better tools for it. On the operations side, the strongest quantified reference is E.ON, which reported a 77% reduction in IT downtime over five years while deploying SAP S/4HANA-based AI for predictive maintenance and customer service, per June 2026 reporting. That is a sustained program result, not a pilot readout. One honesty note buyers should hold onto: much of what gets marketed as grid AI is deterministic automation. Duke Energy’s self-healing technology in Florida avoided more than 950,000 extended outages and saved nearly 6.3 million hours of outage time since January 2024, which is a genuinely impressive result achieved with automated fault detection and rerouting, not machine learning. Both matter. They are different purchases with different risk profiles. For the demand-side analytics half of this topic, see AI for energy consumption analysis. ## Outage Prediction: Machine Learning Before the Storm Predicting outages before weather hits is one of the cleanest utility AI use cases, because the training data (past storms, past damage, infrastructure and vegetation data) already exists. The reference deployment is the UConn Outage Prediction Model built with Eversource, which runs ahead of storms to estimate outage counts across the service territory so crews and mutual aid are staged before the first call. The technology has since moved beyond one utility: Schneider Electric partnered with UConn and Eversource to fold the model into its commercial weather platform for other utilities. The buyer question here is validation: any storm model should be backtested against your own territory’s outage history, not a national average, before its forecasts drive staging decisions that cost real money. ## Meter Data: AMI Analytics and AI at the Grid Edge The United States has about 119 million AMI meters installed, roughly 72% of all electric meters, per EIA’s most recent count. That interval data is the raw material for most of the analytics above, and two production developments stand out. First, anomaly detection inside the meter-data platform: Oracle added AI anomaly detection to its Meter Data Management product in June 2025, aimed at energy-loss detection and faulty-meter identification. Second, AI is moving into the meter itself: Utilidata’s Karman platform, built on an NVIDIA edge module, raised $60.3 million in April 2025 to scale distributed AI across energy infrastructure, and Department of Energy grid awards are funding deployments of more than 150,000 Karman units at Portland General Electric, ComEd, and Duquesne Light. Meter-embedded AI is early, but it is funded, named, and shipping rather than conceptual. What most utilities actually need first is duller: getting AMI, CIS, and GIS data clean and joined enough for any model to use. That groundwork is the subject of utility data management with AI. ## Production Versus Pilot at a Glance | Area | In production today | Still pilot-stage | First question to ask | CIS and billing | Self-service agents (SAP GA Q4 2025, Cayla live), high-bill alerts (Opower) | Autonomous bill correction without human approval | Where does the human approve a bill-changing action? | Customer contact | Call summarization (Oracle, May 2025), agent assist | Bots resolving billing disputes | What happens on escalation, and how fast? | Field work | Inspection computer vision, AI smoke detection (PG&E) | AI-optimized crew scheduling at utility scale | Which utility my size runs this, and what changed? | Grid ops and forecasting | ML load forecasting, EPRI consortium models emerging | Autonomous grid control, grid-specific LLMs | Was the model validated on our load shape? | Outage prediction | Storm outage models (UConn and Eversource) | Real-time restoration optimization | Backtested against our outage history? | Meter data (AMI) | MDM anomaly detection (Oracle, June 2025) | Meter-embedded AI at full scale (Karman rollouts) | Is our AMI data clean enough to train on? ## What Is Still Pilot-Stage Three things stay firmly in the pilot column as of July 2026. Fully agentic workflows that chain actions across CIS, finance, and field systems are being showcased, not routinely operated; SAP’s autonomous enterprise material names RWE in its offshore wind context, and that should be read as a directional signal rather than a deployment template. Grid-specific foundation models from the EPRI consortium are in early access, not in control rooms. And meter-embedded AI is at the start of its rollout curve, with the named deployments above still being installed. “Utilities do not get credit for being early. They get credit for being right. Start where a wrong answer costs you a confused customer, and earn your way toward the places where a wrong answer costs you a feeder.” Nikhil Jathar, founder of AvanSaber ## What to Ask Before Buying Every area above collapses into the same five questions. Is the capability generally available or a roadmap item, with a date either way? Can the vendor name a reference utility of your size running it in production? How was the model validated against your kind of data, your tariffs, your storms, your load shape? Where does a human approve any action that touches a bill or the grid? And what audit trail does it leave for the regulator who will eventually ask? How the three major CIS vendors answer those questions today is laid out in our SAP vs Oracle vs Cayenta AI comparison, and the wider evaluation framework, AI included, is in the guide to choosing the right utilities software. For the full set of articles behind this map, the AI in utilities topic hub lists them in reading order. ## Frequently asked questions Where is AI in utilities already in production? Customer engagement and billing lead. Oracle reported approximately 45 million households enrolled in Opower AI-driven programs as of April 2026, SAP's Utilities Customer Self-Service Agent reached general availability in Q4 2025, and Cayenta's Cayla agent went live at a Florida municipal utility in 2026. Machine-learning outage prediction (the UConn and Eversource model), AI meter-data anomaly detection (Oracle, June 2025), and computer-vision inspection programs (PG&E) are also operational. What is the difference between AI and grid automation like self-healing networks? Self-healing distribution systems, such as the one Duke Energy runs in Florida, detect a fault and reroute power using deterministic automation rules, not machine learning. They deliver large, measurable reliability gains, but they are automation. AI in the strict sense (models trained on data that make predictions) shows up in load forecasting, storm outage prediction, meter anomaly detection, and customer analytics. Buyers should keep the two categories separate when evaluating vendor claims. Should a utility buy AI as a product or as a feature of its CIS and ERP? As of 2026, almost all production AI for utilities ships as a layer on an existing system of record: Joule on SAP, Opower and CC&B AI on Oracle, Cayla on Cayenta. No AI-native utility CIS exists yet. The practical question is not which AI product to buy but how well your platform vendor's AI layer integrates with the data you already have. What should a utility ask a vendor before buying an AI capability? Five things: whether the capability is generally available or a roadmap item, a named reference utility of similar size running it in production, how the model was validated against your kind of data, where the human approval sits for any action that touches a bill or the grid, and what audit trail the system produces for regulators. --- ## AI in Utilities URL: https://www.utilitieslabs.com/ai-in-utilities/ # AI in Utilities AI in utilities is less about a single platform and more about a set of places where machine learning and language models earn their keep: explaining a bill, triaging a service call, guiding a field tech, and reading the legacy data nobody else can. This topic maps where the value is real today, where it is still a pilot, and how to add it without ripping out the systems that keep the lights on. The short answer: AI in utilities is not one product you buy, it is a layer you add to the systems you already run. The value shows up first in billing, customer service, and field operations, and it shows up because those are the places where a fifty-year-old data model has to be explained to a real person who is waiting. This page is the decision map: where AI belongs in a utility architecture and which operating questions to ask before funding it. For a dated inventory of named products, deployments, and current vendor evidence, use AI in utilities: where it actually lands in 2026. For a product shortlist, use the separate AI software buyer guide. Keeping those jobs separate prevents a broad strategy page, an evidence report, and a vendor comparison from competing for the same search intent. ## Why “layer it, do not replace it” is the whole strategy The pitch a utility usually hears is that AI requires a new core system. For most utilities that is a fantasy, and an expensive one. You do not rip out the system of record for a million customers because someone demoed a clever assistant. The decades of edge cases, tariff rules, and regulatory logic baked into the existing platform are the actual asset. So the real work is the opposite of replacement. Leave the system of record where it is, and put AI on top of it. That is the pattern behind AI-native ERP for utilities and the reason it beats a clean-slate rebuild that throws away what the old system knows. ## Where the value is real today Billing transparency. A customer calls because their bill tripled. Behind that one question sit meter reads, tariff rules, weather, and a data model that was never designed to explain itself. An AI layer can read all of that and hand the agent, or the customer, a plain answer. That is the clearest win, and we cover it in AI in utility billing. Customer service. Triage, routing, and first-line answers are a natural fit, because the work is high volume and pattern heavy. The detail is in AI in utility customer service. Field operations. Guiding a technician through a meter swap or a substation inspection is where AI meets the physical world. This is also where augmented reality and AI overlap, which we treat in our work on augmented reality for SAP and utility field service. Making legacy data legible. The most underrated use case is the least glamorous: getting a useful answer out of a system that predates the people maintaining it. That is the spine of AI for utility data management. ## Where it is still a pilot Grid optimization, full digital twins, and autonomous control are real areas of research, but for most utilities they sit closer to a pilot than to a production system you can lean on. The honest framing is to treat them as experiments with clear success criteria, not as line items you can count on this budget cycle. AI that sits next to deterministic controls is production. AI that replaces them is a project, and a careful one. ## How to add AI without betting the reliability of the grid The discipline is the same whether you are posting a billing adjustment or guiding work on a feeder: keep the probabilistic system next to the deterministic one, never inside it. Insist on an audit trail, reversibility, and a human who stays accountable for the outcome. Utilities move deliberately for good reasons, and that instinct is a feature here, not a frustration. The whole point of building the governance scaffolding is that you do not have to choose between AI and control. Start where the answer is concrete and the cost of being wrong is a confused customer rather than a dark neighborhood. Prove it there, then move outward. The map above is ordered roughly that way on purpose. ## In this topic - AI in utilities: the evidence-backed 2026 field map - AI software for utility companies, compared by use case - AI on SAP IS-U and Oracle CC&B - Agentic AI for collections and dunning - AI for utility operations - How AI and machine learning are reshaping utilities - AI in utility billing - AI in utility customer service - AI-native ERP for utilities - AI for utility data management - AI across the meter-to-cash architecture ## Everything else in this topic 19 more articles, A to Z. - AI Analytics on Utility ERP and CIS Data: A Practical Guide - AI and AR for Utility Safety and Leak Detection - AI and XR Digital Twins: Operator Visualization in Utilities - AI and XR for Utility Customer Service: Proactive and Field Support - AI and XR in Smart City Utilities: Grounded Assessment - AI and XR in Utility Customer Service: What Is Real Now - AI Data Analytics in Utilities: Loss, Demand, and Asset Outcomes - AI for Energy Consumption Analysis in Utilities: What Works - AI for Energy Distribution: Cutting Losses and Balancing the Grid - AI for Smart Grid Operations: What Actually Works Today - AI in Water Utility Management: Leaks, Quality, and Pressure - AI Predictive Maintenance for Utilities: Cut Downtime - AI Tools for Predictive Maintenance in Utilities: The Pipeline - AI Transforming the Energy Sector: Honest Trends Overview - Digital Twins for Grid Management: AI, GIS, and ADMS - Digital Twins in Utilities: What's Real vs. What's Still Hype - Predictive Analytics in Utility Management: Forecasting and Assets - Smart Grid Optimization: ADMS, DERMS, and AI for the Distribution Grid - Smart Grids Explained: AMI, ADMS, and DERMS in Utility Operations --- ## AI-Native ERP for Utilities: What Legacy CIS Don't Do Yet URL: https://www.utilitieslabs.com/ai-native-erp-for-utilities-what-legacy-cis-dont-do/ # AI-Native ERP for Utilities: What Legacy CIS Don't Do Yet AvanSaber Research/June 2, 2026/4 min read Every utility CIS vendor is adding AI. None of them has rebuilt their platform as an AI-native system. Understanding that gap, what it means in practice, and how long it is likely to persist is useful for any utility making a platform decision in the next two to three years. ## AI-bolted-on versus AI-native The distinction matters more in practice than in marketing. An AI-bolted-on system starts with a transactional core, often decades old, and adds an AI layer alongside it. The AI reads from the database, produces recommendations, and writes back through the existing APIs. The core data model, the user interface, and the workflow engine were designed for screen-based, menu-driven operation. AI is a capable add-on in this architecture, but the primary interaction model is still the original system. An AI-native system designs the AI as the primary interface from the start. The ledger or record exists to support what the AI needs, not the other way around. Workflows, data structures, and audit trails are built around the assumption that a conversational agent or autonomous process is the normal way to interact with the system, not a special mode. In the general ERP and accounting market, AI-native products are beginning to appear. ERPClaw, an AvanSaber AI-ERP product, is one example of this pattern in general accounting and ERP. It is not a utility CIS and does not handle utility billing, rate structures, or regulatory compliance. It illustrates the architectural pattern, not a utility-specific product roadmap. In utility CIS specifically, every platform on the market today, including SAP IS-U and S/4HANA Utilities, Oracle CC&B, and Cayenta CIS, is in the AI-bolted-on category. That is not a criticism. It is a description of where the market is. ## What legacy utility CIS platforms do well The case for legacy utility CIS platforms is substantial and often undersold by the vendors themselves in their AI marketing. These systems carry thirty or more years of accumulated domain logic that an AI-native system would need to replicate before it could serve as a system of record for regulated billing. Rate structure depth. Utility rates are complex. Time-of-use rates, tiered structures, demand charges, low-income programs, net metering credits, and deregulated-market supply rates all have to bill correctly at scale. SAP IS-U, Oracle CC&B, and Cayenta CIS have implemented and debugged these structures across hundreds of utilities. That depth is not in a language model. Regulatory billing compliance. Disconnection rules, dunning notice requirements, payment plan protections, and low-income customer safeguards are embedded in utility CIS workflow engines. The logic reflects specific regulatory orders and utility tariff filings. Replicating this in an AI-native system without a dedicated regulatory compliance layer is a multi-year project. FI-CA and high-volume AR. SAP’s FI-CA Contract Accounts Receivable and Payable subledger handles millions of contract accounts, payment allocations, dunning schedules, and collections at utility scale. Oracle CC&B handles the equivalent scope. These are not general AR systems. They were built for the volume and complexity utilities generate. Integration ecosystem. Utility CIS platforms have established integrations to OMS, MDM, GIS, SCADA, and payment processors. SAP, Oracle, and Cayenta each bring an established partner and integration ecosystem. An AI-native entrant would need to build or partner for all of it. ## The gaps that AI-native architecture addresses Despite that depth, legacy utility CIS platforms have real gaps that the AI-native pattern addresses, at least in non-utility contexts. Autonomous finance operations. In a properly AI-native system, routine accounting tasks like payment matching, exception resolution, and close processes can run with minimal human initiation. In a bolted-on AI system, AI recommendations still require a human to navigate to the right screen and confirm the action. The friction is lower than it was, but the interaction model is still the original system’s. Conversational interfaces. Staff training on legacy CIS systems is a significant ongoing cost. Menus built for 1990s workflows, dozens of transaction codes, and complex configuration screens are not intuitive to new staff. An AI-native system would allow a staff member to describe what they need in natural language and have the system act. Legacy CIS AI layers are adding this capability, but the underlying screen structure often limits how far they can go. Real-time exception handling. Legacy CIS exception queues are often batch-driven. An AI-native architecture can surface and route exceptions in near real time as the underlying data changes. Bolted-on AI can approach this with the right integration design, but it requires engineering that would not be needed in a purpose-built AI-native system. ## The realistic 2026 outlook for utilities The gap between AI-bolted-on and AI-native matters, but it is not a reason to defer utility CIS decisions while waiting for an AI-native utility platform. No vendor has announced one, and the depth of domain logic required means the timeline for a credible AI-native utility CIS is measured in years, not months. What utilities should be evaluating now is the quality of each vendor’s AI integration architecture. SAP announced more than 200 agents and 50 assistants at Sapphire in May 2026 and is positioning Joule as the primary agentic layer across the SAP estate. Oracle has shipped targeted AI in Opower, call summarization, and MDM anomaly detection. Cayenta’s Cayla agent is in production at mid-market utilities. Each of these is a bolted-on AI layer, and each has genuine value on the right set of tasks. The AI-native pattern is worth watching as it develops in the broader ERP market. When it begins to appear in utility-specific contexts, the vendors already building AI-forward CIS architectures will have the head start. For now, the platform decision is still primarily about domain depth, integration fit, and implementation risk, with AI integration quality as a meaningful but secondary factor. For a practical guide to weighting these factors in a procurement, see how to choose the right utilities software and the overview of utility billing ERP options. ## Frequently asked questions What does AI-native ERP mean? AI-native ERP refers to systems designed from the start with AI as the primary interface and workflow layer, rather than AI added onto an existing transactional core. In these systems, a conversational interface is the default way to interact with the ledger, not a feature added to a screen-based system. In the general ERP market, this pattern is emerging. In utility CIS specifically, it has not yet arrived. What do legacy utility CIS platforms do well that AI-native systems cannot replace? Legacy utility CIS platforms like SAP IS-U, Oracle CC&B, and Cayenta CIS carry decades of rate-structure modeling, regulatory billing logic, FI-CA or CC&B subledger integration, and meter-to-cash process depth. An AI-native system cannot substitute for that domain logic. The CIS is the system of record for regulated billing, and it will remain so. Is there an AI-native option for utility billing today? No. The AI-native ERP pattern exists in the general accounting and ERP market, but no AI-native platform has replicated the utility-specific billing, rate, and regulatory scope of SAP IS-U, Oracle CC&B, or Cayenta CIS. AI-native products like ERPClaw (an AvanSaber AI-ERP product) handle general accounting and ERP, not utility-specific billing or customer information system functions. Should utilities wait for an AI-native CIS before making a platform decision? No. An AI-native utility CIS is not on any vendor's near-term roadmap as a full replacement for IS-U, CC&B, or Cayenta. Utilities replacing aging CIS platforms should evaluate the current market on its merits, including how each platform's AI layer integrates, then revisit as the market evolves. --- ## AI Predictive Maintenance for Utilities: Cut Downtime URL: https://www.utilitieslabs.com/ai-predictive-maintenance-for-utilities-reduce-downtime-cut-costs/ # AI Predictive Maintenance for Utilities: Cut Downtime AvanSaber Research/January 21, 2025/Updated June 2, 2026/3 min read Utilities manage thousands of assets that fail in ways that are rarely sudden. A distribution transformer runs warm for weeks before the insulation breaks down. A pump bearing degrades gradually through a measurable vibration signature. AI predictive maintenance works because that degradation leaves a trail in data that utilities already collect. ## What the Technology Actually Does At its core, AI predictive maintenance applies machine learning to time-series data from sensors and historical records to estimate how far an asset is from a failure state. The output is a risk score or remaining-useful-life estimate for each monitored asset, refreshed as new readings arrive. The inputs vary by asset class. For distribution transformers the key signals are dissolved-gas analysis, oil temperature, load factor from SCADA, and fault history from the work order system. For pumps and compressors the dominant signals are vibration frequency, bearing temperature, and flow deviation from baseline. For overhead feeders the signals include feeder-relay trip counts, load imbalance, and outage-restoration patterns stored in the outage management system. None of this replaces the engineering judgment of a crew planner. The model surfaces anomalies and ranks assets by risk; a qualified engineer reviews the recommendation and creates or defers the work order. The CIS or ERP that holds the asset register remains the system of record. The AI sits alongside it, reading data and writing back draft work orders. ## The Data Pipeline Reliable predictions depend on data quality at every step. SCADA systems capture real-time readings but they are rarely clean enough to feed directly into a model. Spikes from communication errors, gaps from maintenance windows, and inconsistent tagging across substations all need to be addressed before modeling begins. This is often the most time-consuming part of a deployment. AMI platforms from Itron and Landis+Gyr contribute meter-event logs that can reveal transformer loading patterns not visible from SCADA alone. High meter-outage event rates on a single transformer, for example, are an early indicator of overloading or connectivity problems. Work order history from the EAM or ERP provides the labeled failure events the model needs to learn from. If work orders are not consistently coded with failure type and asset ID, the training data is thin and prediction accuracy suffers. Utilities that invest in work order data quality before an AI deployment see meaningfully better results. ## Asset Classes That Benefit Most Distribution transformers are the asset class where predictive maintenance shows the clearest value. The failure modes are well understood, the data signals are measurable, and an unexpected transformer failure can leave hundreds of customers without power while incurring significant emergency repair costs. Submersible and centrifugal pumps in water and wastewater operations are a close second. Pump failures often go undetected until flow drops, by which point secondary damage to the wet well or pressure zone has occurred. Vibration-based anomaly detection can catch bearing wear weeks in advance. Transmission and distribution feeders are harder to model because they are affected by external events (trees, vehicle strikes) that no sensor can predict. AI is useful here for pattern-based risk scoring of aging infrastructure rather than real-time anomaly detection. ## Where This Fits in a Broader Utility Operations Strategy Predictive maintenance on its own does not improve outcomes if the work order process is broken or if crews lack parts. The discipline works when it is connected to a functioning planning and scheduling system, which in turn connects to inventory management in the ERP. For utilities running SAP IS-U or Oracle Utilities, that integration path is well-established. SCADA connectors and EAM modules can route AI-generated work order drafts into standard approval workflows without building custom middleware from scratch. For the complementary view on the tools and data pipelines that support this workflow, see our article on AI tools and the predictive maintenance data pipeline. If you are evaluating vendors or want to discuss how this fits your current platform, the AvanSaber team works on utility operations and CIS/ERP integrations. ## Frequently asked questions What data sources feed an AI predictive maintenance model? The most common inputs are SCADA telemetry (voltage, current, temperature, vibration), meter-event logs from AMI platforms such as Itron or Landis+Gyr, work order history from the EAM or ERP, and manufacturer life-cycle data for each asset class. Does the CIS or ERP need to change to support predictive maintenance AI? Usually not. The AI tool reads data from SCADA and the asset register through APIs or an integration layer. Work order recommendations flow back into the ERP or EAM as draft orders for a planner to approve. The CIS remains the system of record throughout. Which assets benefit most from this approach? Distribution transformers, submersible pumps, compressors, and overhead feeders see the highest return because failures are expensive, unplanned outages affect many customers, and the sensors needed to monitor them are already installed on most modern systems. --- ## AI in Water Utility Management: Leaks, Quality, and Pressure URL: https://www.utilitieslabs.com/ai-revolutionizing-water-utility-management/ # AI in Water Utility Management: Leaks, Quality, and Pressure AvanSaber Research/June 6, 2024/Updated September 24, 2026/3 min read Water utilities face a specific set of operational problems that are different from electric utilities, and AI applications in water management are most valuable when they are designed around those problems: finding where water is leaking before it surfaces, managing pressure to protect aging mains, and catching water quality anomalies early enough to respond before they affect customers. ## Non-Revenue Water and Leak Detection Non-revenue water (NRW) is the most consistently cited operational challenge in water distribution. Physical losses through pipe leaks account for a large share of NRW at most utilities, and finding them is slow and expensive using traditional acoustic surveys alone. AI-based leak detection analyzes pressure transducer readings and flow meter telemetry from SCADA to identify pressure patterns that are consistent with a developing leak. Night-line flow analysis, which compares minimum overnight flow against expected demand, is an established technique; AI extends it by correlating readings across multiple pressure zone sensors simultaneously and flagging anomalies with a geographic zone estimate that directs survey crews to the right neighborhood rather than the whole district. AMI data from platforms such as Itron and Landis+Gyr adds another signal. A cluster of meters reporting unusually low pressure or unusually high consumption at hours when demand should be flat can indicate a distribution main break before it is visible at the surface. The CIS remains the system of record for meter data and billing. The AI layer reads from the AMI system and SCADA and reports back to operations, not to the billing workflow. ## Pressure Zone Management Pressure that is too high accelerates pipe wear and increases leak rates. Pressure that is too low creates compliance and service quality problems. Managing pressure zones to stay within narrow operating bands is increasingly done through automated pressure reducing valve (PRV) control systems guided by predictive models. AI models trained on historical demand patterns and weather data can produce short-horizon pressure demand forecasts that allow PRV control algorithms to adjust set points ahead of demand peaks rather than reacting after the fact. SCADA telemetry feeds these models continuously, and the control decision stays within the SCADA/control system layer. The AI is the forecasting engine; a human operator or a SCADA control script executes the adjustment. ## Water Quality Monitoring Water quality monitoring in distribution systems traditionally relies on a small number of fixed sampling points and scheduled grab samples. AI adds value at two levels. Continuous sensor monitoring for residual chlorine, turbidity, and pH can detect unexpected changes in minutes. AI anomaly detection, applied to these continuous streams, identifies deviations from the asset’s own historical baseline rather than firing only when a reading crosses a fixed threshold, which catches gradual deterioration earlier. This narrows the response window for the operations team. Event detection systems that correlate anomalies across multiple sensors in the distribution network can isolate the probable source zone of a quality change, which is useful when a potential contamination event is developing. The output is an alert to the water quality team and a suggested zone for immediate manual sampling. Laboratory confirmation and regulatory notification still follow standard protocols; AI does not short-circuit that process. ## Demand Forecasting and Infrastructure Planning Short-term demand forecasting helps with pump scheduling and energy management, reducing pumping costs by running high-energy pump stations during off-peak tariff windows. Medium and long-term forecasting supports capital planning decisions about when to upsize transmission mains or add storage capacity. For water utilities running an enterprise CIS, tools like Cayenta CIS (a Harris Computer division) paired with SmartWorks MDM, a sister Harris product Cayenta offers as an add-on, can give demand forecasting models access to clean, validated interval reads rather than raw AMI data, which improves forecast accuracy. ## What AI Does Not Replace AI monitoring does not replace the certified operators who hold responsibility for safe water delivery. It does not replace physical inspection of aging infrastructure. And it does not replace manual water quality sampling required by regulators. What it does is compress the time between an anomaly developing and an operator knowing about it, and direct field crews to higher-probability problem locations rather than having them survey the entire network. For the broader predictive maintenance context across utility asset types, see AI predictive maintenance for utilities: reduce downtime and cut costs. For how utilities structure the underlying data pipeline that makes these applications reliable, see maximizing utility data management with AI. The AvanSaber team advises water utilities on integrating operational AI applications with existing CIS and SCADA environments. ## Frequently asked questions What is non-revenue water and how does AI help reduce it? Non-revenue water (NRW) is water that enters the distribution system but does not generate revenue, mainly through physical leaks, meter inaccuracies, and unauthorized use. AI identifies NRW by correlating pressure sensor readings, flow meter data, and AMI meter events to locate likely leak zones for targeted investigation. Does AI for water quality replace manual sampling? No. AI monitors continuous sensor streams for early warning anomalies and narrows the area where an issue may exist, but regulatory compliance still requires manual sampling and laboratory confirmation. AI speeds the detection and response window; it does not replace the chain of custody. Which systems does AI need to read in a water utility? The main sources are SCADA (pressure transducers, flow meters, pump telemetry), AMI meter data for consumption patterns, LIMS (laboratory information management) for water quality results, and the asset register or work order system for infrastructure history. --- ## AI for Smart Grid Operations: What Actually Works Today URL: https://www.utilitieslabs.com/ai-solutions-for-efficient-smart-grid-management/ # AI for Smart Grid Operations: What Actually Works Today AvanSaber Research/July 31, 2024/Updated June 2, 2026/3 min read Smart grid AI is often described at a high level of abstraction. This article focuses on what utilities actually run in their control rooms today and where AI is adding measurable operational value. For the platform-level view of smart grid technology and longer-range optimization strategies, see smart grid optimization: AI-powered sustainable energy solutions. ## Fault Detection, Isolation, and Restoration The highest-value AI application in current grid operations is augmenting fault detection, isolation, and restoration (FDIR). When a feeder fault occurs, an operator traditionally relies on relay trip signals, customer outage calls, and crew reports to identify the fault location and the best switching sequence to restore power to unaffected sections. ADMS platforms from GE Vernova (GridOS) and Schneider Electric (EcoStruxure ADMS) include AI-assisted FDIR modules that analyze relay event sequences and smart meter outage notifications to estimate the fault location and present a switching plan to the operator. The operator reviews and approves the switching actions before they execute. On automated switch circuits, the system can execute isolation steps while a human confirms restoration switching. The distinction matters: the AI proposes, a human confirms. Oracle Utilities Network Management System (NMS) provides similar capabilities within the Oracle Utilities Application Framework stack, which many utilities already use alongside Oracle CC&B for customer information. ## Volt/VAR Optimization and Conservation Voltage Reduction Volt/VAR optimization (VVO) and conservation voltage reduction (CVR) use AI to find the optimal voltage and reactive power settings across a distribution circuit that minimize losses or reduce customer energy consumption while staying within voltage quality standards. These applications run continuously, adjusting capacitor bank positions and voltage regulator tap settings in near-real time as load changes. The AI component is the optimization engine that solves the power flow equations quickly enough to be useful in an operating environment, not just in planning studies. GE Vernova GridOS and Schneider EcoStruxure ADMS both include production VVO/CVR modules that are deployed at scale across North American utilities. The value depends on feeder characteristics. Circuits with significant reactive load and long radial runs tend to see higher loss-reduction results. Not every feeder benefits equally, and utilities should expect variation across their circuit portfolio. ## DER Integration and Dispatch Distributed energy resources (DERs) including rooftop solar, customer-sited batteries, and demand response are now numerous enough that their collective behavior affects distribution circuit loading in ways that require active management. A DERMS layer, such as AutoGrid (part of Schneider Electric’s ecosystem), manages DER dispatch for grid services: calling on flexible loads or batteries to smooth a peak, or curtailing generation to prevent a feeder overload. The ADMS provides the physical circuit state; the DERMS optimizes the DER portfolio response. Both systems exchange data with the SCADA historian and pass signals back through control system interfaces. For utilities with Oracle or SAP billing platforms, DER participation in demand response programs also has a billing and settlement dimension. When a customer’s battery participates in a grid event and earns a credit, that credit has to flow back through the meter-to-cash process correctly, which means the CIS and billing configuration matter as much as the grid operations software. ## Load Forecasting for Operations Short-term load forecasting at the substation and feeder level is an established AI application that directly supports day-ahead and hour-ahead operational decisions. Models trained on historical SCADA load data, AMI interval reads, and weather forecasts produce feeder-level load predictions that help operators anticipate transformer loading and schedule switching maintenance during low-load windows. These models are typically retrained periodically as grid topology and DER penetration change. A feeder that added a large battery installation or a new commercial customer looks different from a model trained on two-year-old data. ## Honest Limitations AI does not eliminate operator judgment in grid operations. Complex fault scenarios, unusual topology, and coordinated multi-feeder events still require experienced dispatchers. AI tools reduce routine cognitive load and surface information faster, but they are working aids rather than replacements for grid operations expertise. Cybersecurity for AI-assisted control systems also deserves direct attention. Any AI layer that communicates with SCADA must be evaluated for its attack surface, not assumed to be isolated from operational technology networks. For the full context of AI trends across the sector, see AI transforming the energy sector: key trends. For the predictive maintenance side of grid asset management, see AI predictive maintenance for utilities. ## Frequently asked questions What is the difference between an EMS, a DMS, and an ADMS? An energy management system (EMS) handles transmission-level generation and scheduling. A distribution management system (DMS) handles distribution switching and fault isolation. An advanced distribution management system (ADMS) combines the DMS with outage management and adds analytics and automation layers, including AI-assisted Volt/VAR optimization and DER dispatch. Where does DERMS fit alongside ADMS? ADMS manages the physical distribution grid. A DERMS manages the dispatch and coordination of distributed energy resources, such as batteries, rooftop solar, and demand response assets. AutoGrid, now part of Schneider Electric's EcoStruxure ecosystem, is one example. The two systems exchange data but serve different control functions. Does AI in grid operations require replacing legacy SCADA? Not necessarily. Most AI-assisted grid operations tools are designed to read from existing SCADA historians and send recommendations back to operators rather than replacing the SCADA control layer. The SCADA system retains control authority. --- ## AI Tools for Predictive Maintenance in Utilities: The Pipeline URL: https://www.utilitieslabs.com/ai-tools-for-predictive-maintenance-in-utilities-enhancing-efficiency/ # AI Tools for Predictive Maintenance in Utilities: The Pipeline AvanSaber Research/May 2, 2024/Updated June 2, 2026/3 min read The primary overview of AI predictive maintenance goals and asset-class fit is covered in AI predictive maintenance for utilities: reduce downtime and cut costs. This article focuses on the tools and data pipeline that sit behind those outcomes. ## The Four Layers of a Working Pipeline A reliable predictive maintenance program needs four connected layers: data collection, data integration, modeling, and action. Data collection starts with sensors already in the field. SCADA systems capture substation and feeder telemetry continuously. AMI platforms from Itron and Landis+Gyr produce interval reads, outage events, and tamper alerts at scale. Some utilities supplement these with dedicated IoT sensors on transformers for dissolved-gas or acoustic partial-discharge monitoring. The key question at this layer is not whether data exists but whether it is tagged consistently enough to be useful. Data integration moves raw readings into a form that models can use. This typically involves a time-series database or a data historian, a cleansing step that handles gaps and outliers, and a join with the asset register held in the ERP or EAM. SAP BTP and Oracle Utilities Application Framework both offer connectors that help utilities pull asset metadata without duplicating records. Modeling applies machine learning to the clean, integrated data. Supervised models need labeled failure events from work order history to learn what pre-failure patterns look like. Unsupervised anomaly detection can work without labels, which helps for asset classes with few historical failures. Neither approach is universally better; most production deployments combine both. Action is where value is actually created or lost. A well-scored asset that sits in a queue no planner reviews does nothing. The output of the modeling layer should connect to a workflow that routes risk flags to the right crew planners, creates draft work orders in the ERP, and checks parts availability in inventory management. ## Tool Categories and Their Roles Utilities typically draw on tools from several categories rather than a single vendor: - SCADA and historians: PI System (AVEVA/Schneider) and OSIsoft are common choices for storing and accessing time-series telemetry. These systems are usually already in place. - AMI analytics: Itron Analytic Services and the Landis+Gyr Analytics platform both provide meter-event analysis that can feed upstream asset models. - ADMS platforms: GE Vernova GridOS and Schneider Electric EcoStruxure ADMS include built-in analytics for feeder health and can surface anomalies to grid operators in context. - Dedicated AI platforms: Several third-party vendors offer asset health scoring products that integrate with SCADA historians and ERP systems. These sit alongside the CIS and ERP as read-only consumers of asset data. - ERP and EAM: SAP PM (Plant Maintenance) and Oracle Utilities’ work management modules are the destination for approved work orders. The AI layer should write back to these systems, not replace them. ## Integration Patterns That Work The cleanest integration pattern sends asset risk scores to a planner dashboard daily, with exception alerts for assets that cross a high-risk threshold. Planners review the list, accept or reject individual recommendations, and push accepted items to the work order queue in the ERP. This keeps a human in the loop for every maintenance decision while still getting the efficiency benefit of machine-assisted prioritization. SAP IS-U shops can use IDocs or BTP APIs to sync work order data. Oracle Utilities shops typically use the OUAF integration framework. Both paths are mature enough that a greenfield integration project for predictive maintenance data flow is manageable in months rather than years. ## Data Quality Is the Real Constraint Every utility AI vendor will tell you their models are powerful. The honest bottleneck in most deployments is data quality: inconsistent asset ID tagging across SCADA and the ERP, work orders closed without failure codes, and sensors that have drifted out of calibration. Addressing these before a modeling project starts is the difference between a pilot that proves value and a pilot that produces noise. For utilities exploring this in the context of broader utility data management, see our article on maximizing utility data management with AI. For the strategic overview of AI trends across the energy sector, see AI transforming the energy sector: key trends. The AvanSaber consulting team advises on data readiness assessments and integration architecture for utilities pursuing predictive maintenance programs. ## Frequently asked questions What is the difference between rule-based alarming and AI-based anomaly detection? Rule-based alarms fire when a single reading crosses a fixed threshold. AI anomaly detection looks at combinations of readings over time and flags when the overall pattern deviates from that asset's own baseline, catching gradual degradation that threshold alarms miss. Do utilities need a dedicated AI platform, or can this be done in the existing ERP? Most utilities use a separate analytics layer that reads from SCADA, AMI, and the ERP and writes back work order drafts. The ERP and CIS remain authoritative for asset records and work orders; the AI layer adds the scoring capability without replacing those systems. --- ## AI Transforming the Energy Sector: Honest Trends Overview URL: https://www.utilitieslabs.com/ai-transforming-energy-sector-key-trends/ # AI Transforming the Energy Sector: Honest Trends Overview AvanSaber Research/March 24, 2024/Updated June 2, 2026/3 min read Accurate assessments of AI in the energy sector are more useful than optimistic ones. Some AI applications are well-established and operating at scale. Others are genuinely in early stages. This overview covers both, with links to articles that go deeper on each use case. ## Where AI Is Delivering Results Today Predictive asset maintenance is the most mature category. Utilities that have invested in sensor infrastructure, clean work order history, and an integration between SCADA data and their EAM or ERP can run AI models that score transformer, pump, and feeder risk accurately enough to shift maintenance from time-based to condition-based. The detailed mechanics are covered in AI predictive maintenance for utilities. Volt/VAR optimization (VVO) and conservation voltage reduction (CVR) are running in production at many North American utilities on platforms from GE Vernova GridOS and Schneider Electric EcoStruxure ADMS. These applications produce measurable reductions in distribution losses and customer energy consumption on circuits where they are deployed. Short-term load forecasting at the substation and feeder level has been an operational tool for years. Machine learning models have improved forecast accuracy meaningfully over classical statistical approaches, particularly for circuits with significant DER penetration where historical patterns are less stable. Fault detection and outage management on ADMS platforms now include AI-assisted fault location and switching recommendations that compress the time from fault to customer restoration. Oracle Utilities NMS and GE Vernova GridOS both include production implementations. The grid operations details are in AI solutions for smart grid management. Customer energy analytics from platforms like Oracle Opower (behavioral insights) and Bidgely (load disaggregation) are used by utilities to run demand response programs and deliver personalized energy reports. These connect to the CIS billing and rate configuration layer. Water utility applications including leak detection, pressure optimization, and water quality anomaly detection are a growing category with production deployments. The specifics are covered in AI in water utility management. ## What Is Still Maturing Fully autonomous distribution switching without human confirmation remains primarily in research and pilot status. The liability and regulatory complexity around automated control decisions in a grid with thousands of customers has slowed deployment of anything beyond pre-programmed automation sequences. AI for complex customer interactions, such as AI agents that negotiate payment arrangements or handle regulatory complaints, is promising but not yet reliable enough to operate without human review in most utility contact center environments. Current deployments work best as routing and triage aids, not autonomous resolution systems. Generative AI tools like SAP Joule and similar offerings are beginning to appear in utility ERP workflows for knowledge retrieval and workflow guidance, but their integration with transaction-level CIS and billing processes requires careful governance to avoid incorrect data modifications. ## The CIS and ERP as the Foundation A pattern that appears across every working AI deployment in utilities is that the CIS and ERP are not competitors to AI; they are the foundation it depends on. Clean meter data from the CIS, accurate asset records from the ERP, and reliable work order history from the EAM are what make AI models produce useful outputs rather than noisy ones. Utilities that approach AI as a replacement for their core platforms tend to encounter integration problems that consume most of the promised efficiency gains. Utilities that treat the CIS and ERP as the systems of record and design AI as a read-query-recommend layer against those systems tend to deploy faster and sustain results. For the SAP and Oracle platform context, see integrating AI with SAP and Oracle for utility operations and top AI software for utility companies. For a deeper look at managing the data foundation that AI depends on, see maximizing utility data management with AI. ## Frequently asked questions Is AI in utilities mostly hype or is it delivering results today? Both statements are partially true. Applications like Volt/VAR optimization, load forecasting, and predictive maintenance on specific asset classes are running in production at utilities globally and are delivering measurable results. Broader claims about AI replacing utility planners or fully autonomous grid operation remain far from production reality. How does AI interact with existing CIS and ERP platforms? In almost every current deployment, the CIS and ERP remain the systems of record for customer data, billing, and asset registers. AI tools read from these systems and from operational sources like SCADA and AMI, then write back recommendations or draft transactions. The CIS and ERP are not being replaced by AI; they are being used as the authoritative data foundation that AI needs to function reliably. --- ## AI and XR in Utility Customer Service: What Is Real Now URL: https://www.utilitieslabs.com/ai-xr-customer-service-next-gen-utility-support-solutions/ # AI and XR in Utility Customer Service: What Is Real Now AvanSaber Research/August 22, 2025/Updated June 2, 2026/1 min read “AI and XR for next-gen utility customer service” bundles a real shift with a speculative one. The honest split: AI tied to the customer information system is delivering value now, while XR mostly helps the people who support customers, not the customers themselves. ## AI that works today Two patterns are in production at utilities: - Customer-facing AI. Chatbots and voice assistants that answer the high-volume questions (what is my bill, did my payment post, when will power return) by reading the account from the CIS. They deflect calls only when wired to live data, so the CIS connection is the whole game. - Agent assist. AI that drafts responses, summarizes account history, and surfaces the right next action from SAP IS-U, Oracle CC&B, or Cayenta CIS so human agents resolve faster. This is often higher-value than full automation because it keeps a person in the loop. The engagement layer matters here: Cayenta’s Silverblaze portal, Oracle’s digital self-service and Opower, and SAP’s self-service for IS-U are where these surface to customers. ## Where XR actually helps XR (AR and VR) on the customer side is mostly experimental. Its real near-term payoff is internal: remote-expert assistance for field crews and guided field service, where a support specialist sees what a technician sees and annotates it. That is a support function, not a customer channel. The field-work detail is in SAP augmented reality for utilities. ## The takeaway Invest in CIS-connected AI (chatbots and agent-assist) now, because it deflects calls and speeds resolution against real account data. Treat customer-facing XR as a pilot, and use AR where it belongs, supporting field and support staff. The broader CX picture is in customer experience in utilities. ## Frequently asked questions Is AI useful for utility customer service today? Yes, in two places: customer-facing chatbots that handle bill, payment, and outage questions, and agent-assist that drafts responses and surfaces account context from the CIS. Both depend on a live connection to SAP IS-U, Oracle CC&B, or Cayenta CIS so answers reflect the real account. Where does XR fit in utility customer service? XR (AR/VR) is niche on the customer side. Its real near-term use is supporting the field and support staff (guided field service, remote expert assistance), not end customers. Treat customer-facing XR as experimental. --- ## AR System Integration in SAP Utilities: Q4 2026 Deprecation URL: https://www.utilitieslabs.com/ar-system-integration-sap-utilities-2026-deprecation/ # AR System Integration in SAP Utilities: Q4 2026 Deprecation AvanSaber Research/August 7, 2026/7 min read Utilities that started an augmented reality programme on SAP now have a date to work against. SAP’s mobile services documentation carries a deprecation notice for the platform’s AR feature, with removal scheduled for the fourth quarter of 2026. The notice is short, easy to miss, and it does not mean what most summaries of it claim. This piece separates what is actually going away from what is not, sets out the integration paths that remain, and gives a replan checklist for anyone with a pilot in flight. ## Key takeaways - SAP states that “the Augmented Reality (AR) feature in SAP Mobile Services is deprecated and will be removed from the mobile services cockpit UI in Q4 2026”, and advises customers to plan for alternative solutions before that date. - The scope is the platform configuration surface, not augmented reality on SAP data. Field applications and partner AR extensions are unaffected. - The durable integration contract was never the AR feature. It is the OData service layer over SAP IS-U and S/4HANA Utilities, which is what every remaining path depends on. - Partner extensions now carry the richest utility workflows. TeamViewer Frontline, for example, routes a live AR call from the work order a technician is executing and returns session recordings, images, and chat transcripts into SAP document management. - Utilities with a running pilot should audit which surface it depends on this quarter, not next year. Most pilots were scoped before the notice existed. ## What SAP actually deprecated The notice appears in the SAP Mobile Services documentation, in the administration guide for Fiori UI configuration. Its wording is unambiguous: “The Augmented Reality (AR) feature in SAP Mobile Services is deprecated and will be removed from the mobile services cockpit UI in Q4 2026.” The guidance attached to it is equally short: plan for alternative solutions before Q4 2026. SAP does not name a successor. The feature remains functional during the notification phase, which is normal for a deprecation of this type and is also why it is easy to leave unattended until the removal lands. What this covers is the AR capability administered from the mobile services cockpit: the platform surface where AR content was configured for SAP mobile applications. That is a narrow slice of any real deployment. It is not the field application, not the device layer, and not the data path into your ERP. ## What the deprecation does not touch Three things survive intact, and they account for most of what a utility AR programme actually consists of. SAP Service and Asset Manager. The mobile field application for maintenance and service work is a separate product with its own roadmap. Crews keep the same work order, notification, and confirmation flows they use today. Partner AR extensions. Independent AR platforms integrate with SAP through their own certified extensions rather than the deprecated cockpit feature. TeamViewer Frontline is the clearest example in asset management: its xAssist extension provides augmented reality remote assistance calls from smart glasses or a mobile device, its xInspect extension drives AR-assisted inspection workflows, and the integration detects the SAP user and tracks the work order being executed so the session stays traceable. Session recordings, meter readings, troubleshooting images, call records, and chat transcripts pass back into SAP document management. The OData layer. Everything of consequence in an AR integration flows over OData services, typically brokered through SAP Business Technology Platform. That contract is untouched. ## The three integration paths that still hold Once the cockpit feature is out of the picture, a utility has three architectures to choose between. They differ less in what the technician sees than in who owns the integration. Path one: field application plus a partner AR extension. SAP Service and Asset Manager stays the system the crew works in, and an AR extension is triggered in context from the work order. The vendor owns the integration and the round trip of captured media back into SAP. This is the shortest route to a working deployment and the most constrained, because the workflows available are the ones the extension supports. Path two: a direct OData integration to IS-U or S/4HANA Utilities. The AR client reads equipment, functional location, work order, and service notification records straight from the ERP, and writes back confirmations, measurement documents, and attachments. You own the service definitions and the mapping. This is the right path when the AR workflow is specific to your operation and no packaged extension models it, and it is the path that ages best, because the integration surface is a standard interface rather than a product feature. Path three: a standalone AR platform with a middleware layer. The AR vendor never talks to the ERP directly. An integration layer on SAP Business Technology Platform mediates, holds the mapping, and manages offline reconciliation. It costs more to build and it is the only sensible design when several AR tools, or several plants with different SAP releases, have to be served at once. The choice is a maintenance decision, not a technology decision. Path one puts the burden on the vendor, path two on your integration team, path three on your platform team. ## What this means for a utility specifically Utility AR work does not fail on the AR. It fails on the data the overlay draws from, which is why the deprecation is a useful prompt. The objects that matter are the ones already carrying your field work: equipment and functional location records, service notifications, work orders, and, for metering, the order and measurement document chain. If those records are incomplete or the functional location hierarchy does not match what a crew finds on site, no AR front end improves the outcome. It surfaces the gap faster, which is valuable, but that is a data programme rather than an AR programme. Our overview of what a utility CIS actually covers sets out where those records live in an IS-U landscape. Connectivity is the second constraint, and substations are the reason. An AR workflow that assumes a live session fails in exactly the locations where hands-free guidance is most useful. Offline capture with deferred sync is the requirement, and it needs to be tested in the worst site you have, not the best. Third, safety approval sits outside the software. Anything worn or held during switching or energised work goes through the same review as any other field equipment, and that review runs on its own calendar. Utilities that treat it as a procurement afterthought lose a quarter. ## A replan checklist before Q4 2026 - Identify the dependency. Determine whether any live or piloted AR workflow is configured through the mobile services cockpit. If it is, it has a removal date. - Inventory the workflows, not the tools. List the specific tasks AR supports today: inspection rounds, remote expert calls, guided replacement procedures. Tools are replaceable; the workflow list is what you are protecting. - Pin the integration contract. Document the OData services each workflow reads and writes. This is the artefact that makes a vendor change a configuration exercise instead of a rebuild. - Re-test offline behaviour. Deferred sync paths tend to be the least exercised part of a pilot and the first to break under a new integration path. - Decide the ownership model. Vendor extension, in-house OData integration, or a platform middleware layer. Pick deliberately, because it determines who is on call when a sync fails at a substation. - Set a measurable success metric before you migrate. Inspection completion time and first-time-fix rate are objective and available in the same systems the AR workflow already touches. Our longer treatment of SAP augmented reality in utility field service covers the deployment scoping question, and the AR plus ERP asset management piece covers the work order integration point in more depth. ## Where AR still does not pay The deprecation is a reasonable moment to cut the parts of a programme that were never tied to a task. AR remains hard to justify for general field work where a rugged tablet showing the same record performs identically, for training scenarios that a well-made video covers, and for any workflow where the underlying asset data is known to be unreliable. It earns its place on repeatable, high-consequence tasks: substation inspection, complex switching, and first-time-fix support on equipment a crew sees rarely. ## The verdict The removal of an admin feature from a cockpit UI is a small event that happens to be well timed. It forces the question most AR programmes never answered, which is what the integration actually rests on. If the answer is a clean OData contract over IS-U or S/4HANA Utilities and a short list of workflows with measurable outcomes, the deprecation costs you a configuration change. If nobody can answer it, that is worth knowing in 2026 rather than in the quarter the feature disappears. If you want a second set of hands on that audit, the AvanSaber team works on SAP utility integrations case by case. ## Frequently asked questions What exactly is SAP deprecating in Q4 2026? The Augmented Reality feature inside SAP Mobile Services. SAP's own documentation states that the feature is deprecated and will be removed from the mobile services cockpit UI in Q4 2026, and tells customers to plan for alternative solutions before that date. It is the platform-side AR configuration surface that goes away, not augmented reality as a way of working with SAP data. Does the deprecation break AR in SAP Service and Asset Manager? No. SAP Service and Asset Manager remains the field application, and partner AR extensions that plug into it continue to work through their own integrations. What disappears is the AR configuration path administered from the mobile services cockpit, so any deployment that relied on that surface needs a replacement before Q4 2026. How does an AR client actually connect to SAP IS-U data? Through OData services, usually mediated by SAP Business Technology Platform. The AR front end reads equipment, functional location, work order, and service notification records, then writes results back as confirmations, measurement documents, or attachments. The integration contract is the OData service definition, which is why the AR vendor choice matters less than the quality of the service layer underneath it. What should a utility do if an AR pilot is already running? Establish which surface the pilot depends on. If AR content is configured in the mobile services cockpit, that path has a removal date and needs a migration plan now. If the pilot runs on a partner extension or a direct OData integration, the deprecation does not affect it, but the audit is still worth doing because most pilots were scoped before the notice was published. Is AR worth keeping in the utility field toolkit at all? On specific, repeatable tasks, yes. Substation inspection, complex switching, and first-time-fix support are the workflows where surfacing asset data hands-free changes the outcome. As a general-purpose field technology it is still hard to justify, and the deprecation is a reasonable moment to cut the parts of a programme that were never tied to a measurable task. --- ## Best Utility Technology in 2026: Cloud ERP, AI, and CIS URL: https://www.utilitieslabs.com/best-utility-technology-2025-cloud-erp-ai-bots-smart-tools/ # Best Utility Technology in 2026: Cloud ERP, AI, and CIS AvanSaber Research/August 22, 2025/Updated June 1, 2026/1 min read Utility technology lists tend to be long and vague. The useful version is short and ordered: get the system of record right first, then layer data and AI where they pay off. This guide ranks what matters for a utility in 2026. ## Start with the system of record Everything else depends on a solid CIS and billing platform: - SAP IS-U / S/4HANA Utilities for large, SAP-estate utilities. - Oracle CC&B for large, Oracle-estate utilities. - Cayenta CIS for municipal and mid-market utilities that want an integrated suite. If this layer is wrong, no amount of AI on top fixes it. The Oracle vs SAP comparison and the Cayenta CIS review cover the selection. ## Then data and analytics Meter data management (SmartWorks, Oracle MDM, or SAP MDUS) and analytics on top of HANA or a cloud warehouse turn raw meter reads into something a utility can act on: loss detection, demand insight, and customer usage programs. ## Then AI, where it is specific AI earns its place on defined tasks: outage prediction, load forecasting, AI-assisted customer service, and metering anomaly detection. Adopt it where it targets one measurable outcome. Treat any broad “AI platform for utilities” pitch with the same procurement discipline you would apply to a core system. ## The order that works Get the CIS right, make the meter data usable, then add AI to specific processes. Utilities that follow that order get value. Utilities that buy AI before fixing the system of record usually do not. ## Frequently asked questions What technology should a utility prioritize in 2026? Start with the system of record: a CIS and billing platform that fits your size (SAP IS-U or S/4HANA Utilities, Oracle CC&B, or Cayenta CIS for the mid-market). Layer meter data management and analytics on top, and add AI where it targets a specific, measurable task rather than as a general upgrade. Is AI worth adopting for utilities yet? In narrow places, yes: outage prediction, load and demand forecasting, AI-assisted customer service, and anomaly detection in metering. The returns come from pointing AI at one well-defined process, not from buying a broad AI platform. --- ## All Articles URL: https://www.utilitieslabs.com/blog/ # All articles The full UtilitiesLabs archive: 106 articles on SAP IS-U, Oracle CC&B, Cayenta CIS, and utility billing. ## 2026 - Oracle CC&B Data Model: How Utility Billing ResolvesAvanSaber Research · Aug 11, 2026 - AR System Integration in SAP Utilities: Q4 2026 DeprecationAvanSaber Research · Aug 7, 2026 - Utility Customer Experience: A CIS-Centric GuideAvanSaber Research · Jul 25, 2026 - AI in Utilities: Where It Actually Lands in 2026AvanSaber Research · Jul 10, 2026 - SAP CIS for Utilities: What IS-U and FI-CA Actually CoverAvanSaber Research · Jun 30, 2026 - SAP vs Oracle vs Cayenta: How They Are Adding AI in 2026AvanSaber Research · Jun 14, 2026 - SAP Joule for Utilities: Agentic AI on IS-U and S/4HANAAvanSaber Research · Jun 8, 2026 - AI-Native ERP for Utilities: What Legacy CIS Don't Do YetAvanSaber Research · Jun 2, 2026 - SAP On-Prem AI U-Turn: Joule on ECC and S/4HANA for UtilitiesAvanSaber Research · May 19, 2026 - Agentic AI for the Utility Back Office: Collections and DunningAvanSaber Research · May 17, 2026 - Cayenta Cayla: An AI Assistant Inside a Utility CISAvanSaber Research · May 11, 2026 - Oracle Opower and Energy & Water Intelligence: Utility AI in 2026AvanSaber Research · Apr 21, 2026 - The Utility CIS Modernization Wave: Legacy to Cloud and AIAvanSaber Research · Apr 14, 2026 - From Meter-Centric to Service-to-Cash: AI and Utility CIS ArchitectureAvanSaber Research · Apr 9, 2026 ## 2025 - Digital Twins for Grid Management: AI, GIS, and ADMSAvanSaber Research · Dec 2, 2025 - AI and XR in Utility Customer Service: What Is Real NowAvanSaber Research · Aug 22, 2025 - Best Utility Technology in 2026: Cloud ERP, AI, and CISAvanSaber Research · Aug 22, 2025 - Customer Experience in Utilities: The CIS-Led Digital TransformationAvanSaber Research · Aug 22, 2025 - Utility ERP and CIS Implementation: Best PracticesAvanSaber Research · Aug 22, 2025 - AI for Utility Operations on SAP IS-U and Oracle CC&BAvanSaber Research · Aug 22, 2025 - Smart Grid Optimization: ADMS, DERMS, and AI for the Distribution GridAvanSaber Research · Aug 22, 2025 - AI and ERP Implementation for Utilities: A 2026 Practitioner ChecklistAvanSaber Research · Aug 6, 2025 - SAP S/4HANA and Oracle Cloud Utilities: What Changed in 2024-25AvanSaber Research · Apr 9, 2025 - Cayenta CIS Review (2026): Fit, Features and AlternativesAvanSaber Research · Mar 29, 2025 - AI Predictive Maintenance for Utilities: Cut DowntimeAvanSaber Research · Jan 21, 2025 - SAP BTP for Utilities: Extend IS-U, Keep the Core CleanAvanSaber Research · Jan 16, 2025 - SAP BTP AI and Joule for Utility Operations on IS-U and S/4HANAAvanSaber Research · Jan 10, 2025 ## 2024 - AI and AR for Utility Asset Inspection and Field ManagementAvanSaber Research · Dec 11, 2024 - AI for Energy Distribution: Cutting Losses and Balancing the GridAvanSaber Research · Nov 5, 2024 - AI and XR for Utility Customer Service: Proactive and Field SupportAvanSaber Research · Nov 5, 2024 - AI and XR Digital Twins: Operator Visualization in UtilitiesAvanSaber Research · Nov 3, 2024 - AI and AR for Utility Safety and Leak DetectionAvanSaber Research · Oct 29, 2024 - FLISR and Self-Healing Grids: How ADMS Uses AI to Restore PowerAvanSaber Research · Oct 11, 2024 - AI and XR in Smart City Utilities: Grounded AssessmentAvanSaber Research · Aug 31, 2024 - SAP Augmented Reality for Utilities: Field Service Guide (2026)AvanSaber Research · Aug 2, 2024 - AI for Smart Grid Operations: What Actually Works TodayAvanSaber Research · Jul 31, 2024 - AI for Utility Operations: The Use Cases That Pay OffAvanSaber Research · Jul 21, 2024 - AI for Energy Consumption Analysis in Utilities: What WorksAvanSaber Research · Jul 21, 2024 - AI and XR in Utility ERP: SAP Joule/BTP vs Oracle Opower and OUAF AIAvanSaber Research · Jul 21, 2024 - AI Software for Utility Companies: 2026 Buyer Guide by Use CaseAvanSaber Research · Jul 21, 2024 - XR Visualization of Real-Time SAP IS-U and S/4HANA DataAvanSaber Research · Jul 1, 2024 - VR for ERP Process Simulation and Design Review in UtilitiesAvanSaber Research · Jun 10, 2024 - AI in Water Utility Management: Leaks, Quality, and PressureAvanSaber Research · Jun 6, 2024 - AR + ERP for Utility Asset Management and Work OrdersAvanSaber Research · May 16, 2024 - AI Tools for Predictive Maintenance in Utilities: The PipelineAvanSaber Research · May 2, 2024 - AI in Utility Customer Service: Chatbots and Agent Assist on the CISAvanSaber Research · Apr 19, 2024 - AI Analytics on Utility ERP and CIS Data: A Practical GuideAvanSaber Research · Apr 15, 2024 - XR for ERP Data Visualization: Niche Value and Real ConstraintsAvanSaber Research · Apr 2, 2024 - AI in Utility Billing: Where It Helps on IS-U, CC&B, and CayentaAvanSaber Research · Mar 31, 2024 - AI Transforming the Energy Sector: Honest Trends OverviewAvanSaber Research · Mar 24, 2024 - VR for Utility Workforce Training: What WorksAvanSaber Research · Mar 15, 2024 - Utility Data Management with AI: MDM, Quality, and IntegrationAvanSaber Research · Mar 2, 2024 - AR and VR with Oracle Utilities ERP: Honest Scope and Use CasesAvanSaber Research · Jan 30, 2024 - AI Data Analytics in Utilities: Loss, Demand, and Asset OutcomesAvanSaber Research · Jan 10, 2024 ## 2023 - Deregulated-Market ERP for Utilities: SAP IS-U and CC&BAvanSaber Research · Dec 23, 2023 - SAP IDocs for Utilities: Integrating IS-U and Meter DataAvanSaber Research · Dec 23, 2023 - SAP ABAP for Utility IS-U: Performance, Clean Core, and BTPAvanSaber Research · Nov 29, 2023 - Cayenta CIS Data Model and Cayla AI: What Operators Need to KnowAvanSaber Research · Nov 12, 2023 - AI Across the Oracle Utilities Suite: CC&B, Opower, and NMSAvanSaber Research · Nov 4, 2023 - Cayenta CIS as a SAP Alternative: The Mid-Market CaseAvanSaber Research · Oct 30, 2023 - Oracle Database Tuning for CC&B and Utility Billing JobsAvanSaber Research · Oct 24, 2023 - Oracle CC&B in Deregulated Markets: Switching and Market MessagingAvanSaber Research · Oct 15, 2023 - AI in SAP Utilities: Joule, BTP, and Billing ExceptionsAvanSaber Research · Oct 10, 2023 - SAP Analytics Cloud and BTP AI: BI for Utilities on SAP IS-UAvanSaber Research · Sep 29, 2023 - Oracle AI and Data Models for Utility Analytics on CC&B DataAvanSaber Research · Sep 19, 2023 - Innovation in Utilities: What's Actually Changing in CIS and ERPAvanSaber Research · Sep 18, 2023 - SAP and Oracle Utility Suite Roadmaps: Where Each Vendor Is HeadingAvanSaber Research · Aug 23, 2023 - SAP and Oracle in Utility Operations: Billing ThroughputAvanSaber Research · Aug 23, 2023 - Modern Utility Software, Mapped: CIS, ERP, MDM, OMS, and GISAvanSaber Research · Aug 23, 2023 - Buying Modern Utility Software: A Practical Evaluation GuideAvanSaber Research · Aug 23, 2023 - Oracle Utilities Pre-Project Readiness: Scoping ConsiderationsAvanSaber Research · Aug 4, 2023 - SAP HANA for Utilities: Real-Time Data and S/4HANAAvanSaber Research · Jul 30, 2023 - Customer Service and Process Transformation via SAP IS-UAvanSaber Research · Jul 28, 2023 - CC&B Implementation Failure Modes: What Goes WrongAvanSaber Research · Jul 26, 2023 - S/4HANA Utilities: The Forward View of HANA in Utility ManagementAvanSaber Research · Jul 22, 2023 - Real-Time Interval Data Processing with SAP HANA UtilitiesAvanSaber Research · Jul 16, 2023 - FI-CA Day-to-Day: How SAP Utilities Finance Runs Financial OperationsAvanSaber Research · Jul 11, 2023 - SAP HANA for Utility Operations: Faster Billing and ManagementAvanSaber Research · Jul 9, 2023 - What an SAP IS-U Implementation Changes Operationally at a UtilityAvanSaber Research · Jul 6, 2023 - Oracle Utilities Implementation: A Practical Guide to CC&B ProjectsAvanSaber Research · Jul 4, 2023 - Smart Water Utility Software: SCADA, NRW, and CIS for WaterAvanSaber Research · Jul 1, 2023 - GIS in Utilities: Esri ArcGIS, Network Models, and IntegrationAvanSaber Research · Jun 28, 2023 - How Regulatory Change Lands on CIS and Billing SystemsAvanSaber Research · Jun 26, 2023 - Digital Twins in Utilities: What's Real vs. What's Still HypeAvanSaber Research · Jun 24, 2023 - Automation in Utility Meter-to-Cash: CIS, RPA, and AI-Driven SavingsAvanSaber Research · Jun 21, 2023 - 5G in the Utilities Sector: Real Uses, Real LimitsAvanSaber Research · Jun 19, 2023 - Sustainability in Utilities: Software That Makes It MeasurableAvanSaber Research · Jun 17, 2023 - Disaster Recovery for Utility IT: CIS, OMS, and Billing UptimeAvanSaber Research · Jun 15, 2023 - Building the Business Case for SAP Utilities Finance and FI-CAAvanSaber Research · Jun 13, 2023 - SAP Utilities Finance: FI-CA, Billing, and the Utility Revenue ProcessAvanSaber Research · Jun 13, 2023 - Oracle vs SAP for Utilities: 2026 ERP Comparison (CC&B, IS-U, S/4HANA)AvanSaber Research · Jun 11, 2023 - Utility Customer Portals: Self-Service That Cuts CallsAvanSaber Research · Jun 11, 2023 - How to Choose Utility CIS and Billing Software: A Selection GuideAvanSaber Research · Jun 8, 2023 - Blockchain in Utilities: Peer-to-Peer Trading and REC TrackingAvanSaber Research · Jun 8, 2023 - Predictive Analytics in Utility Management: Forecasting and AssetsAvanSaber Research · Jun 8, 2023 - Renewables and Utility Software: Billing, DERMS, and ForecastingAvanSaber Research · Jun 8, 2023 - OT/IT Cybersecurity for Utilities: NERC CIP and BeyondAvanSaber Research · Jun 8, 2023 - SAP IS-U for Utilities: Modules, S/4HANA Utilities Migration, and FitAvanSaber Research · Jun 8, 2023 - Smart Grids Explained: AMI, ADMS, and DERMS in Utility OperationsAvanSaber Research · Jun 8, 2023 - Emerging Utility Tech: AMI, DERMS, AI, and Cloud CISAvanSaber Research · Jun 7, 2023 - IoT and AMI Sensors: The Data Path Into MDM, CIS, and ADMSAvanSaber Research · Jun 7, 2023 - Digital Transformation in Utilities: A CIS and Data-First GuideAvanSaber Research · Jun 7, 2023 - The Future of Utilities: Business Model Shifts to WatchAvanSaber Research · Jun 7, 2023 - AI and Machine Learning in Utilities: Hub OverviewAvanSaber Research · Jun 7, 2023 - Utility Data Platforms: AMI, SCADA, and CIS Data at ScaleAvanSaber Research · Jun 7, 2023 - Cloud for Utility Software: S/4HANA, Oracle, and CayentaCloudAvanSaber Research · Jun 7, 2023 --- ## Cayenta Cayla: An AI Assistant Inside a Utility CIS URL: https://www.utilitieslabs.com/cayenta-cayla-ai-assistant-utility-cis-review/ # Cayenta Cayla: An AI Assistant Inside a Utility CIS AvanSaber Research/May 11, 2026/Updated September 24, 2026/4 min read Utility AI has spent several years as a conference theme. In 2026, some CIS vendors are moving past the slide deck and into production. Cayenta, a division of Harris Computer, is one of them. Its AI product, Cayla (Cayenta Artificial Intelligence), is now described on the Cayenta product pages as a live capability, and a real utility deployment signed in March 2026 gives it a concrete reference point. This review covers what Cayla is, what the Lakeland deployment tells us, and where its limits are at this early stage. ## What Cayla is: Agent and In-App Assistant Cayenta describes Cayla in two distinct delivery modes, per its product pages. The first is Cayla Agent, which manages customer inquiries across voice and digital channels. This is the outward-facing layer: a customer calls or contacts the utility through a digital channel, and Cayla Agent handles the interaction, routing and resolving without necessarily requiring a human agent. The second is the In-App Assistant, oriented toward internal staff. Per Cayenta’s description, it can answer questions, analyze accounts, summarize operations, and execute workflows. Demonstrated actions include starting a move-in and setting up a payment plan. These are not AI generating a report for a manager to act on later. They are AI taking action inside the CIS directly on behalf of a staff user, which puts them in the agentic category. The distinction matters for governance. An AI that answers questions carries different risk than one that executes transactions. Utilities deploying agentic workflow tools need clear policies on which actions require human confirmation and what the rollback procedure is for an incorrectly executed transaction. ## The Lakeland deployment: what the record shows The clearest public evidence of Cayla in production comes from Lakeland, Florida. The City of Lakeland Commission approved an agreement on March 2, 2026 for Lakeland Electric to implement Cayla, built in partnership with Dial AI. The agenda memo targets a production go-live around May 1, 2026, integrated with Lakeland Electric’s existing Cayenta CIS. That makes Lakeland a first-mover, not a long-tenured reference site. At the time this post publishes, the deployment is weeks old at most. There is no multi-year performance record to evaluate yet. What the agreement confirms is that Cayla is not vaporware. A municipality committed budget, went through a commission approval process, and signed with a specific partner (Dial AI) and a specific go-live target. For a product that only recently appeared on the Cayenta product pages, that is meaningful early validation. ## Where Cayla fits inside the Cayenta CIS The Cayenta suite is broader than the core CIS, as covered in the detailed Cayenta CIS review. Beyond Cayenta’s own work, financial and HCM modules, Cayenta offers add-ons powered by sister Harris products: ServiceLink for mobile workforce, Silverblaze for customer self-service portals, SmartWorks for meter data management, and DataVoice for outage management. Cayla sits inside this stack. Its primary advantage is that it already has authorized access to account data, billing history, service agreements, and CIS workflows. No integration project is required to give the AI access to data for a move-in or payment plan setup. Standalone AI tools bolted onto a CIS through API connections are feasible but add cost and a surface area for synchronization errors. For a Cayenta customer, Cayla removes that tax. Cayla Agent and the Silverblaze self-service portal address overlapping but different problems: Cayla handles inbound real-time inquiries while Silverblaze handles asynchronous self-service. The combination matters for utilities focused on contact-center deflection, a topic explored in the case for customer portals in utility services. ## How Cayla compares to Oracle and SAP utility AI at a high level Oracle’s utility AI portfolio is larger and older. Oracle Opower has run utility programs since 2009, and Oracle reports 44.6 million residential households enrolled as of March 2026. Oracle added AI call summarization to its Customer Platform in May 2025 and AI anomaly detection in Meter Data Management in June 2025. These are incremental additions to an enterprise platform with a deep installed base. SAP Joule is the cross-application AI assistant SAP is embedding across its S/4HANA suite, including S/4HANA Utilities, targeting similar workflow acceleration but in the SAP ecosystem. Cayla’s proposition is narrower scope and deeper CIS integration for a specific customer segment. Cayenta customers at the municipal and mid-market tier get an AI that runs inside the tools they already use, without a separate AI platform procurement. That matters for a 200,000-account municipal utility not staffed to run a large-scale AI program. For the full CIS comparison, the Oracle vs SAP utilities analysis provides enterprise context. ## Honest limits at this stage Cayla is early. The honest assessment of any AI product with one reference site on a short deployment timeline is that the long-term operating record does not exist yet. Prospective buyers should ask: - What is the error rate on agentic transactions (move-ins, payment plan setup), and how are exceptions handled? - What governance controls restrict which actions the In-App Assistant can execute without human confirmation? - What does the Dial AI partnership mean for support, model ownership, and continuity if the partnership changes? - How does Cayla handle utility-specific edge cases: life-support accounts, delinquency restrictions, complex rate structures? A CIS-native AI that can answer questions and execute transactions inside the system of record is a credible design for utility customer service. The open question is execution at scale over time. Utilities considering Cayla should track the Lakeland deployment’s public reporting and request updated reference sites before committing. ## Frequently asked questions What is Cayenta Cayla AI? Cayla (Cayenta Artificial Intelligence) is the AI layer built into the Cayenta CIS suite. It ships in two forms: Cayla Agent, which handles customer inquiries across voice and digital channels, and an In-App Assistant for internal staff that can answer questions, analyze accounts, summarize operations, and execute workflows such as move-ins and payment plan setup. Has Cayla been deployed at a real utility? Yes. The City of Lakeland, Florida (Lakeland Electric) signed an agreement in March 2026 to implement Cayla, built with partner Dial AI, targeting a production go-live around May 2026, integrated with its existing Cayenta CIS. How does Cayla compare to Oracle Opower or SAP Joule for utilities? They target different problems. Cayla is a CIS-embedded workflow assistant handling customer service transactions. Oracle Opower is a behavioral engagement and demand-side management platform. SAP Joule is a cross-suite AI assistant for SAP applications. Cayla's advantage is tight CIS integration for Cayenta customers; its limitation is that it is an early product with a short live deployment record. Does Cayla work with CIS platforms other than Cayenta? No. Cayla is a Cayenta product, designed to run inside the Cayenta CIS suite. It is not an independent AI layer that can be dropped onto other CIS platforms. --- ## Cayenta CIS Review (2026): Fit, Features and Alternatives URL: https://www.utilitieslabs.com/cayenta-cis-review-best-alternative-to-sap-oracle-for-utilities/ # Cayenta CIS Review (2026): Fit, Features and Alternatives AvanSaber Research/March 29, 2025/Updated September 24, 2026/6 min read Cayenta CIS comes up whenever a municipal or mid-sized utility decides SAP IS-U and Oracle Customer Care and Billing (CC&B) are heavier than the problem in front of them. The question this review answers is narrow and practical: when is Cayenta the right call, and when do you still need one of the giants? ## What Cayenta CIS actually is Cayenta is a customer information system and utility ERP suite. On its website the company describes itself as “a division of N. Harris Computer Corporation”, and Harris lists Constellation Software as its parent. Cayenta’s company history runs from its founding in 1983 as Solutions for Government, through the Cayenta name in 2000 and joining Harris in 2004, to the CayentaCloud launch in 2020 and the Cayla AI layer in 2025. That ownership matters. Constellation describes a buy-and-hold approach to the vertical software businesses it acquires, so Cayenta is not a venture-funded product at risk of a sudden pivot. Cayenta says it serves electric, water, gas and fiber utilities, from municipalities, counties and co-ops to investor-owned utilities, across the United States, Canada and the Caribbean. As of September 2026, its solutions page splits the line-up into three groups. Cayenta’s own modules - Customer Information System, the billing and customer system of record, with a 360 customer view, configurable workflows and a flexible rates engine. - Cayla (Cayenta AI), in two forms: Cayla Agent handles customer inquiries across voice and digital channels, and the In-App Assistant helps staff answer questions, analyze accounts, summarize operations and run workflows. Our Cayla review covers it in depth. - ERP modules: Work Management (work orders, labor and material costing), Financial Management (multi-fund accounting, budget control) and Human Capital Management (time and attendance, employee self-service). Add-ons powered by sister Harris Utilities products - Mobile Workforce Management, powered by ServiceLink, a field service management product for utilities that says it is part of Harris Computer Systems. - Customer Engagement Portal, powered by Silverblaze, a cloud-hosted portal, mobile app and smart forms product that describes itself as part of the Harris family. - Meter Data Management, powered by SmartWorks, “a division of N. Harris Computer Corporation” that markets its platform as SmartWorks Compass. - Outage Management System, powered by DataVoice, an outage management vendor that Harris acquired in October 2017 and whose line also covers customer IVR, web and mobile outage communication. Deployment and services - CayentaCloud: single-tenant SaaS in dedicated environments, backed by SOC 2 certification. On-premises deployment is still offered. - Services: Cayenta leads implementations itself and, after go-live, offers CayOps subscription support, CayEngage, learning services, health checks, managed services for on-premises customers and integration services (services page). Most Cayenta reviews stop at “flexible CIS” and never draw this line, yet it is exactly what a buyer needs. The CIS and ERP modules are Cayenta’s own products. The four add-ons are separate Harris business units with their own websites and teams, which Cayenta sells as integrated options. In your RFP, ask who supports each piece after go-live and whether each add-on integration is standard product or project work. ## Cayenta vs SAP IS-U vs Oracle CC&B | Dimension | Cayenta CIS | SAP IS-U / S/4HANA Utilities | Oracle CC&B | Typical buyer | Municipal, county, co-op, mid-sized; also sells to investor-owned | Large IOU, SAP-estate utilities | Large IOU, Oracle-estate utilities | Billing engine | Cayenta CIS with configurable rates engine | IS-U billing on FI-CA contract accounting | CC&B rating and billing | MDM | SmartWorks (sister Harris product, Cayenta add-on) | Third-party MDM connected through SAP’s MDUS interface | Oracle Utilities Meter Data Management | OMS | DataVoice (sister Harris product, Cayenta add-on) | Third-party or SAP partner | Oracle Utilities Network Management System | Mobile workforce | ServiceLink (sister Harris product, Cayenta add-on) | SAP Field Service Management or a partner product | Oracle Field Service | Vendors behind the full stack | Cayenta plus sister Harris businesses | SAP plus third-party MDM and OMS | Mostly Oracle | Deregulated-market depth | Adequate for most retail models | Deep, with configuration effort | Deep, with configuration effort The pattern is consistent. Cayenta’s case rests on breadth from one vendor family: the CIS, the ERP modules and the four add-ons come from Cayenta and its sister Harris businesses, which cuts the number of unrelated vendors to integrate. SAP and Oracle win when scale, complex market messaging, or an existing SAP or Oracle estate dominate the decision. Two dated facts shape the comparison. SAP ends mainstream maintenance for Business Suite 7 core applications, the ECC platform that classic SAP IS-U runs on, at the end of 2027, with optional extended maintenance to the end of 2030 (SAP maintenance strategy). A utility on classic IS-U therefore faces a CIS decision either way. And Oracle’s CC&B documentation lists release 25.4 as the latest (checked September 2026) and describes the product as designed for utilities of all sizes, so Oracle is not only a large-utility option on paper. ## Where Cayenta is the right call Choose Cayenta when most of these hold: you serve a municipal, cooperative, or mid-sized utility; you want CIS, ERP, MDM, OMS and mobile workforce from one vendor family rather than assembled from unrelated vendors; and a multi-year SAP or Oracle program would be out of proportion to your size. The integrated suite is the real advantage, because every outside vendor adds an interface to build, test and maintain, and that integration work quietly dominates utility IT budgets. ## Where you still want SAP or Oracle Choose SAP IS-U, S/4HANA Utilities, or Oracle CC&B when you are a large investor-owned utility, when you operate in a complex deregulated market with heavy supplier-message exchange, or when you already run SAP or Oracle across the rest of the business and the CIS needs to sit inside that estate. For a head-to-head on those two, read the Oracle vs SAP for utilities comparison. ## The honest verdict Cayenta CIS is a credible alternative to SAP and Oracle, not a budget compromise. For the right size of utility it is often the better choice, because Cayenta positions its CIS as a complete meter-to-cash lifecycle in one system, with the sister add-ons filling in field, portal, meter data and outage needs. The caution is procurement discipline: Cayenta does not publish pricing, so validate scope, service types, add-ons and integration count against a real quote before you treat any cost figure as fact. To score Cayenta against SAP and Oracle on the same terms, use our utility CIS selection guide and RFP scorecard. ## How to check Cayenta CIS user reviews and references Searches for the most trusted or highest rated CIS based on user reviews usually end at a star rating. For a utility CIS that is a weak signal, because a rating from a handful of reviewers cannot tell you how the system runs at a utility of your size, service mix and deployment model. Better evidence, in order: - Reference calls with matching utilities. Ask Cayenta for two or three references that match your profile: electric, water, gas or fiber; municipal, co-op or investor-owned; CayentaCloud or on-premises. Ask about data conversion, billing accuracy in the first cycles after go-live, upgrade effort and support response times. - The add-on question. Ask each reference which of ServiceLink, Silverblaze, SmartWorks and DataVoice they run, and who they call when something breaks between the CIS and an add-on. - Current users in person. Harris runs the Harris Customer Training Conference (HCTC) for users of Harris software; the 2026 event is in Denver, December 8 to 10 (Cayenta home page). - One scorecard for every vendor. Put the reference answers into the same scoring model you use for SAP and Oracle, so no vendor wins on presentation alone. ## Sources and disclosure Facts on this page were checked on September 25, 2026 against Cayenta’s home, solutions, about and services pages; the ServiceLink, Silverblaze, SmartWorks and DataVoice sites; Harris’s Constellation Software page; and the SAP and Oracle sources linked above. For the enterprise alternatives, see Oracle’s Customer Care and Billing documentation and the SAP for Utilities portal. Disclosure: UtilitiesLabs is backed by AvanSaber, which operates a SAP utilities consulting practice. That is a commercial relationship that could bias us toward SAP. We manage it by naming it, citing independent sources for comparative claims, and reviewing SAP, Oracle, and Cayenta to the same bar. See our editorial policy. ## Frequently asked questions Is Cayenta CIS a real alternative to SAP and Oracle for utilities? For municipal, cooperative and mid-sized utilities, yes. Cayenta pairs its CIS with its own work, financial and HCM modules and with add-ons from sister Harris Utilities businesses, so fewer outside vendors need to be integrated than in a typical SAP IS-U or Oracle CC&B program. For the largest investor-owned utilities with complex deregulated-market needs, SAP and Oracle still set the ceiling on scale and market-message maturity. Who owns Cayenta? Cayenta describes itself as a division of N. Harris Computer Corporation, which it joined in 2004. Harris lists Constellation Software as its parent. Cayenta was founded in 1983 as Solutions for Government and took the Cayenta name in 2000. What modules does Cayenta CIS include? Cayenta's own line-up is the Customer Information System, Work Management, Financial Management and Human Capital Management, plus the Cayla AI layer and CayentaCloud hosting. Mobile workforce management, a customer engagement portal, meter data management and outage management are sold as Cayenta add-ons powered by sister Harris products: ServiceLink, Silverblaze, SmartWorks and DataVoice, in that order (Cayenta solutions page, checked September 2026). How much does Cayenta CIS cost? Cayenta does not publish list pricing on its website. The real number depends on utility size, service types, the deployment model (CayentaCloud single-tenant SaaS or on-premises), and how many add-ons and integrations are in scope. Compare it with SAP IS-U or Oracle CC&B only on scoped quotes that cover the same modules. Where can I find Cayenta CIS user reviews? A public star rating from a handful of reviewers says little about how the system will run at your utility. The stronger evidence is reference calls with utilities that match your size, service mix and deployment model, plus conversations with current users, for example at the annual Harris Customer Training Conference. Score every shortlisted vendor on the same scorecard. --- ## Customer Experience in Utilities: The CIS-Led Digital Transformation URL: https://www.utilitieslabs.com/customer-experience-digital-transformation/ # Customer Experience in Utilities: The CIS-Led Digital Transformation AvanSaber Research/August 22, 2025/Updated June 2, 2026/2 min read Utility customer experience is unusual: customers rarely choose their utility, contact is dominated by billing and outages, and satisfaction is mostly decided by whether the basics work. So a CX program that does not start at the customer information system is decoration. The transformation that matters is making the CIS-connected experiences reliable and self-served. ## The CIS decides the experience Accurate, understandable bills come from the rating and billing engine in SAP IS-U, Oracle CC&B, or Cayenta CIS. When a bill is wrong or confusing, no front-end design saves the experience. The first CX move is billing accuracy and clear bill presentment, which is a CIS configuration problem, not a portal problem. ## Self-service that deflects calls The highest-volume calls are “what is my bill” and “am I out.” Moving those online is the core of the digital transformation: bill view and payment, usage history, outage reporting, and service start/stop/transfer. Cayenta’s Silverblaze engagement portal, Oracle Opower and Oracle’s digital self-service, and SAP’s customer self-service for IS-U all exist to surface CIS data to the customer. The detail is in the customer portals guide. ## Proactive and omnichannel communication Mature CX gets ahead of the call: outage alerts, high-bill alerts, and payment reminders pushed by SMS, email, and app, driven by events from the CIS and OMS. The shift is from reactive call handling to proactive notification. ## Energy insight as engagement Showing customers their own consumption (and how to lower it) is both a satisfaction and a demand-side lever. Oracle Opower built its name on behavioral energy reports, and Bidgely disaggregates usage by appliance from smart-meter data. These run on the meter data the CIS and MDM already hold. ## The takeaway Treat utility CX as a CIS-led program: get billing right, move the high-volume interactions to self-service, communicate proactively, and add energy insight. The systems that make or break it are the same ones covered across UtilitiesLabs, starting with the Oracle vs SAP comparison and the Cayenta CIS review. ## Frequently asked questions What drives customer experience for a utility? The billing and customer information system, more than any standalone CX tool. Accurate bills, working self-service (bill view, payment, outage reporting, service moves), and proactive communication all depend on a live connection to the CIS (SAP IS-U, Oracle CC&B, or Cayenta CIS). --- ## Digital Twins for Grid Management: AI, GIS, and ADMS URL: https://www.utilitieslabs.com/digital-twins-in-utilities-transform-grid-management-with-ai-xr/ # Digital Twins for Grid Management: AI, GIS, and ADMS AvanSaber Research/December 2, 2025/Updated June 2, 2026/4 min read Utility grid digital twins have moved from concept to working tooling over the past several years, but the term still attracts hype that obscures what the technology actually does today. This piece covers the practical picture: what a grid digital twin is built from, where AI adds genuine value, and where XR fits, which is narrower than most vendor decks suggest. ## What a Grid Digital Twin Actually Is A utility grid digital twin is a live network model synchronized with the physical grid through several data layers. The foundation is typically GIS, Esri ArcGIS is the most common platform, which holds the spatial topology of lines, substations, and assets. On top of that sits the ADMS (Advanced Distribution Management System): platforms like GE Vernova’s ADMS, Schneider Electric EcoStruxure, or Oracle Network Management System provide the real-time switching state, fault indicators, and load flow calculations. SCADA telemetry feeds ADMS with near-real-time measurements. AMI data from smart meters, Itron Riva or Landis+Gyr Gridstream, for example, adds per-endpoint load visibility. Asset records and maintenance history live in the ERP or enterprise asset management system. Pulling all of these into a unified model is the hard integration work that vendors rarely highlight. The result is not a 3D rendered visualization. It is a topological and parametric model that can run load flow, fault isolation, and restoration (FLISR) scenarios in software before operators commit to live switching. ## Where AI Genuinely Helps AI layers provide value in a few specific areas within this stack: Predictive failure scoring on assets such as transformers and underground cables uses sensor data, maintenance records, and weather inputs to rank assets by near-term failure probability. This feeds directly into maintenance scheduling in the ERP work-order module. Load forecasting and DER dispatch are also well-established. DERMS platforms like AutoGrid use ML models to balance distributed energy resources against grid capacity constraints. This is operational, not a pilot. Fault location in distribution networks, particularly underground feeders where acoustic and impedance-based methods struggle, is an area where AI pattern recognition on SCADA telemetry is showing measurable improvement over rule-based approaches. What AI does not do reliably yet: autonomous switching decisions in live networks. Human operators retain control through the ADMS console, and that is appropriate given the safety stakes. ## The Real Role of XR in Grid Management XR, AR and VR together, has a specific, bounded role here. It is not a real-time operational tool for grid management. Operators do not don headsets to run switching procedures. Where XR proves useful is in design review and training. A VR rendering of a planned substation allows engineers to walk through cable routing and clearance issues before construction. That is a legitimate productivity gain during capital project design. AR overlays on tablets or headsets during field inspection can surface asset records and last-maintenance dates from the ERP without the technician pulling out a separate device. Platforms connecting to SAP IS-U or S/4HANA Utilities asset modules can deliver this today; see the detailed treatment of AR on SAP platforms here. VR training simulations for substation switching procedures, energization sequences, and emergency response scenarios let new operators practice in a zero-risk environment before touching live equipment. This is one of the highest-value XR applications in utilities. Framing XR as a primary grid management interface, the “digital twin cockpit” described in many vendor presentations, overstates where the technology is. The smart grid optimization article covers the AI-driven operational side in more depth. ## Integration Complexity Is the Real Barrier The honest constraint on grid digital twin adoption is not the visualization layer. It is data integration. Getting GIS, ADMS, AMI, and ERP asset data into a consistent, synchronized model requires middleware, data governance, and ongoing reconciliation. Many utilities have GIS topology that diverges from the as-built field record. That gap undermines model accuracy before any AI or XR layer is applied. Utilities that have made the most progress typically started with a single substation or feeder as a pilot, resolved the data quality problems at that scope, and then expanded. Organizations considering this path should plan the data integration work as the primary workstream, not the visualization tools. ## Scope and Honest Assessment Grid digital twins built on GIS, ADMS, and asset data are practical today for planning simulations, predictive maintenance, and operator training. AI adds measurable value in fault prediction and load optimization. XR is useful in design review and training, not in day-to-day grid operations. The ERP and ADMS remain the systems of record. The “metaverse for grid operators” framing is marketing; useful work is happening, but it is more targeted than that framing implies. For a broader look at AI in utility operations, see AI for optimizing utility operations. ## Frequently asked questions What data sources feed a utility grid digital twin? Primarily GIS (such as Esri ArcGIS), SCADA telemetry, ADMS network models (GE Vernova, Schneider EcoStruxure, Oracle NMS), and asset records from the ERP. AMI data from meters (Itron, Landis+Gyr) adds load granularity. Does XR replace ADMS or the ERP in grid operations? No. ADMS and the ERP/CIS remain the systems of record. XR is a visualization layer, useful for design review and training, but operators still work from the ADMS console for real switching and outage management. Where does AI add genuine value in a grid digital twin? Fault location prediction, load forecasting, DER dispatch optimization (DERMS such as AutoGrid), and predictive maintenance scoring on specific assets like transformers and switchgear. Is 'metaverse for utilities' a real procurement category? Not yet at any scale. Most utility AR/VR budgets go to field inspection tooling and training simulation, not immersive metaverse environments. --- ## Emerging Utility Tech: AMI, DERMS, AI, and Cloud CIS URL: https://www.utilitieslabs.com/emerging-technologies-in-the-utilities-industry-what-to-watch/ # Emerging Utility Tech: AMI, DERMS, AI, and Cloud CIS AvanSaber Research/June 7, 2023/Updated September 24, 2026/3 min read Utilities are navigating a technology refresh cycle unlike any in recent memory. Four categories stand out as genuinely consequential right now: advanced metering infrastructure, distributed energy resource management, AI-driven analytics embedded in CIS and ERP platforms, and cloud-native customer information systems. Each is at a different maturity point, and the implications for your back-office systems differ substantially. ## Advanced Metering Infrastructure and the Data Flood Itron and Landis+Gyr remain the dominant AMI hardware vendors, but the real challenge is what happens after meters are installed. Head-end systems generate interval reads at 15- or 60-minute granularity across millions of endpoints. That data must flow into a meter data management (MDM) system, Oracle Utilities MDM or equivalent, before it reaches billing and customer care. Utilities that installed AMI a decade ago often struggle with the data pipeline. SAP IS-U’s meter reading workbench was not designed for interval data volumes at scale, and many shops rely on middleware or custom IDocs to bridge the gap. The organizations seeing real returns from AMI are those that have also modernized the MDM-to-CIS handoff. ## DERMS: Managing the Grid Edge Distributed energy resource management systems handle DER dispatch, forecasting, and grid coordination. GE Vernova and Schneider EcoStruxure Grid are active in this space. DERMS sits between the grid operator layer (SCADA/EMS) and the customer layer (CIS/billing) and must exchange data with both. For billing teams, the downstream effect is billing complexity: net metering, virtual net metering, time-of-use rates for EV charging, and demand response settlement all require CIS platforms capable of handling non-traditional rate structures. Oracle CC&B and SAP IS-U both support configurable rate engines, but configuration depth matters, a system set up for flat residential rates will need significant rework for TOU or demand-charge billing. ## AI and Machine Learning in Utility Operations AI is entering utility operations through two doors: predictive asset analytics and customer-facing automation. On the grid side, machine learning models applied to SCADA and sensor data can flag transformer degradation or cable fault risk before failures occur. On the customer side, platforms like Cayenta CIS incorporate tools such as Cayla AI for call deflection and self-service. The honest constraint is data quality. AI models are only as good as the underlying data, and utilities with fragmented meter data, manual billing adjustments, and inconsistent CIS records will find AI initiatives underperform expectations until foundational data hygiene work is done. This is less a technology problem than a process and governance one. ## Cloud CIS and the Platform Modernization Question SAP S/4HANA Utilities and Oracle Utilities Application Framework are both available in hosted or cloud configurations, but neither is a greenfield SaaS product. Meanwhile, cloud-native CIS vendors are gaining ground in the mid-market. Cayenta CIS and its broader Harris Computer ecosystem (ServiceLink mobile workforce, Silverblaze self-service portal, SmartWorks meter data management, all offered by Cayenta as add-ons) represent a full-stack alternative worth serious evaluation for utilities outside the large investor-owned tier. The decision framework for platform modernization is covered in detail in our guide to choosing the right utilities software. For a direct comparison of the two incumbent platforms, see Oracle vs. SAP for utilities implementations. ## What to Prioritize The watch-list hierarchy for most utilities: first, get the AMI-to-MDM-to-CIS data pipeline working reliably; second, address rate engine limitations before DER penetration forces a billing crisis; third, evaluate AI tools against realistic data quality baselines; and fourth, assess whether the incumbent CIS platform can carry the organization for another decade or whether a migration is unavoidable. Technology choices compound over time. A CIS selected today will shape operational capability through the 2030s. The organizations making the best decisions are approaching this as a systems architecture question rather than a vendor selection one. For deeper analysis of how these trends reshape long-range utility strategy, see our future of utilities predictions piece, which focuses on the structural business model shifts rather than specific technologies. If you want an independent assessment of your current platform stack against these trends, AvanSaber’s team is available for a consultation. ## Frequently asked questions What is the difference between AMI and smart meters? Advanced Metering Infrastructure (AMI) is the full two-way communications network, meters, head-end systems, and data management layer. Smart meters are the endpoint devices within that network. Do smaller utilities need DERMS? Any utility with meaningful rooftop solar penetration or battery storage programs will eventually need distributed energy resource management. The threshold is lower than most assume. Can cloud CIS replace SAP IS-U or Oracle CC&B? Cloud-native CIS platforms are now viable for mid-market utilities. Cayenta CIS from Harris Computer is one option worth evaluating alongside Oracle and SAP before committing to a major platform refresh. --- ## FLISR and Self-Healing Grids: How ADMS Uses AI to Restore Power URL: https://www.utilitieslabs.com/empowering-self-healing-grids-with-ai-for-autonomous-operations-in-utilities/ # FLISR and Self-Healing Grids: How ADMS Uses AI to Restore Power AvanSaber Research/October 11, 2024/Updated June 2, 2026/3 min read Self-healing grid is a widely used phrase, but the mechanics behind it are specific. The capability rests on a well-defined function called Fault Location, Isolation, and Service Restoration, commonly abbreviated FLISR, which is executed by an Advanced Distribution Management System (ADMS). AI is entering this space, but its role is advisory: the ADMS remains the authoritative system that opens and closes switching devices. ## What FLISR Actually Does When a fault occurs on a distribution feeder, a FLISR-enabled ADMS runs a three-step sequence. First, it uses fault indicator data and protective relay signals to locate the faulted segment. Second, it automatically opens the switching devices that bracket that segment, isolating it from the energized network. Third, it identifies alternate supply paths and recloses sectionalizing switches to restore power to the customers who were de-energized but are not on the faulted segment itself. On a well-configured feeder with automated switches, this entire sequence can complete in under two minutes without dispatcher involvement. Platforms such as GE Vernova GridOS, Schneider Electric EcoStruxure ADMS, and Oracle Utilities Network Management all implement FLISR as a core module rather than an add-on. ## Where AI Adds Value Without Replacing the ADMS The ADMS holds the network topology model, the switch states, and the protective device settings. AI models do not have direct write access to switching devices. Instead, they work with the data the ADMS exposes. Practical AI use cases that utilities are deploying today include: - Pre-fault pattern recognition. Models trained on AMI interval data from Itron or Landis+Gyr meters and distribution sensor streams can flag feeder sections showing voltage anomalies that precede insulation failures. The output is a maintenance work order recommendation, not an automated switch action. - Switching plan optimization. Before a planned outage, AI can evaluate hundreds of possible switching sequences against load forecasts and suggest the plan that minimizes customer-minutes interrupted. The ADMS dispatcher reviews and approves the plan before execution. - Post-fault load rebalancing. After FLISR has restored customers via alternate paths, AI can recommend further load transfers to reduce overloading on the alternate feeder. Again, a controlled workflow in the ADMS confirms the action. SAP BTP AI services and Oracle’s embedded analytics in Oracle Utilities Network Management are the two most common enterprise-grade integration points for these recommendation workflows. Neither bypasses the human-in-the-loop requirement for switching operations. ## Voltage and VAR Optimization Working Alongside FLISR Volt/VAR optimization (VVO) is a related ADMS function that continuously adjusts capacitor banks and voltage regulators to maintain voltage within acceptable limits. Like FLISR, VVO is an automated ADMS function. AI-based forecasting of distributed energy resource (DER) output, particularly solar generation variability, can improve the setpoints VVO uses, but the optimization loop itself runs inside the ADMS. DERMS platforms such as AutoGrid handle the DER side, coordinating battery storage dispatch or demand response, and feed aggregated flexibility signals to the ADMS. This is a co-ordination architecture, not a replacement of the ADMS control plane. ## Implementation Realities Achieving full FLISR automation requires adequate automated switch density on feeders, a reliable communication network to those switches, and an accurate network model in the ADMS. Utilities running aging infrastructure on manual switches get little value from AI-assisted FLISR recommendations until the underlying switching hardware is upgraded. For utilities exploring this path, a broader review of grid intelligence strategy belongs in the smart grid optimization and AI context, and the operational AI layer sits within the wider AI for utility operations discussion. Oracle Utilities practitioners can review the Oracle Utilities pillar for platform-specific guidance, and SAP-based utilities should look at SAP IS-U integration patterns. Avansaber’s advisory team works with utilities on ADMS selection, FLISR configuration, and the AI integration layer. Reach out via Avansaber Contact to discuss your distribution automation roadmap. ## Frequently asked questions What is FLISR in distribution automation? Fault Location, Isolation, and Service Restoration (FLISR) is an automated sequence built into ADMS platforms. When a fault is detected, the system identifies the faulted segment, opens switching devices to isolate it, then recloses alternate paths to restore power to unaffected customers, all within seconds to minutes. Does AI replace the ADMS in a self-healing grid? No. The ADMS is the system of record and executes switching logic. AI models running on platforms such as SAP BTP AI or alongside Oracle Utilities Network Management analyze patterns and recommend pre-approved switching plans, but a human operator confirms any action outside normal automated parameters. Which vendors offer ADMS with FLISR capabilities? GE Vernova GridOS, Schneider Electric EcoStruxure ADMS, and Oracle Utilities Network Management are the three most widely deployed platforms in North American and European distribution utilities. --- ## AI and XR in Smart City Utilities: Grounded Assessment URL: https://www.utilitieslabs.com/empowering-smart-cities-ai-and-xr-for-sustainable-utilities/ # AI and XR in Smart City Utilities: Grounded Assessment AvanSaber Research/August 31, 2024/Updated June 2, 2026/4 min read “Smart city” has been an active vendor category for over a decade. The honest verdict is mixed: some specific AI applications in utilities are genuinely mature, a few XR uses are proving their value, and a large share of the broader vision remains aspirational. This piece separates the working pieces from the hype. ## What Is Actually Working in Smart City Utility AI AI applications in utilities that are generating real operational outcomes today fall into a short list. AMI-driven analytics are the most broadly deployed. Itron and Landis+Gyr smart meter networks produce interval data that supports load forecasting, non-technical loss detection, and, in water utilities, pressure and consumption anomaly detection that helps identify distribution leaks. These are production systems, not pilots. DERMS platforms, with AutoGrid as a notable example, use ML-based forecasting to dispatch distributed energy resources, rooftop solar, batteries, demand response, against grid constraints. Several grid operators in North America and Europe are running these in production to defer distribution upgrades. ADMS platforms from GE Vernova, Schneider EcoStruxure, and Oracle NMS use AI-assisted FLISR (fault location, isolation, and service restoration) to reduce outage duration. The AI element here is primarily pattern matching on historical switching data to recommend restoration paths, with human operators approving the final actions. Customer-facing AI, chatbots and usage advisors, is rolling out through CIS platforms. Cayenta CIS (Harris Computer) has added AI tools through its Cayla AI module. SAP C4U includes customer engagement analytics tied to S/4HANA Utilities. Oracle CC&B integrations with Oracle CX layer in AI-driven service suggestions. These are incremental improvements to existing CIS workflows, not replacements for the billing system. ## Where XR Fits in a Smart City Utility XR in a smart city utility context is useful in three specific scenarios, and it does not belong in a fourth. Field inspection: AR overlays on mobile devices or headsets surface asset records, last-inspection dates, and wiring diagrams when a technician is standing at a piece of equipment. This reduces paper-based lookups and speeds up fault isolation. The data comes from the ERP asset module or GIS. Crew training: VR simulations for substation operations, gas pipeline emergency response, and distribution switching let new workers practice procedures safely. A 45-minute VR session simulating a substation energization sequence is far safer and cheaper than equivalent live training. See also VR for utility workforce training. Design and capital project review: VR walkthroughs of planned substations or pipeline reroutes let engineers and planners identify clearance and routing conflicts before breaking ground. This is a design tool, not an operations tool. The scenario where XR does not fit: routine customer billing, CIS operations, or SCADA monitoring. The systems of record for those functions are the CIS (SAP IS-U, Oracle CC&B, Cayenta) and the ADMS. An operator running switching procedures is working from an ADMS console, not a mixed-reality headset. ## The “Metaverse for Utilities” Problem Several consulting firms and vendors have published material on the “utility metaverse”, persistent virtual environments where operators manage grid assets, customer service runs in 3D, and field crews collaborate immersively. This framing is mostly marketing. The practical barriers are not primarily technical. They are organizational: utilities run on safety-critical processes with regulatory oversight. Changing the operator interface for grid switching or the intake workflow for a CIS requires regulatory approval, workforce retraining, and extensive testing. Utilities move carefully for good reasons. The realistic near-term outcome is incremental AR tooling for specific field tasks, not a wholesale shift to immersive operating environments. ## Data and System Integration Remains the Hard Problem Smart city utility initiatives that have stalled most often did so because of data integration problems, not technology limitations. Smart meters, GIS, ADMS, CIS, and ERP all hold different views of the same physical infrastructure and customer base. Getting those views synchronized well enough to support AI analytics requires data governance work that is unglamorous but essential. Utilities that have made genuine progress on smart city AI have typically invested heavily in their data platform first, MDM (master data management) for customer and asset records, standardized AMI data pipelines, and ERP data quality improvement, before layering analytics on top. For the AI and operations angle, see AI for utility operations and the smart grid optimization coverage. ## An Honest Summary Smart city utility AI is real in a bounded set of applications: AMI analytics, DERMS dispatch, ADMS-assisted outage management, and CIS-embedded customer tools. XR is useful for field inspection support, crew training, and design review. The broader “metaverse for utilities” vision is not supported by current procurement, regulatory, or organizational reality. The ERP and CIS stay at the center of utility operations regardless of what the smart city stack adds around the edges. ## Frequently asked questions Which smart-city utility AI applications are working today? Demand forecasting, AMI-based leak detection (water utilities), DER dispatch via DERMS, and outage management via ADMS are production-grade. Most others are in pilot or vendor-demo stage. What is XR's realistic role in a smart city utility context? Field inspection support via AR overlays, VR training for crews, and design review for infrastructure projects. XR is not replacing SCADA dashboards or CIS billing operations. Does smart-city tech replace the CIS or ERP? No. SAP IS-U, Oracle CC&B, Cayenta CIS, and similar platforms remain the customer and billing systems of record regardless of smart-city overlays. --- ## AR + ERP for Utility Asset Management and Work Orders URL: https://www.utilitieslabs.com/enhancing-asset-management-with-ar-erp-integration/ # AR + ERP for Utility Asset Management and Work Orders AvanSaber Research/May 16, 2024/Updated June 2, 2026/3 min read The case for AR in utility field operations is specific and practical: a technician standing at a transformer, switchgear cabinet, or meter installation benefits from seeing asset history, open work orders, and wiring diagrams without pulling out a laptop or radio-calling the back office. That is the genuine value proposition. It does not extend to billing, customer service, or SCADA operations. ## The Asset Data Problem AR Solves Field crews working from paper printouts or radio dispatch miss context that sits in the ERP. A maintenance technician might not know that an asset was flagged for follow-up during a prior inspection, or that a part was replaced under warranty. When AR is connected to the ERP asset module, SAP Plant Maintenance (PM) within IS-U or S/4HANA Utilities, or the EAM layer in Oracle Utilities, that history becomes available on-site, tied to the physical asset through QR codes, barcodes, or visual recognition. The workflow impact is concrete: fewer return trips for missed information, faster fault isolation, and better capture of maintenance data back into the ERP at job close. The ERP remains the system of record throughout. AR is the access layer, not the data store. ## Work Order Integration Is the Core Integration Point The most practical AR-ERP integration for utilities centers on work orders. In SAP IS-U, work orders flow through the PM module with connections to financial settlement in FI-CA. In Oracle CC&B or Cayenta CIS environments, field service work orders may live in a separate EAM system or within the CIS service order module. An AR application connected via API to the ERP work order module can: - Display the active work order number and task description when the technician scans an asset - Surface parts lists and torque/calibration specs from the asset master - Capture photo and notes attachments back to the work order record - Flag completion steps that trigger ERP status updates This is different from an AR application that operates independently of the ERP. The value comes from the integration, not from the AR rendering itself. ## Inspection and Remote Expert Use Cases Beyond work orders, two other field scenarios justify AR in utility asset management: Inspection workflows: Distribution line patrol, substation equipment inspection, and gas infrastructure visual inspection all benefit from AR-assisted checklists. An inspector can annotate what they see against a structured form that writes directly to the asset record. Drones with thermal or visual AR overlays are extending this to aerial inspection of transmission assets, though integration with ERP asset data at that level is still maturing. Remote expert assist: A field technician encountering an unfamiliar fault condition can connect via AR video to a specialist who sees the technician’s view and can annotate the physical environment with guidance markers. This reduces the need to dispatch a specialist to site for every non-routine situation. Platforms like PTC Vuforia, Scope AR, or TeamViewer Frontline support this pattern. For the SAP-specific implementation picture, how AR connects to IS-U and S/4HANA Utilities asset modules through IDocs and APIs, the AR integration with SAP article covers that in detail. ## Where AR Does Not Help To be direct: AR does not improve billing accuracy, CIS customer record management, meter reading processing, or regulatory reporting. Those functions run in IS-U FI-CA, Oracle CC&B, or Cayenta CIS and do not have a meaningful AR use case. AR also does not reduce the need for ERP data quality. If asset records in the ERP are incomplete or incorrect, AR surfaces incorrect information to the field. Data quality remediation is a prerequisite, not an afterthought. ## Realistic Deployment Scope Most current utility AR deployments use rugged tablets rather than headsets. Camera-based AR on a tablet with a protective case is more practical in outdoor utility environments, rain, gloves, dust, than a headset. Headsets like the RealWear Navigator are gaining traction for specific indoor substation scenarios where hands-free operation is a genuine advantage. A focused pilot on a single asset type, say, pad-mount transformer inspection, with ERP integration scoped to that asset class is a more tractable starting point than a broad field AR rollout. That also allows measurement of the efficiency gain before committing to larger hardware and integration investment. For the broader context on AI and field asset management, see AI for utility operations. ## Frequently asked questions What ERP asset data does an AR overlay typically surface in the field? Asset ID, installation date, last maintenance record, nameplate specs, and open work orders. In SAP IS-U or S/4HANA, this pulls from PM (Plant Maintenance) or the asset accounting module. In Oracle CC&B environments, it may come from the integrated EAM. Does AR replace the work order system? No. Work orders are created, dispatched, and closed in the ERP or field service module. AR surfaces the relevant work order data to the technician on-site; it does not generate or finalize the order. What are the realistic hardware options for field AR in utilities? Rugged tablets with camera-based AR are the most common deployment today. Headsets (HoloLens 2, RealWear Navigator) are used in specific inspection and remote-assist scenarios. Full headset deployment across a large field crew is still uncommon. Which utilities have deployed AR for field operations? Enel Group has publicized AR for field maintenance. Several North American distribution utilities have run pilots with RealWear devices for substation inspection. Broad production deployments remain limited compared to the pilot count. --- ## SAP Augmented Reality for Utilities: Field Service Guide (2026) URL: https://www.utilitieslabs.com/enhancing-erp-integrating-ar-with-sap/ # SAP Augmented Reality for Utilities: Field Service Guide (2026) AvanSaber Research/August 2, 2024/Updated September 27, 2026/14 min read “SAP augmented reality” returns mostly manufacturing and supply chain results. For utilities the useful question is narrower: where does AR on SAP IS-U or S/4HANA Utilities actually help a field crew, and what does the implementation require? The answer centers on field service, meter work, and asset inspection, not general enterprise digitization. Updated 9 August 2026: rewritten against the current product landscape. SAP has now published a deprecation notice for the Augmented Reality feature in SAP Mobile Services, with removal from the mobile services cockpit UI scheduled for Q4 2026, so the route table below marks what is still supported and what is not. The deeper treatment of that change is in what the Q4 2026 deprecation actually changes. ## Key takeaways - There is no single “SAP AR product”. There are four routes, and one of them now has a removal date: SAP states that “the Augmented Reality (AR) feature in SAP Mobile Services is deprecated and will be removed from the mobile services cockpit UI in Q4 2026”. - The route that carries most real deployments is SAP Service and Asset Manager, whose AR views and visual work instructions target first-time-fix rate, extended where needed by a partner package. - The durable integration contract is not any AR feature. It is the OData service layer over SAP IS-U or S/4HANA Utilities, which every route reads and writes through. - For utilities the value concentrates in four field tasks: substation inspection, meter work, first-time fix, and switching. AR does not touch billing, customer records, or regulatory reporting, and no honest business case should claim it does. - Asset data quality is a prerequisite, not an outcome. Where IS-U records are wrong, AR renders wrong data in three dimensions, faster. ## Is SAP discontinuing augmented reality? Two SAP mobile AR components are going away in 2026, and the integration approach behind utility field AR is not. SAP has deprecated the Augmented Reality feature in SAP Mobile Services, with removal from the mobile services cockpit UI scheduled for Q4 2026. Separately, it has deprecated the SAP Fiori for iOS AR components (the FioriAR package), marked deprecated in 2025 and scheduled for removal in 2026 because they rely on outdated third-party frameworks; that repository is now archived. Both are SAP-side building blocks, not the field-AR story as a whole. What carries real utility deployments is untouched by either removal: SAP Service and Asset Manager with its built-in AR views, partner extensions layered on top of it, and direct OData integration to SAP IS-U or S/4HANA Utilities. A pilot built on those three routes keeps working after Q4 2026. The route with the removal date is the platform-side configuration surface, not the field application and not the data contract. The full picture of what the Q4 2026 deprecation changes is covered separately. ## The four routes, and which are still supported Most confusion about SAP augmented reality comes from treating it as one product. It is four different things with different owners, different costs, and now different lifecycles. | Route | What it is | Who owns the integration | Status as of August 2026 | SAP Service and Asset Manager | The mobile field application, with built-in AR views and visual work instructions aimed at first-time fix | SAP, with your Basis and functional teams | Current, and the default starting point | Partner extension on top of SSAM | Packaged AR bought as an extension. TeamViewer Frontline provides xAssist remote-assistance calls and xInspect inspection workflows, triggered in context from the work order the technician is executing; ViewAR SERVICE AR is a comparable SAP-partner route | The vendor | Current, richest workflow coverage | Direct OData build | Your own AR client reading equipment, functional location, work order, and notification records straight from IS-U or S/4HANA Utilities | You | Current, and the route that ages best | AR feature in SAP Mobile Services | The platform-side AR configuration surface in the mobile services cockpit | SAP platform team | Deprecated. Removal from the cockpit UI in Q4 2026 Two practical notes. Captured field media does come back: notes and images taken during hands-free workflows are stored in the SAP backend and synced, so the documentation trail lands in S/4HANA rather than in a vendor silo. And the deprecation only touches the fourth row. If your pilot was built on the first three, nothing about it changes. “On a utility rollout the AR headset is the easy part. The hard part is whether the asset record in IS-U is clean enough to overlay. Where the SAP data is good, augmented reality earns its place on substation inspection and meter work. Where it is not, AR just renders bad data in three dimensions.” Nikhil Jathar, founder of AvanSaber and former SAP consultant ## What SAP augmented reality means in a utility Augmented reality in this context is a presentation and capture layer sitting over SAP data. The system of record stays SAP IS-U or S/4HANA Utilities. An AR application reads asset master data, work orders, and service notifications that already exist in SAP, then surfaces them to a worker standing in front of the equipment. The worker sees relevant data overlaid on the physical environment and can capture field input back into SAP without leaving the work site. Three routes exist to build this layer on SAP for utilities: SAP Asset Manager, SAP’s standard mobile field service application, includes built-in AR capabilities. It connects natively to IS-U and S/4HANA Utilities work orders and asset records and is the lowest-integration-effort path for a utility that already uses SAP Asset Manager for field service dispatch. Third-party AR platforms, PTC Vuforia, Scope AR, and TeamViewer Frontline being the most deployed in utility environments, connect to SAP through OData services exposed by IS-U or S/4HANA Utilities. They add capabilities like remote expert video overlay and industrial AR annotation that the SAP-native option does not cover natively. Hands-free hardware, specifically RealWear Navigator headsets, is a separate hardware layer that works with both the SAP-native and third-party software routes. It matters most for tasks where a technician needs both hands free: substation switching, meter installation in confined spaces, and lineman operations at height. ## Four utility field uses that are operational The utility-specific value of SAP AR concentrates in four areas. Each maps directly to data the utility already holds in IS-U or S/4HANA Utilities. Asset and substation inspection. Overlay the asset record, last-inspection data, and the inspection checklist on the equipment in view, then capture results directly to the SAP work order. For substation equipment with complex inspection procedures and regulatory documentation requirements, the structured capture to the asset record is often the primary value, more concrete than the AR visualization itself. Meter reading and meter work. Guide a technician to the correct meter, display the installation and configuration history from IS-U, and reduce mis-reads and mis-installs. For smart meter deployment programs and meter exchange work, AR-guided installation reduces repeat visits caused by installation errors on the first visit. Field service and first-time fix. Surface the open work order, asset history, and next correct step so a crew completes more jobs on the first visit. Reducing unnecessary return trips is one of the largest controllable costs in utility field operations, and first-time-fix rate is where AR-assisted field service shows the clearest measurable impact. Lineman safety and switching. Present switching steps and hazard data in the field crew’s line of sight for high-risk tasks, with each confirmation captured back to SAP. For complex switching sequences that require documented step-by-step execution, AR-guided switching reduces procedure errors and creates an automatic audit trail in the asset record. Each of these ties to data a utility already holds in SAP IS-U or S/4HANA Utilities. That is what makes them operational rather than aspirational. For the detailed picture of how AR connects to ERP asset modules, work orders, and inspection data flows, AR and ERP for utility asset management covers the integration specifics. ## Augmented reality for utility field service Utility field service is where SAP AR earns its keep, and the framing that matters is utility-specific: field crews, meter technicians, and substation teams working assets that already live in SAP IS-U or S/4HANA Utilities. This is not generic field service management software. It is an AR presentation layer over the utility’s own asset and work-management data, delivered through SAP Service and Asset Manager and the OData services that expose IS-U to the field. The dispatch and work-order backbone stays in SAP. A crew receives the assigned work order, and the AR layer surfaces the asset record, service history, and the next correct step against the equipment in front of them. For a meter technician that means the installation and configuration history from IS-U at the meter, not back at the truck. For a substation team it means the inspection checklist and isolation data tied to the specific functional location. The measure that moves is first-time fix: fewer return trips because the crew had the right asset data, captured to the work order, on the first visit. Two constraints keep this honest for a utility. The value shows up on defined, repeatable tasks tied to clean IS-U records, not across the whole field operation at once. And connectivity is worst exactly where the assets are, so offline capture with deferred sync back to SAP is a requirement, not an extra. For the work-order and asset-management mechanics underneath this, AR and ERP for utility asset management has the detail. ## Integration architecture The integration pattern is consistent across SAP-native and third-party approaches. The AR application calls OData services exposed by SAP IS-U or S/4HANA Utilities to read: - Asset master records: installation dates, nameplate data, maintenance history, equipment ratings - Work orders with task lists, parts requirements, and step-by-step completion sequences - Service notifications and structured inspection checklists - Safety-relevant records including isolation procedures and switching step sequences On job completion, the AR application writes captured data back through the same OData interfaces: inspection results, photo attachments, completion confirmations, and exception flags. The asset record and work order in IS-U reflect the field results without manual back-office data entry. The integration scope should be bounded. A pilot that connects to one workflow object type, such as the inspection work order, is a manageable integration project. Expanding to cover the full breadth of IS-U objects in the first deployment adds complexity without proportional benefit and makes measuring success harder. Treat the AR tool as a field client. It reads from and writes to SAP. It does not replace SAP as the system of record and should not be designed to operate independently of the IS-U data model. ## A substation inspection in SAP AR, step by step Here is what the integration does on one concrete task, described as capability rather than a specific deployment. The point is which SAP IS-U objects the AR layer reads and writes, so the scope of a pilot is clear before anyone buys hardware. - The crew opens the assigned inspection work order in SAP Service and Asset Manager. The AR layer resolves the substation asset from its functional location and equipment records in IS-U. - On the equipment in view, it overlays the asset master data (nameplate, ratings, installation date), the last inspection result, and the structured inspection checklist attached to the work order. - The technician works the checklist hands-free. Each reading, photo, and exception is captured against the work order and the equipment record. - Measurement readings write back as measurement documents on the equipment. Photos attach to the notification or work order. Completion and any exception flags post to SAP. - The asset record and work order in IS-U reflect the field result with no back-office re-keying, which is the documentation trail a regulator expects. Every object named here already exists in a utility’s IS-U or S/4HANA Utilities data model. That is what keeps the walkthrough grounded: the AR layer surfaces and captures records that are already there, and a pilot that connects only the inspection work order and its equipment is a bounded, measurable integration rather than an open-ended one. ## Manufacturing AR versus utility field AR The public SAP AR material, and most of what ranks for the term, is about the factory floor: assembly guidance, warehouse picking, and machine repair against product and production data. That is a real use, and it is not the utility use. A utility is not assembling a product on a line. It is inspecting a substation, exchanging a meter, and switching a feeder, against asset and work-management data in IS-U rather than production data. The distinction matters for a buyer. Manufacturing AR value cases lean on throughput and assembly-error reduction. The utility case leans on first-time fix, inspection data quality in the asset record, and a defensible audit trail for safety-critical work. Reading a manufacturing AR business case and assuming it transfers is the fastest way to scope a utility pilot wrong. ## What a first pilot should look like Utilities that run successful first AR pilots consistently apply the same scoping rule: one workflow, one asset type, one crew, and a defined success metric. Substation inspection is the most common starting point because it satisfies each condition. The inspection checklist maps to a defined asset record structure in IS-U. Hands-free operation in the substation environment has immediate practical value. The success metric, inspection completion time and data completeness in the asset record, is objective and measurable within weeks of deployment. Define the success metric before deployment, not after. For first-time-fix rate improvement, set a baseline from the prior period’s work order data before the pilot begins. For inspection data quality, establish what a complete record looks like in IS-U before any AR capture begins. Pilots without a pre-defined metric rarely generate the evidence needed to justify a broader rollout. Hardware choice is a pilot decision, not a pre-decision. Most pilots begin with rugged tablets using camera-based AR because they are practical outdoors, work with gloves, and survive utility field environments. Headsets add clear value in specific indoor scenarios, substation switchgear rooms, meter vaults, and underground distribution chambers, where hands-free operation is a genuine operational requirement rather than a convenience. | Hardware | Where it fits in utility work | What to watch | Rugged tablet, camera AR | Default for outdoor field work: meter exchange, overhead inspection, general work orders | Glove operation and screen legibility in direct sun decide adoption more than AR quality does | Hands-free headset (RealWear Navigator class) | Switchgear rooms, meter vaults, underground chambers, any task needing both hands | Safety approval runs on its own calendar and sits outside the software project | Smart glasses with remote assist | Remote expert calls on rare equipment, where the value is the person on the other end | Connectivity, which is worst exactly where the equipment is most obscure The connectivity point deserves more weight than it usually gets in a business case. A workflow that assumes a live session fails in the substation basement. Offline capture with deferred sync is a requirement, not a nice-to-have, and it should be tested at the worst site in the territory rather than the one nearest the office. ## What it costs, honestly Nobody publishes real numbers for this, and any vendor total cost of ownership slide should be treated as marketing rather than evidence. What can be said without inventing figures is where the money actually goes, and it is not where first-time buyers expect. The AR software licence is rarely the largest line. The larger ones are the integration work to expose and map the right OData services, the asset-data remediation the pilot exposes, the device fleet and its safety approval, and the training and process redesign that has to survive crew turnover. A programme that budgets for licences and headsets and not for the other four is the standard way this fails quietly. The corollary is that the first pilot should be scoped to make those costs visible early, which is another argument for one workflow, one asset type, one crew. ## Where SAP AR does not pay Being specific about the failure cases is more useful than another list of benefits. - Where the asset data is known to be poor. AR surfaces the record faster; it does not correct it. Fix the record first or the pilot measures your data problem instead of your AR question. - Where a tablet showing the same record performs identically. If nothing about the task requires hands free or spatial context, AR is an expensive way to display a work order. - Training scenarios a good video covers. The AR training case is real in a few high-consequence procedures and oversold everywhere else. - Anything in billing to cash. Billing, customer records, and regulatory reporting run in IS-U FI-CA and the billing engine, and have no meaningful AR use case at all. - Programmes without a measured task. If nobody can name the metric before deployment, the rollout has no way to prove itself and will be cut in the next budget round regardless of how well it works. ## Where to draw the line Generic enterprise AR and VR for ERP, broader innovation strategy, and non-utility applications sit with AvanSaber. UtilitiesLabs keeps the focus on utility field work tied to SAP IS-U and S/4HANA Utilities data, because that is where the use cases are concrete and the data already exists. AR does not improve billing accuracy, customer record management, or regulatory reporting. Those functions run in IS-U FI-CA or the billing engine and have no meaningful AR use case. A utility should not expect AR to touch the billing-to-cash process. Data quality is a prerequisite, not an outcome. If asset records in IS-U are incomplete or incorrect, the AR layer surfaces incorrect information to the field. A data quality assessment of the relevant asset class should precede the pilot, not follow it. ## The verdict SAP augmented reality earns its place in utilities on specific, repeatable field tasks: substation inspection, meter work, first-time-fix support, and safety-critical switching. The technology works when it is narrowly scoped, tied directly to IS-U or S/4HANA Utilities asset data, and measured against a defined outcome. Pilot on one workflow. Keep SAP IS-U or S/4HANA Utilities as the system of record. Scale only where the field data shows measurable improvement in first-time-fix rate, inspection data completeness, or switching error reduction. For the full SAP IS-U module map and how field service and asset management fit into the broader SAP utilities architecture, start with the SAP IS-U platform overview. ## Frequently asked questions What is SAP augmented reality used for in utilities? The practical uses are field service and safety: overlaying work-order and asset data for crews on site, guiding meter and substation inspection, and supporting lineman tasks during switching. The data behind the overlay comes from SAP IS-U and S/4HANA Utilities asset and work-management records. It is a presentation and capture layer, not a replacement for the system of record. Does augmented reality integrate with SAP IS-U? Yes. AR front-ends consume asset master data, work orders, and service notifications from SAP IS-U or S/4HANA Utilities through OData services or standard APIs, then present that data hands-free to a field worker. SAP Asset Manager, SAP's standard field service application, includes built-in AR capabilities for this purpose. Third-party AR platforms connect through the same OData interfaces. What hardware is used for SAP AR in utility field work? Rugged tablets with camera-based AR are the most common deployment. RealWear Navigator headsets are used in substation inspection and indoor maintenance where hands-free operation matters. Full headset deployment across a large field crew is uncommon; most utilities start with tablets before evaluating headsets for specific workflows. Is SAP augmented reality worth it for a utility? It pays off on specific, repeatable tasks: substation inspection, meter work, complex switching, and first-time-fix support, where surfacing the right asset data reduces errors and unnecessary return trips. It is not a general-purpose field technology. The ROI case is strongest when the utility already has clean asset data in IS-U and a clearly scoped pilot workflow. Is SAP augmented reality being discontinued? No, but two SAP building blocks are going away. SAP's documentation states that the Augmented Reality feature in SAP Mobile Services is deprecated and will be removed from the mobile services cockpit UI in Q4 2026, and advises customers to plan for alternative solutions before that date. Separately, the SAP Fiori for iOS AR components in the SAP BTP SDK were marked deprecated in 2025 and are scheduled for removal in 2026. Both are building blocks, not the field application. SAP Service and Asset Manager, partner AR extensions, and direct OData integrations to IS-U or S/4HANA Utilities are unaffected. Which SAP product actually provides augmented reality? SAP Service and Asset Manager is the field application, and it carries AR views and visual work instructions aimed at raising first-time-fix rates. Beyond that, most production deployments add a partner extension such as TeamViewer Frontline or ViewAR SERVICE AR, which trigger AR workflows in context from the work order and return captured media into the SAP backend. There is no standalone product called SAP AR. How should a utility pilot SAP AR? Scope the pilot to one workflow, one asset type, and one crew before any broad rollout. Substation inspection is a common starting point because the inspection checklist maps cleanly to the asset record in IS-U, the value of hands-free operation is immediate, and the success metric (inspection completion time and data quality) is objective and measurable within the first weeks. --- ## VR for Utility Workforce Training: What Works URL: https://www.utilitieslabs.com/enhancing-erp-training-with-vr/ # VR for Utility Workforce Training: What Works AvanSaber Research/March 15, 2024/Updated September 24, 2026/3 min read VR training for utility field crews is one of the better-substantiated AR/XR applications in the industry. The safety stakes of utility work, high-voltage switching, gas pressure systems, confined spaces, make the case for consequence-free simulation practice straightforward. This piece covers what the technology delivers today, which scenarios it fits, and how it connects to the rest of the utility technology stack. ## Why Utility Field Training Is a Good VR Fit Most enterprise VR training enthusiasm is overstated. But utilities are genuinely one of the better fits for the technology, for a simple reason: the cost of a mistake during live training is high. Miswiring a substation during an energization procedure, mishandling a gas main during a pressure test, or entering a confined space without correct atmospheric checks, these carry injury and equipment damage risk that makes simulation practice genuinely valuable, not just convenient. A 30-to-45-minute VR simulation of a switching procedure lets a new substation operator practice the sequence, make errors, and receive corrective feedback without any live equipment involved. The same operator then approaches the real energization sequence with procedural familiarity that classroom instruction does not build as effectively. This is a bounded claim: VR builds procedural familiarity and reduces hesitation. It does not replace supervised live practice, and it does not replace the formal qualification processes utilities maintain for safety-critical roles. ## Scenarios with a Real Track Record Substation switching and energization: VR simulations of switchgear operation, bus configurations, and energization sequences are the most mature utility VR training application. Several North American and European utilities have deployed these. The scenario fidelity matters, the VR model should reflect the actual equipment types the operator will encounter, not a generic switchyard. Gas distribution emergency response: Simulating a reported gas odor response, excavation near a main, and pressure isolation procedures are scenarios where VR practice reduces the variance in crew response. The procedural steps are fixed; VR is an efficient way to build and reinforce them. Arc flash and electrical safety awareness: VR scenarios that demonstrate the physics of an arc flash event, which cannot be safely demonstrated in live training, give workers visceral context for why PPE and clearance procedures exist. This is a knowledge-building application more than procedural training. Control room operations: VR and desktop simulation of SCADA and ADMS interfaces allows control room operators to practice outage response scenarios without touching the live system. This is particularly useful for training on low-frequency, high-consequence events like large storm outages. ## What VR Training Does Not Do VR does not train the customer-facing or administrative parts of utility work. Billing adjustments in SAP IS-U FI-CA, service order processing in Oracle CC&B, or customer account management in Cayenta CIS are software workflows trained through ERP/CIS walkthroughs and sandbox environments, not VR. The VR for ERP system simulations article covers that specific angle. VR also does not substitute for regulatory qualification requirements. A lineman still needs supervised live-line training and a formal sign-off process that regulators and unions require. VR accelerates the path to that sign-off; it does not replace it. ## Integration with HR and Compliance Records The operational connection between VR training and the broader HR/ERP stack matters for compliance tracking. Most utilities are required to document training completions for safety-critical roles. VR platforms need to generate completion records that flow to the LMS (Learning Management System) or directly to SAP HCM, SuccessFactors, or Oracle HCM. Without that integration, training completion data lives in a silo and creates audit risk. The integration is typically straightforward, most modern VR training platforms support xAPI or SCORM connectors to standard LMS systems, but it requires configuration work during deployment. Plan for this from the start rather than treating it as an afterthought. ## Hardware and Deployment Practicalities Standalone headsets (Meta Quest 3 or equivalent) cover most field training simulation scenarios. They are affordable enough to deploy multiple units per training center without a large capital commitment. For high-fidelity control room simulations, tethered PC-based systems with larger display environments are more appropriate. Outdoor field deployment of VR headsets for on-site practice remains uncommon and impractical. The value is in training centers and classroom settings before crews go to the field, not in the field itself. For the AR field support side, how AR overlays assist crews during actual field work, see the SAP augmented reality hub and AR for utility asset management. ## Frequently asked questions What utility training scenarios are best suited to VR? Substation switching procedures, energization sequences, gas leak emergency response, confined space entry, arc flash awareness, and new-hire orientation to complex infrastructure. Scenarios where a mistake in live training has safety or equipment consequences. Does VR training replace the ERP or CIS training? They address different training needs. ERP/CIS training (SAP IS-U, Oracle CC&B, Cayenta) is typically software walkthrough and role-play. VR training is for physical field and operations procedures. Both are needed; neither replaces the other. What hardware is typically used for utility VR training? Standalone headsets (Meta Quest 3, HTC Vive XR Elite) for most simulation scenarios. Tethered PC-based systems for high-fidelity control room simulations. Rugged headsets for field-adjacent use. How does VR training connect to ERP records? Training completions can be logged to SAP HCM or SuccessFactors, Oracle HCM, or a standalone LMS. The VR platform triggers a completion record; the HR/ERP system holds the compliance record. --- ## 5G in the Utilities Sector: Real Uses, Real Limits URL: https://www.utilitieslabs.com/exploring-the-potential-of-5g-in-the-utilities-sector/ # 5G in the Utilities Sector: Real Uses, Real Limits AvanSaber Research/June 19, 2023/Updated June 2, 2026/3 min read Private 5G networks are moving from pilot projects to cautious production deployments at a small but growing number of electric and gas utilities. The technology is genuinely useful in specific scenarios, but the industry-wide transformation narrative oversells what 5G actually delivers today. ## Where 5G Fits in the Utility Technology Stack The utility communications landscape already has several layers: licensed-band RF mesh (Itron OpenWay Riva, Landis+Gyr Gridstream), cellular LTE backhaul, fiber on the transmission side, and older SCADA radio. 5G does not replace these; it adds capacity and low latency in situations where existing options fall short. The three most concrete use cases are: Private campus networks at substations. A 5G private network (operating on licensed or CBRS spectrum) can connect protection relays, reclosers, and distributed energy resource controllers with sub-10 ms latency. GE Vernova and Schneider EcoStruxure ADMS platforms can absorb telemetry at this speed for fault detection and isolation. Dense-urban AMI backhaul. In cities where building penetration defeats RF mesh, 5G small cells on utility poles carry meter reads from Itron or Landis+Gyr devices to the AMI head-end. That head-end then delivers interval data to an Oracle MDM or SAP MDG-U for validation and settlement before the CIS processes it for billing. Field crew connectivity. Technicians running GIS-connected work management applications on tablets benefit from high-bandwidth 5G where Wi-Fi is unavailable. Esri ArcGIS field apps and SAP Work Manager both require reliable data sync; 5G reduces the offline-mode workarounds that slow field operations. ## Integration With CIS and OMS 5G is a transport technology, not an application. Data flowing over a 5G network still must land somewhere meaningful. For electric utilities, that means reaching the OMS (outage management system) for switching orders, the ADMS for volt-VAR optimization, and ultimately the CIS for customer notification and billing. SAP IS-U and Oracle CC&B both consume interval reads from the MDM via batch or near-real-time feeds; the communication path to the meter is abstracted away from those systems entirely. Utilities running Oracle CC&B and those on SAP IS-U need to evaluate 5G at the AMI head-end integration layer, not at the CIS level. The CIS does not care whether the meter read arrived over cellular, RF mesh, or 5G. ## Honest Assessment of Deployment Barriers The cost structure is demanding. Private 5G requires spectrum licensing or CBRS coordination, purpose-built radios, and a 5G core. For a distribution utility covering thousands of square miles, building coverage density that rivals an existing RF mesh network is expensive and slow. Cybersecurity obligations compound the challenge. Any device connecting to the operational network must satisfy NERC CIP controls (for electric utilities in North America). A 5G endpoint that touches protection or metering infrastructure is subject to the same access management, patch management, and incident reporting requirements as any other grid device. The radio access network adds attack surface that must be accounted for in the CIP compliance program. A further constraint: regulatory data residency requirements in some jurisdictions restrict where meter data can be processed. Cloud-hosted 5G core deployments that route data through shared infrastructure may require additional legal review. ## What to Watch Vendors including Ericsson and Nokia are partnering with utilities on private network builds. ERCOT in Texas, as the largest deregulated market in North America, represents a proving ground where settlement accuracy at scale requires reliable interval data delivery. Watch for 5G to expand in high-density meter areas in deregulated markets first. For utilities evaluating the technology, start with a narrow scope: one substation or one urban meter cluster. Prove out the integration with the existing AMI head-end and MDM before making grid-wide commitments. The IoT and sensor data path into MDM and CIS is the actual integration problem to solve, and 5G is one option among several for the transport segment. See AvanSaber’s contact page for help scoping a realistic 5G pilot for your system architecture. ## Frequently asked questions What is the main use case for 5G in electric utilities? Private 5G campus networks for substations and distribution automation, plus high-bandwidth AMI backhaul where legacy RF mesh is congested. Does 5G replace existing AMI communication networks? Rarely a full replacement. Most deployments layer 5G alongside existing Itron or Landis+Gyr RF mesh to fill coverage gaps or handle high-density urban meter clusters. How does 5G data reach the CIS or MDM? 5G carries meter reads and sensor telemetry to the AMI head-end, which then feeds an Oracle MDM or equivalent system before posting consumption data to the CIS for billing. --- ## Blockchain in Utilities: Peer-to-Peer Trading and REC Tracking URL: https://www.utilitieslabs.com/how-blockchain-technology-is-transforming-the-utilities-industry/ # Blockchain in Utilities: Peer-to-Peer Trading and REC Tracking AvanSaber Research/June 8, 2023/Updated June 2, 2026/3 min read Blockchain attracts a lot of attention in utility industry publications. The reality is more modest: a small number of proof-of-concept platforms exist, a handful have progressed to limited pilots, and none has replaced a mainstream energy market settlement mechanism. That does not mean the technology is irrelevant, but a clear-eyed view is useful before committing budget. ## What Blockchain Actually Does in the Energy Context A blockchain is a distributed append-only ledger where participants agree on the state of records without a central authority. For energy, that property is most useful when multiple unaffiliated parties need to agree on who generated what, when, and who bought it. Two use cases fit that description. Renewable Energy Certificate tracking. RECs (also called Guarantees of Origin in Europe) represent proof that one megawatt-hour of electricity was generated from a renewable source. The existing registry systems (PJM GATS, WREGIS in the western US) are centralized. A blockchain ledger could reduce the administrative overhead of cross-registry transfers and lower the risk of double-counting. This is a real problem worth solving, and it is the blockchain application in utilities with the most plausible near-term value. Peer-to-peer energy trading. Power Ledger, an Australian company, built a platform allowing solar prosumers to sell excess generation directly to neighbors on the same distribution feeder. WePower used a token-based model to let renewable project developers pre-sell energy output. Both ran documented pilots in Australia, Estonia, and other markets. Neither has achieved the transaction scale or regulatory acceptance needed to operate as a primary settlement mechanism. ## Why Pilots Have Not Scaled Several structural barriers explain the gap between pilot success and production adoption. Settlement in a regulated market is a legal construct, not just a technical one. ERCOT, as the operator of the Texas deregulated market, runs financial settlement through a centralized clearing function with defined latency and dispute resolution. A blockchain-based settlement layer would require ERCOT rule changes, FERC acceptance in interstate markets, and state PUC approval in others. That is a multi-year regulatory process. The CIS sits downstream of settlement. SAP IS-U and Oracle CC&B receive posted consumption and revenue data after settlement is complete. Even if a blockchain layer handled P2P trades, the CIS would need a new upstream integration to consume those results for billing. That integration does not currently exist in any of the major platforms. Transaction throughput is a second barrier. Public blockchains (Ethereum, for example) process a few dozen transactions per second. A large utility with millions of meters generates interval reads every 15 minutes around the clock. High-throughput private chains solve the speed problem but reintroduce centralization, reducing the decentralization argument. ## Where Blockchain May Still Have a Role REC tracking is the most defensible application. The deregulated market context creates genuine demand for transparent, cross-party attribution of renewable generation. Corporate sustainability programs are driving REC procurement at scale, and auditors want better provenance records than current registry systems provide. Smart contracts (self-executing code on a blockchain) could also automate demand response payments between a utility and a fleet of distributed assets. This is closer to production than P2P trading, because the number of counterparties is smaller and the regulatory path is more navigable. ## An Honest Verdict Blockchain is not transforming utility billing or CIS operations in any meaningful current sense. The transformation narrative gets ahead of what the technology can actually deliver within the regulated utility construct. The strongest near-term opportunity is REC attribution for corporate renewable buyers and utilities seeking to substantiate sustainability claims. For utilities evaluating the space, keep blockchain experiments narrow, scoped to the settlement or attribution layer, and separate from core CIS/ERP systems. Choosing the right utilities software for billing and customer management still rests on proven platforms, not distributed ledgers. If you want to understand where blockchain pilots might fit in your operations, AvanSaber can help frame the right scoping question. ## Frequently asked questions What is the most credible blockchain use case in utilities? Renewable Energy Certificate (REC) tracking, where an immutable ledger reduces double-counting risk. P2P energy trading platforms exist but remain small-scale pilots. Have any blockchain energy trading platforms scaled beyond pilots? Power Ledger and WePower ran documented pilots in Australia and Europe. Neither has reached the transaction volume of a major bilateral energy market like ERCOT. Does blockchain integrate with SAP IS-U or Oracle CC&B billing? No current production integration exists. P2P settlement would require a new settlement layer that sits upstream of the CIS, not inside it. --- ## Digital Twins in Utilities: What's Real vs. What's Still Hype URL: https://www.utilitieslabs.com/how-digital-twins-are-revolutionizing-the-utilities-industry/ # Digital Twins in Utilities: What's Real vs. What's Still Hype AvanSaber Research/June 24, 2023/Updated June 2, 2026/3 min read The term “digital twin” covers a wide range of maturity levels, from a simple CAD model linked to a spreadsheet up to a real-time, physics-based simulation of an entire distribution network. For utility professionals evaluating the technology, the gap between vendor marketing and production-ready capability is significant. This article separates the well-established use cases from the ones still being proven. ## What a Utility Digital Twin Actually Requires A digital twin is only as useful as the data it consumes. For a grid or network twin, three data layers must be accurate and current: - Network topology. Esri ArcGIS is the dominant GIS platform in utilities, and it is typically the authoritative source for feeder configurations, transformer locations, and asset attributes. If the GIS model is stale, every simulation built on top of it is unreliable. - Real-time operational state. SCADA feeds and ADMS telemetry from platforms such as GE Vernova GridOS or Oracle Utilities Network Management provide switch positions, load readings, and voltage measurements that bring the static topology to life. - Asset health data. Age, maintenance history, and sensor readings (oil temperature, partial discharge, vibration) from the ERP or a dedicated asset performance management system give the twin its predictive dimension. Many utilities have all three layers in separate systems with poor synchronization. Closing that gap is the unglamorous prerequisite work before any twin delivers value. ## Where Digital Twins Deliver Measurable Value Today Substation and transformer modeling is the most mature use case. Equipment vendors and independent platforms can build thermal and electrical models of transformers with enough fidelity to predict winding hot-spot temperatures under load, estimate remaining insulation life, and trigger maintenance before a catastrophic failure. This requires continuous monitoring, not just periodic testing. Distribution network simulation supports planning work: evaluating the impact of a new industrial customer, sizing a capacitor bank, or modeling the voltage effects of rooftop solar penetration on a feeder. The ADMS network model is the foundation, but planners typically run simulations in a separate tool that can test scenarios without touching the live system. Field workforce support is an emerging use case. Technicians with access to a spatially accurate GIS-linked twin can visualize underground cable routes, locate buried assets, and see the last-known condition of a structure before arriving on site. Esri’s Utility Network model is purpose-built for this. ## What Is Still Hype or Early Stage Real-time city-scale network twins exist in pilot form at a handful of utilities, but operating one reliably at scale requires data integration maturity that most organizations have not yet achieved. The sensor refresh rates, data quality controls, and model update pipelines needed are substantial infrastructure investments. AI-driven autonomous decision-making inside the twin is not production reality. AI models can run scenario analysis on twin data and surface recommendations, but the twin does not issue switching commands. Control remains in the ADMS, with human oversight on any action outside automated parameters. This is the right architecture: the CIS or ADMS is the system of record; the twin is an analytical layer alongside it. Complete system-of-systems twins that integrate grid, water, and gas networks in a single federated model are being researched but are not deployed at scale. ## The GIS-ADMS Integration Gap Is the Real Bottleneck For most utilities, the limiting factor is not a lack of twin technology. It is that the GIS model diverges from the ADMS network model over time as construction records lag field changes. Any digital twin strategy has to address this data governance problem first. Practitioners building out this capability should review smart grid optimization approaches for the grid-side context, and AI for utility operations for the analytics layer. The Oracle Utilities and SAP IS-U pillars cover how the ERP and CIS integrate with grid data systems. For a vendor-neutral assessment of where digital twin investment makes sense for your utility, Avansaber’s advisory team can help scope the data integration prerequisites alongside the twin platform selection. ## Frequently asked questions What data feeds a utility grid digital twin? A grid twin typically ingests the network topology from Esri ArcGIS (the GIS of record), real-time SCADA and ADMS telemetry, AMI interval data, and asset health records from the ERP or asset management system. Without an accurate, maintained GIS model, the twin is unreliable. Is a digital twin the same as the ADMS network model? They overlap but are not identical. The ADMS network model is optimized for real-time switching and load flow calculations. A digital twin is broader: it may incorporate thermal aging models, historical outage records, civil asset data, and simulation scenarios that the ADMS does not maintain. Which asset classes benefit most from digital twins today? Large rotating equipment (turbines, generators), high-voltage transformers, and underground cable networks see the most mature use cases because their failure modes are well-modeled and the sensor data needed to feed the twin is already installed. --- ## IoT and AMI Sensors: The Data Path Into MDM, CIS, and ADMS URL: https://www.utilitieslabs.com/how-iot-is-revolutionizing-the-utilities-industry/ # IoT and AMI Sensors: The Data Path Into MDM, CIS, and ADMS AvanSaber Research/June 7, 2023/Updated June 2, 2026/3 min read The phrase “IoT in utilities” covers two distinct categories of device that have very different data paths and different destinations in the software stack. Conflating them produces a blurry picture. This article separates them and follows each data flow to its destination. ## Category One: Advanced Metering Infrastructure AMI devices, primarily electric meters from Itron and Landis+Gyr, are the largest-scale IoT deployment in the utility industry. A US investor-owned utility serving a few hundred thousand customers may have millions of interval meter reads arriving every 15 minutes. That volume cannot land directly in a CIS. The data path works like this: - The meter reads consumption and stores interval data locally. - RF mesh (Itron OpenWay, Landis+Gyr Gridstream) or cellular backhaul carries reads to the AMI head-end system. - The head-end forwards interval data to the Meter Data Management system. Oracle MDM (part of Oracle Utilities) and SAP MDG-U are the dominant platforms at large utilities; some Cayenta CIS sites use lighter MDM connectors within the Harris Computer ecosystem. - The MDM performs Validation, Estimation, and Editing (VEE): checking for missing intervals, flagging meter tamper events, and filling short gaps with approved estimation algorithms. - Clean, validated consumption data posts to the CIS, where SAP IS-U or Oracle CC&B picks it up to generate bills. The MDM is the most often underestimated component in AMI rollouts. Projects that go straight from head-end to CIS without a proper MDM layer frequently struggle with billing accuracy. See our Oracle Utilities overview and SAP IS-U guide for how each platform handles the MDM integration. ## Category Two: Grid Sensor and Distribution Automation Devices Distribution automation equipment includes smart reclosers, capacitor bank controllers, line sensors, and substation IEDs (intelligent electronic devices). These do not feed the billing system. Their data destination is the SCADA/EMS layer or, increasingly, the ADMS. GE Vernova GridOS and Schneider Electric EcoStruxure Grid both ingest this telemetry for fault detection, isolation, and restoration (FDIR) and for volt-VAR optimization. Communication protocols are predominantly DNP3 over serial or IP and IEC 61850 for modern substation automation. This data path is entirely separate from the AMI-to-MDM-to-CIS flow. Customer outage notification is the integration point that bridges the two domains. The ADMS detects a fault and identifies affected customers using the GIS network model; it then passes outage records to the OMS. The OMS or the CIS triggers customer notifications via SMS, IVR, or a customer portal. ## Water and Gas Sensor Applications Water utilities deploy pressure and flow sensors throughout distribution networks to detect non-revenue water loss. These sensors feed supervisory control systems and, in more modern implementations, asset management platforms. The CIS still handles billing, but it receives water consumption from AMR/AMI devices using a similar head-end and MDM pattern as electric metering. Gas distribution automation uses electronic pressure regulators and leak detection sensors whose telemetry goes to SCADA, not the CIS directly. ## Predictive Maintenance: A Real Benefit With Realistic Scope Vibration and temperature sensors on rotating equipment (pumps, compressors, transformer cooling fans) generate condition data that feeds predictive maintenance workflows. SAP PM (Plant Maintenance) and IBM Maximo are the common destinations. The value is real: catching a bearing failure before it causes an unplanned outage is straightforward to quantify. What requires caution is the assumption that every sensor feed is immediately actionable. Building the analytics layer to turn raw vibration data into a reliable maintenance trigger takes time and domain expertise. ## The Integration Challenge The diversity of protocols, head-end systems, and MDM platforms makes IoT integration in utilities one of the more complex middleware problems in industrial software. SAP IDocs, Oracle SOA-based web services, and REST APIs all appear in the same integration architecture depending on system age. Before adding new sensor sources, map the existing data flows from device to application. The question is rarely “can we collect this data?” and more often “where does it land, and who maintains the integration?” AvanSaber advises utilities on these integration architectures. ## Frequently asked questions Where does AMI meter data go before it reaches the CIS? From the meter it travels over RF mesh or cellular to the AMI head-end, then to a Meter Data Management system (Oracle MDM or SAP MDG-U) for validation and VEE, then to the CIS for billing. What is VEE in the context of smart metering? Validation, Estimation, and Editing. The MDM checks interval reads for gaps or anomalies, estimates missing values using approved algorithms, and produces a clean dataset for the CIS. How does IoT sensor data reach the ADMS? Distribution automation devices (reclosers, capacitor banks, smart switches) push telemetry via DNP3 or IEC 61850 to the SCADA/EMS or directly to an ADMS like GE Vernova GridOS or Schneider EcoStruxure. --- ## Utility ERP and CIS Implementation: Best Practices URL: https://www.utilitieslabs.com/implementation-best-practices/ # Utility ERP and CIS Implementation: Best Practices AvanSaber Research/August 22, 2025/Updated June 2, 2026/2 min read Most utility ERP and CIS programs do not fail on features. They fail on data, rate complexity, cutover, and partner fit. These are the practices that consistently de-risk a SAP IS-U, Oracle CC&B, or Cayenta CIS implementation, drawn from how these programs actually run. ## Scope around meter-to-cash, not modules The spine of the program is the revenue process: meter read and validation, rating, billing, invoicing, payment, collections, and dunning (in SAP, the FI-CA subledger). Scope and test against that end-to-end flow, including the awkward cases (estimated reads, prorated bills, move-in/move-out) rather than a module checklist. ## Treat rate complexity as the long pole Tiered, time-of-use, seasonal, and deregulated-market rates multiply test cases faster than anything else. Inventory every rate, rider, and tax early, and build the rate configuration and its test matrix first. Late-discovered rate logic is the most common schedule killer. ## Cleanse legacy data before migration, not after Bad meter, account, and balance data in the old CIS resurfaces as billing errors after cutover. Profile and cleanse in the legacy system first, then migrate, then reconcile. Plan the data work as a first-class workstream, not a final-week task. ## Run parallel billing before cutover Bill the same population in the old and new systems and reconcile the results. Parallel billing is the single most effective proof that the new platform bills correctly, and it is what lets you switch off the legacy system with confidence. ## Phase the cutover Big-bang cutovers on high-volume billing concentrate risk. Phasing by service type (electric, then gas, then water) or by region contains blast radius and lets the team learn before scaling. ## Choose the partner as carefully as the platform Partner quality decides program outcomes more often than the software does. Weight regional implementation experience and reference programs heavily. For platform selection itself, start with the Oracle vs SAP comparison, the Cayenta CIS review, and the utility software selection guide. For the SAP-specific integration layer, see the SAP IDocs guide. ## Frequently asked questions Why do utility CIS implementations fail? Rarely the software. The common causes are underestimated rate complexity, dirty legacy data migrated as-is, big-bang cutovers on high-volume billing, and a weak implementation partner. Each is avoidable with the practices below. How long should a utility CIS implementation take? A mid-to-large utility CIS or billing program commonly runs one to three years depending on service types, customer count, rate complexity, and integration scope. Phasing by service or region is normal and reduces risk. --- ## Digital Transformation in Utilities: A CIS and Data-First Guide URL: https://www.utilitieslabs.com/implementing-digital-transformation-in-the-utilities-sector-a-step-by-step-guide/ # Digital Transformation in Utilities: A CIS and Data-First Guide AvanSaber Research/June 7, 2023/Updated June 2, 2026/3 min read Digital transformation in utilities gets framed as a technology procurement exercise more often than it should. The utilities that make real progress treat it as a data and process problem first, then ask which platform can support the improved processes. This guide is organized around that distinction. ## Start With the Billing and Customer Data Foundation Before selecting a new CIS or ERP, utilities need an honest audit of current data quality. Billing history, account hierarchies, rate assignments, meter-to-account linkages, and service location records are the raw material that any new platform will inherit. In SAP IS-U environments, this means reviewing the IDoc transaction history, the FI-CA contract account structure, and how consumption data flows from MDM into billing. In Oracle CC&B shops, the focus is typically the OUAF configuration layer and how customizations have been layered over the base product. Poor data quality surfaces late in CIS migrations and causes the most costly delays. Build a data profiling exercise into the early project phase, not as a box to check, but as a risk management tool. ## Define Outcomes Before Selecting Technology A digital transformation roadmap without defined outcomes becomes a technology refresh that solves yesterday’s problems. Useful outcome targets for utilities include: time-to-bill after meter read, call handle time for billing inquiries, percentage of meter reads processed without manual intervention, and customer portal adoption rates. Once outcomes are defined, the platform selection becomes a filtering exercise rather than a beauty contest. Our guide to choosing utilities software covers evaluation criteria in detail, and our implementation best practices guide addresses vendor management and project governance. ## The CIS-ERP-MDM Integration Architecture The meter-to-cash process spans at least three systems: the meter data management platform, the CIS (which holds the customer, account, and rate), and the financial system (which handles receivables, collections, and ledger posting). In large utilities, SAP IS-U integrates FI-CA for collections and the broader SAP S/4HANA financials stack. Oracle CC&B operates within the Oracle Utilities Application Framework and can connect to Oracle Financials. Cayenta CIS from Harris Computer handles the full billing and customer management stack for mid-market utilities. The integration seams between these systems are where transformation projects most often introduce errors. Map the data flows before committing to a go-live date. A pilot covering a representative sample of account types, rate structures, and billing scenarios will surface edge cases that UAT scripts typically miss. ## Change Management Is Not Soft Work CIS replacements require billing staff, customer service representatives, and field operations personnel to change deeply ingrained workflows. Training timelines are routinely underestimated. The organizations with the smoothest go-lives are those that involved frontline staff in UAT and process design, not just IT and project managers. NERC CIP compliance teams and cybersecurity staff also need to be part of the planning process if the new platform touches any operational technology data, which is increasingly the case as AMI data flows through the same infrastructure as customer records. ## Phased Rollout Reduces Go-Live Risk A phased approach, starting with a limited geography or customer class, gives the project team the ability to identify integration failures, billing calculation errors, and performance problems without exposing the full customer base. This is standard advice, but the pressure to consolidate data centers or decommission legacy licenses often pushes utilities toward big-bang go-lives. Resist that pressure where possible. ## Monitor Against the Outcomes You Defined Post-go-live, track the outcome metrics established in the planning phase. Time-to-bill, manual exception rates, and customer contact volumes are leading indicators that something in the billing engine or the meter data pipeline is not working as expected. Platforms like Oracle CC&B and SAP IS-U both provide operational dashboards, but custom reporting against the defined KPIs is usually necessary. Digital transformation in utilities is a multi-year program, not a project with a single end date. The organizations that sustain momentum are those that treat the initial CIS go-live as the beginning of a continuous improvement cycle rather than the finish line. For an independent assessment of your current systems and a realistic transformation roadmap, contact AvanSaber. ## Frequently asked questions Where do most utility digital transformation projects fail? Data quality and integration. Projects that underestimate the effort to clean up billing history, reconcile meter data, and map legacy account structures to a new system tend to run long and over budget. Should a utility replace its CIS before or after implementing AMI? There is no universal answer, but AMI deployments that precede CIS modernization often create technical debt when the new meter data volumes outpace what the old CIS can process. Sequencing matters. How long does a CIS replacement typically take? For a mid-size utility, 18 to 36 months is a realistic range for a full CIS replacement. Larger investor-owned utilities on SAP IS-U or Oracle CC&B projects have run longer. --- ## Sustainability in Utilities: Software That Makes It Measurable URL: https://www.utilitieslabs.com/incorporating-sustainability-practices-in-the-utilities-industry/ # Sustainability in Utilities: Software That Makes It Measurable AvanSaber Research/June 17, 2023/Updated June 2, 2026/3 min read Sustainability commitments in the utility industry are increasingly subject to regulatory scrutiny and investor pressure. The gap between a policy commitment and a verifiable, auditable result runs directly through the software systems that manage metering, billing, customer programs, and financial reporting. This article focuses on the systems side of sustainability, where the real implementation work happens. ## Demand Response Programs Require CIS Flexibility Demand response is the most mature software-enabled sustainability lever available to distribution utilities. A DR program pays customers to reduce load during peak events. Billing for that credit, tracking program enrollment, and settling with grid operators requires a CIS platform with a configurable rate engine. Oracle CC&B handles complex rate structures natively, including tiered rates, time-of-use rates, and demand charge credits. SAP IS-U’s FI-CA module manages the financial settlement side. Cayenta CIS, used by many mid-market utilities, supports DR program enrollment through its customer portal ecosystem. The limiting factor at most utilities is not the platform capability but the rate configuration effort required to build and maintain program logic, which tends to be underestimated during program design. ## DER Integration and the Net Metering Billing Problem As residential solar penetration grows, utilities face a billing challenge: net metering requires reading the meter in both directions, calculating the netting period (monthly, annual), applying the correct export rate, and presenting a bill that customers can understand. This is harder than it sounds when the underlying CIS was configured for simple residential consumption billing. Distributed energy resource management systems from GE Vernova and Schneider EcoStruxure handle the grid-side coordination, forecasting DER output, managing dispatch signals, and communicating with the ADMS layer. But the customer-facing settlement and billing remains a CIS responsibility. Utilities that have not updated their billing configuration to handle bidirectional metering accurately are creating billing disputes that erode customer trust and generate regulatory complaints. ## ESG Reporting Depends on Data Integration Credible environmental, social, and governance reporting requires the ability to trace generation source data to specific customer accounts and time intervals. That means the SCADA/EMS generation data, the AMI interval consumption data from Itron or Landis+Gyr meters, and the financial general ledger must all be reconcilable. SAP S/4HANA Utilities includes sustainability-adjacent reporting modules, and Oracle Utilities has reporting frameworks that can pull from MDM and billing. However, the data quality prerequisite is strict. Utilities with fragmented MDM implementations, manual meter reading exceptions, or billing adjustments not tied to verified read data will produce ESG reports that cannot withstand audit. The investment in data quality described in our digital transformation guide is therefore a prerequisite for credible sustainability reporting, not a separate workstream. ## Water Utilities and Non-Revenue Water For water utilities, the sustainability equivalent of demand response is non-revenue water (NRW) reduction. Water that is produced, treated, and pumped but never billed represents both an operational cost and an environmental waste. SCADA-integrated leak detection systems, pressure monitoring, and district metered area (DMA) analytics are the tools available. These feed into the CIS only at the billing reconciliation point, but the operational data infrastructure that supports NRW management is closely related to what is needed for sustainability reporting. ## Honest Assessment: Sustainability as an Outcome, Not a Feature Software vendors increasingly market sustainability as a product feature. The honest framing is that software enables sustainability only when the underlying data infrastructure is sound and the rate and program structures are correctly configured. A utility can purchase a DERMS platform and still fail to demonstrate measurable demand reduction if program enrollment, dispatch settlement, and billing are not working together. Our modern software solutions overview covers the broader technology landscape. For the specific question of how DER trends are changing what utility platforms must do, see the impact of renewable energy on utility providers. For help assessing whether your current CIS and billing infrastructure can support your sustainability program goals, contact AvanSaber. ## Frequently asked questions What software supports demand response program management? CIS platforms with configurable rate engines (Oracle CC&B, SAP IS-U, Cayenta CIS) handle the billing side. DERMS platforms from GE Vernova or Schneider EcoStruxure handle the dispatch and settlement side. How do utilities report on ESG commitments accurately? Credible ESG reporting requires interval meter data from AMI, generation mix data from the SCADA/EMS layer, and financial data from the ERP. Utilities without MDM-to-ERP integration often resort to manual reconciliation, which introduces reporting risk. Is demand response different from energy efficiency? Yes. Demand response is a dispatchable reduction in load, typically compensated through a rate program. Energy efficiency is a permanent reduction in consumption. Both require CIS support for program enrollment, billing credits, and measurement and verification. --- ## Innovation in Utilities: What's Actually Changing in CIS and ERP URL: https://www.utilitieslabs.com/innovation-in-utilities-how-modern-software-is-driving-change/ # Innovation in Utilities: What's Actually Changing in CIS and ERP AvanSaber Research/September 18, 2023/Updated June 2, 2026/3 min read The word “innovation” in the utility software industry often refers to incremental improvements dressed in ambitious language. This article focuses on changes that are genuinely altering how utilities operate their CIS, ERP, and grid management systems, rather than on aspirational technology narratives. This piece takes a different angle than our modern software solutions overview, which covers the technology landscape broadly. Here the focus is on what has already changed in production deployments. ## Interval Billing: The Change That Reshapes CIS Architecture The most consequential change in utility CIS over the past decade is not a software platform replacement; it is the shift from estimated or monthly-read billing to interval-based billing driven by AMI data. When a utility moves from monthly meter reads to 15-minute interval data from Itron or Landis+Gyr meters, the bill calculation workload increases by orders of magnitude. SAP IS-U and Oracle CC&B were both designed around periodic billing cycles. Adapting them to consume interval data from an Oracle MDM or SAP MDG-U and calculate time-of-use rates required significant platform evolution. SAP S/4HANA Utilities introduced changes to the FI-CA engine and the Energy Data Management module to handle interval rate calculation at scale. Oracle CC&B’s successor, Oracle Customer to Meter (C2M), rebuilds the rating engine with interval data as the native input. This is not a minor upgrade. It affects rate configuration, bill presentation, customer service workflows, and the integration between the MDM and CIS. Utilities that have completed AMI rollouts but have not yet updated their rate structures are carrying billing systems that are not using the available data. ## AI in Production: Narrow But Real Vendor claims about AI in utility software span a wide range of maturity levels. The applications that are genuinely in production use today: Collections and credit risk. Several CIS platforms, including SAP IS-U and Oracle CC&B, offer predictive arrears models that score accounts by likelihood of non-payment. These models inform dunning workflow configuration, allowing utilities to differentiate outreach for high-risk versus low-risk accounts. The models use billing history, payment history, and demographic inputs. Customer service automation. Cayenta CIS’s Cayla AI assistant handles call deflection for common inquiries: outage status, bill explanation, payment arrangement. This is a production capability, not a pilot. It integrates with Cayenta’s Silverblaze customer portal for consistent self-service across channels. Predictive maintenance scoring. Asset management platforms (SAP PM, IBM Maximo) can consume sensor data from field devices to score maintenance priority. The operational value is real on well-instrumented assets; the challenge is data availability and quality for assets that lack sensors. ## Grid Software: ADMS and DERMS On the grid operations side, the ADMS (Advanced Distribution Management System) represents genuine capability advancement over older OMS platforms. GE Vernova GridOS and Schneider EcoStruxure Grid both provide fault detection, isolation, and restoration (FDIR) automation that reduces outage duration without dispatcher intervention. Volt-VAR optimization running continuously reduces technical losses on distribution feeders. Distributed Energy Resource Management Systems (DERMS) are newer and less uniformly deployed. As solar, storage, and EV charging penetration increases, utilities need software to manage the two-directional power flows these resources create. Integration between the DERMS, ADMS, and the CIS (which must handle net metering billing) is an active area of platform development. ## What Has Not Changed Much The core billing and financial settlement engines in SAP IS-U and Oracle CC&B remain architecturally similar to their early-2000s designs. Both platforms have added modules and modernized UIs, but the fundamental rate engine, billing run, and FI-CA payment processing logic have not been replaced. For utilities on these platforms, the innovation is mostly at the edges: portals, analytics, AI-assisted workflows, and AMI integration. Cayenta CIS occupies a different position: it is a more modern codebase designed for small to mid-size utilities, with a lower total cost of ownership than the major platforms for straightforward billing configurations. For utilities evaluating where to invest in genuine operational improvement rather than software marketing, AvanSaber provides independent analysis of specific platform capabilities against your requirements. ## Frequently asked questions What is the most significant recent innovation in utility CIS? The shift from monthly batch billing to interval-based bill calculation driven by AMI data. This enables time-of-use rates and demand charges without manual meter reads. How is AI being used in utility software today? AI-powered collections and arrears prediction in CIS platforms, Cayla AI in Cayenta CIS for customer service automation, and predictive maintenance scoring in asset management systems. What is the difference between innovation in utility OT versus IT? OT innovation (ADMS, FDIR, volt-VAR optimization) is driven by grid reliability and cost mandates. IT innovation (CIS, ERP, portals) is driven by customer expectations and billing complexity from new rate structures. --- ## AI for Utility Operations on SAP IS-U and Oracle CC&B URL: https://www.utilitieslabs.com/integrate-ai-with-sap-oracle-streamline-utility-operations/ # AI for Utility Operations on SAP IS-U and Oracle CC&B AvanSaber Research/August 22, 2025/Updated June 1, 2026/1 min read “Integrate AI with SAP and Oracle” is a broad promise. For utility operations the useful version is specific: AI helps on a handful of data-rich tasks around SAP IS-U and Oracle CC&B, and it helps most when the ERP stays the system of record. This guide covers where it pays off and how to wire it safely. ## Where AI helps on utility ERP - Billing-exception triage. Large utilities generate thousands of billing exceptions. A model can rank and route them so analysts work the ones that matter first. - Demand and load forecasting. Consumption and weather data feed forecasts that improve procurement and operations. - AI-assisted customer service. Drafting responses and surfacing the right account context from the CIS speeds up agents. - Metering anomaly detection. Spotting losses, tampering, and faulty meters in interval data. ## How to integrate it safely Keep SAP IS-U or Oracle CC&B as the system of record. AI reads from the ERP, produces a prediction or recommendation, and writes back a suggested action that a person or a controlled workflow approves. That pattern captures the value without letting a model post unreviewed changes to billing or customer records. ## Where this fits For the platform context, see the SAP IS-U guide and the Oracle vs SAP comparison. ## Frequently asked questions How does AI integrate with SAP IS-U or Oracle CC&B? AI sits alongside the system of record, not inside it. It reads data from IS-U or CC&B (billing exceptions, consumption, service history), produces a prediction or recommendation, and writes back a suggested action that a person or a controlled workflow confirms. The ERP stays the source of truth. What utility tasks is AI good at today? Billing-exception triage, demand and load forecasting, AI-assisted customer service, and metering anomaly detection. These are bounded, data-rich tasks where a model can be measured against a clear outcome. --- ## GIS in Utilities: Esri ArcGIS, Network Models, and Integration URL: https://www.utilitieslabs.com/leveraging-geospatial-information-systems-gis-in-utility-management/ # GIS in Utilities: Esri ArcGIS, Network Models, and Integration AvanSaber Research/June 28, 2023/Updated September 24, 2026/3 min read Geospatial data is foundational infrastructure for a utility, not just a mapping convenience. Every work order, outage record, and customer notification that depends on location accuracy ultimately traces back to whether the GIS network model is correct. Esri ArcGIS is the dominant platform in the utility GIS market, and understanding how it connects to other enterprise systems is essential for any utility planning a technology investment. ## The Network Model: What GIS Actually Stores For an electric distribution utility, the GIS holds the topological network model: which transformer feeds which secondary, which switch isolates which feeder section, where each service point connects to the network. For a water utility, it holds pipe diameter, material, age, and valve connectivity. This is not just cartography. The connectivity data is what enables automated outage tracing in the OMS and hydraulic modeling for water distribution. Esri’s ArcGIS Utility Network, the current architecture replacing the older Geometric Network, enforces referential integrity across the network topology. A transformer cannot exist without belonging to a substation; a service point must connect to a distribution conductor. This constraint model is what keeps the GIS accurate enough to trust for operations. ## Integration With OMS and ADMS The OMS (outage management system) consumes the GIS network model to predict customer impact when a device operates or fails. When a recloser opens, the OMS traces downstream through the GIS topology to count the affected service points and generate customer notifications. GE Vernova and Schneider EcoStruxure each maintain their own internal network model but synchronize it from the GIS as the system of record. The ADMS extends the OMS with volt-VAR optimization and fault location. Both functions require an accurate impedance model that is derived from the GIS attributes (conductor type, length, transformer kVA). An out-of-date GIS produces incorrect ADMS results, which is why utilities that have invested heavily in distribution automation must also invest in GIS data quality programs. ## Integration With CIS and Customer Data The GIS and the CIS communicate primarily at the service point level. Each customer account in Oracle CC&B, SAP IS-U, or Cayenta CIS has a service point that corresponds to a GIS feature. During an outage, the OMS passes affected service point IDs to the CIS or a separate customer notification platform. Customer portals display outage status by pulling location data from this same chain. Field work management is a second integration path. SAP Work Manager, Oracle Field Service, and Cayenta Work Management create work orders that include GIS coordinates and network asset identifiers. Crews in the field use Esri ArcGIS Field Maps or comparable mobile apps to record as-built changes, which must then be posted back to the authoritative GIS. ## Integration With Work Management Asset replacement and construction projects depend on accurate GIS records to estimate material needs and route crew travel. When a new transformer is installed, the GIS record must be updated with the new asset ID, and that ID must link to the asset management system (SAP PM or IBM Maximo) to start the maintenance lifecycle. Utilities that run SAP IS-U frequently use SAP PM alongside it; the GIS-to-PM asset linkage is a recurring integration project. ## Data Quality: The Persistent Challenge The value of GIS depends entirely on data accuracy. Many utilities hold legacy GIS records that reflect as-designed rather than as-built conditions. Conductors replaced during emergency repairs are not always added back to the GIS. Transformer ratings differ between GIS and engineering records. These gaps produce incorrect outage predictions and faulty field dispatches. A phased GIS data quality program, typically starting with the highest-voltage feeders and working down to secondary, is standard practice before connecting the GIS to an ADMS or new OMS. The modern software transformation overview places GIS accuracy as a prerequisite for most advanced grid applications. For help mapping out a GIS integration architecture with your OMS, CIS, or work management platform, contact AvanSaber. ## Frequently asked questions What role does Esri ArcGIS play in utility network management? ArcGIS maintains the authoritative network model: the topology of pipes, wires, valves, and switches. It feeds the OMS and ADMS with the connectivity data they need for outage tracing and switching. How does GIS connect to the CIS for customer impact analysis? During an outage, the OMS uses the GIS network model to trace upstream from a fault and identify affected service points. Those service point IDs link to customer accounts in the CIS for notification. Does Esri ArcGIS Utility Network replace older geometric network models? Yes, ArcGIS Utility Network (introduced in ArcGIS Pro) replaces the older Geometric Network. Migration is a significant project for utilities on long-lived ArcMap deployments. --- ## Predictive Analytics in Utility Management: Forecasting and Assets URL: https://www.utilitieslabs.com/leveraging-predictive-analytics-for-efficient-utility-management/ # Predictive Analytics in Utility Management: Forecasting and Assets AvanSaber Research/June 8, 2023/Updated June 2, 2026/3 min read Predictive analytics in utilities is a broad term covering several distinct applications. This article organizes those applications by the business problem they address and is direct about the data requirements that determine whether each application works in practice. ## Demand Forecasting Short-term demand forecasting, from a few hours to a few days ahead, is used by grid operators for economic dispatch decisions and by distribution engineers for switching and maintenance scheduling. Machine learning models trained on SCADA load history, AMI interval reads, and weather data have become the standard approach, improving on classical regression models particularly for circuits with significant variable renewable generation or demand response assets. Medium-term forecasting, from months to several years, informs infrastructure investment decisions: when to upgrade a substation transformer, when a distribution circuit will need reconductoring, or when a pressure zone needs additional storage. These models need clean AMI consumption data and growth trend information from planning systems. Water utilities apply demand forecasting to predict daily and seasonal water demand, which drives pump scheduling and reservoir management decisions. The signal sources are the same: meter data, weather, and historical consumption, often with population and development data added for medium-term planning. ## Asset Risk Scoring Asset risk scoring ranks infrastructure assets by their probability of failure within a planning horizon. For electric utilities the primary asset classes are distribution transformers, underground cables, and overhead feeder sections. For water utilities they are mains, pumps, and valves. Models for asset risk combine age and vintage data from the asset register with operational data from SCADA (loading history, fault events) and AMI (outage events attributable to an asset). The output is a ranked list of assets for each planning cycle, which a capital planning team uses to prioritize replacement and refurbishment. This is described in detail in AI predictive maintenance for utilities and AI tools for predictive maintenance: the data pipeline. The relevant point for predictive analytics broadly is that the model is only as useful as the asset register data feeding it. Utilities whose ERP or EAM records are incomplete or inconsistently maintained get unreliable risk scores. ## Revenue Protection and Loss Analysis Predictive analytics can identify accounts with consumption patterns that are statistically inconsistent with their premise type, historical behavior, or neighboring accounts. This is useful for revenue protection programs targeting meter tampering or unauthorized service, and for identifying meter or MDM data errors that are causing billing inaccuracies. The model flags candidates for investigation; a field crew confirms. False positives are common enough that treating model output as a work queue for investigation, not as a billing action, is the correct process design. The CIS remains the system of record and billing is not adjusted until investigation is complete. ## Customer Behavior and Churn Analytics Oracle Opower applies behavioral analytics to CIS and AMI data to identify customers who are most likely to benefit from demand response enrollment or energy efficiency programs. Uplight offers similar capabilities for customer engagement programs. These tools help utilities target outreach to customers where the probability of enrollment is higher, improving program cost-effectiveness. For utilities in deregulated markets, predictive churn modeling identifies customers who are likely to switch to a competitive supplier, enabling proactive retention outreach. The deregulated market section of this site covers the CIS configuration requirements specific to competitive retail environments. ## Data Requirements Are the Honest Constraint Every application in this article works better with more complete, cleaner data. The practical sequencing for utilities is to assess data quality before committing to a predictive analytics program, address the most critical gaps (particularly in work order coding and interval read completeness), and then build models against a data set that is representative of the actual system. Vendors will often propose model designs before data quality work is complete. That ordering tends to produce pilots that cannot be reproduced in production. The honest approach is to treat data quality assessment as part of the project scope from the start. For the utility data management context, see maximizing utility data management with AI. For the broader picture of AI trends across the sector, see AI transforming the energy sector. The AvanSaber team advises on predictive analytics program design for utilities on SAP, Oracle, and Cayenta platforms. ## Frequently asked questions What is the minimum data history needed for a reliable demand forecast model? Most practitioners recommend at least two full years of interval load data to capture seasonal patterns reliably. For substation-level forecasting, SCADA load history combined with AMI interval reads typically provides sufficient depth. Shorter histories are workable but require more careful validation of seasonal extrapolation. How does predictive analytics differ from descriptive reporting? Descriptive reporting tells you what happened, such as total consumption last month by rate class. Predictive analytics uses historical patterns to estimate what is likely to happen next: which customers are likely to churn to a community choice aggregator, which transformers are likely to fail before the next inspection cycle, or what load will look like on a hot afternoon three days from now. Should predictive model outputs go directly into billing or rate decisions? No. Model outputs should be reviewed by qualified analysts before affecting billing or rate actions. Predictive models produce probabilities and estimates, not confirmed facts. The CIS remains the system of record; model outputs inform decisions made by people. --- ## XR for ERP Data Visualization: Niche Value and Real Constraints URL: https://www.utilitieslabs.com/leveraging-xr-for-erp-data-visualization/ # XR for ERP Data Visualization: Niche Value and Real Constraints AvanSaber Research/April 2, 2024/Updated June 2, 2026/3 min read Extended reality covers virtual reality (VR), augmented reality (AR), and mixed reality (MR). In an ERP context, the practical question is whether presenting data in three-dimensional or immersive form produces better decisions than a well-designed two-dimensional dashboard. The honest answer is: sometimes, for specific data types, and at meaningful development cost. This post maps out where that value genuinely appears and where it does not, particularly for utilities running Oracle or SAP ERP and CIS systems. ## Where Dimensional Representation Adds Meaning Not all ERP data benefits from 3D or immersive representation. Financial aggregates, AR aging reports, and billing exception queues are inherently tabular; rendering them in a virtual environment adds cognitive overhead without adding information. The data types that gain from spatial representation share a common trait: they describe relationships that have a physical or network topology dimension. Examples in a utility context: - Network and grid topology: Electric distribution feeders, gas pipeline segments, or water main networks are genuinely spatial. Viewing feeder loading, outage segments, or pressure zones in a 3D model drawn from NMS or GIS data can reveal spatial correlations that a tabular list of circuit IDs does not. - Asset and infrastructure hierarchies: Plant maintenance records in SAP or Oracle showing transformer, switchgear, and cable assets are easier to navigate when rendered as a spatial asset tree than when browsed as nested table rows. - Supply chain and logistics flows: Multi-site utilities with material warehousing can benefit from 3D warehouse layouts overlaid with inventory levels from the ERP. These are legitimate cases. They are also specific use cases, not a general argument that all ERP data should move to XR. ## How XR Connects to an ERP System XR applications do not have native connectors to SAP IS-U, Oracle CC&B, or OUAF. An XR data visualization tool typically retrieves data through the ERP’s published APIs: SAP’s OData services via the Business Technology Platform or Gateway, or Oracle Utilities’ REST services. The data is then rendered in a 3D engine such as Unity or Unreal, or in a browser-based WebXR environment. The latency of this approach matters. ERP APIs designed for transactional use add 200 to 500 milliseconds of round-trip overhead, which is acceptable for dashboard refresh but not for near-real-time monitoring. Utilities that need truly real-time XR overlays, such as live grid state during a restoration event, need a separate streaming data path, typically from the SCADA or ADMS layer, not the ERP directly. ## The Cost Argument Is Honest and Significant Building a production XR data visualization for an ERP dataset requires 3D development skills that are separate from ERP implementation expertise. The XR application must be maintained through ERP patch cycles. API changes in CC&B patch bundles or SAP Support Packages can break XR data feeds. Device procurement and management for VR headsets or AR glasses adds infrastructure cost. For most utilities, a well-designed Oracle Analytics Cloud dashboard or SAP Analytics Cloud story delivers 80 to 90 percent of the decision support value at a fraction of the XR development cost. XR investment is justified when the specific data type is spatial, when the decision audience regularly works in that spatial context, or when there is a safety or training simulation value that two-dimensional tools cannot replicate. ## Relationship to the Broader ERP-AR Discussion This post focuses on data visualization. The question of AR for live field operations, particularly overlaying ERP work order and asset data on physical equipment, is addressed in the companion posts on Oracle ERP with AR and VR and XR for real-time SAP data visualization. The foundational discussion of how AR intersects with ERP AR modules is at Enhancing ERP: Integrating AR with SAP. For utility ERP teams assessing where visualization investment will have the most impact, AvanSaber can help evaluate whether XR or conventional analytics better fits the specific data and decision context you are working with. ## Frequently asked questions What ERP data types benefit most from XR visualization? Geospatial and network topology data, supply chain and logistics flows, and plant maintenance asset hierarchies are the categories most commonly cited as gaining clarity from 3D or immersive visualization versus traditional 2D dashboards. Does XR work with standard ERP APIs? XR applications connect to ERP systems via REST APIs, OData, or SOAP services. Most major ERPs expose these endpoints, but the APIs were designed for transactional use, not real-time streaming, which affects the freshness and latency of data visible in an XR environment. Is XR for ERP visualization commercially available or still experimental? A small number of commercial tools exist, primarily in the industrial and field-service space. For general ERP financial or billing data, XR visualization remains largely custom-built and is not yet mainstream. --- ## XR Visualization of Real-Time SAP IS-U and S/4HANA Data URL: https://www.utilitieslabs.com/leveraging-xr-for-real-time-sap-data-visualization/ # XR Visualization of Real-Time SAP IS-U and S/4HANA Data AvanSaber Research/July 1, 2024/Updated June 2, 2026/3 min read SAP IS-U on ECC and S/4HANA Utilities together manage billing, device management, meter reading, and financial settlement for electric, gas, and water utilities. Most of this data is viewed through standard SAP Fiori apps, SAP Analytics Cloud reports, or legacy SAPGUI transactions. XR visualization layers are a non-standard addition, but for specific data types, they provide a genuinely different way of interpreting operational state. This post is specifically about the SAP side of the utility XR question. The companion post on XR for general ERP data visualization covers Oracle and cross-platform considerations. The AR-for-SAP financial module angle is covered separately at Enhancing ERP: Integrating AR with SAP. ## SAP IS-U Data Types With a Spatial Dimension SAP IS-U manages connections and meter points against a premises and installation hierarchy. That hierarchy has a physical location: a street address, a grid coordinate, or a network topology position. When that location data is joined with operational state, the result is inherently spatial. Scenarios where XR adds value over a conventional SAP dashboard: - Outage and restoration context: Displaying FI-CA account counts by premise location over a GIS network model, drawn from SAP IS-U meter installation records, lets an operations manager see revenue exposure from an outage spatially rather than as a count in a table. - Field service and PM asset inspection: A technician using AR glasses during a substation inspection can surface SAP Plant Maintenance equipment records, maintenance history, and open work orders against the physical asset in their field of view, without switching to a mobile device. - Billing exception geography: Accounts with open billing exceptions or disconnection orders, extracted from IS-U contract account data, can be plotted on a 3D grid model to identify concentrations that point to a systematic metering or tariff configuration issue. ## Integration Architecture on SAP BTP Real-time XR visualization of SAP data requires a defined integration path. SAP Business Technology Platform provides the recommended current approach: - SAP API Management on BTP governs and rate-limits OData API calls from the XR application to IS-U or S/4HANA Utilities. - SAP Event Mesh can publish IS-U business events, such as meter reading exceptions or billing run completions, to the XR layer without the application polling SAP constantly. - SAP HANA Cloud can serve a denormalised spatial view of IS-U data, joining installation location, device, and account status, which the XR engine queries for its rendering feed. This architecture keeps the XR application decoupled from the core SAP system. When IS-U is patched or when a migration to S/4HANA Utilities occurs, the API Management layer absorbs most of the API change management rather than requiring the XR application to be rebuilt. For a detailed look at SAP BTP’s broader role in utility integrations and extensions, see Unlocking Business Potential with SAP BTP. ## S/4HANA Utilities and the Clean-Core Consideration Under SAP’s clean-core guidance, custom Z-code inside the IS-U core should be minimised. This applies equally to any XR data feed that was historically built as a custom ABAP report or RFC function module. On S/4HANA Utilities, the preferred path is to expose IS-U data through standard OData services or SAP BTP side-by-side extensions rather than calling bespoke RFCs from an XR application. This matters practically because a custom RFC-based XR data feed will not survive a technical upgrade from ECC IS-U to S/4HANA Utilities without rework. Building the integration via BTP from the outset avoids that rework. ## Honest Scope: Where XR Adds Less Value General SAP IS-U billing transactions, rate schedule maintenance, FI-CA financial postings, and IDOC monitoring are procedural tasks. They do not benefit from spatial or immersive representation. Pushing these workflows into an XR environment adds device and development overhead without improving the decision or the outcome. The value of XR in an SAP utilities context is narrow but real for spatial operational data. For teams evaluating whether an XR initiative alongside an SAP IS-U or S/4HANA Utilities programme is worth the investment, AvanSaber can assess what your specific operational data and use cases justify. ## Frequently asked questions What SAP data is most useful to visualise in XR for a utility? Network asset hierarchies in SAP Plant Maintenance (PM), FI-CA account balances by geographic area, and field service order locations pulled from SAP Work Manager are the most spatially meaningful SAP datasets for utility XR use cases. How does an XR application connect to SAP IS-U or S/4HANA Utilities? XR applications typically use SAP OData APIs published through SAP BTP API Management or SAP Gateway. For near-real-time scenarios, SAP Event Mesh on BTP can push change events to the XR layer without constant polling. Does SAP offer any native XR tools for utilities? SAP has offered SAP Augmented Reality as an experimental capability and collaborates with third-party spatial computing vendors. There is no packaged SAP IS-U or S/4HANA Utilities XR product; integration requires custom development. --- ## Oracle Database Tuning for CC&B and Utility Billing Jobs URL: https://www.utilitieslabs.com/mastering-oracle-top-coding-tips-for-efficient-database-management/ # Oracle Database Tuning for CC&B and Utility Billing Jobs AvanSaber Research/October 24, 2023/Updated June 2, 2026/4 min read Oracle Database sits at the centre of every Oracle Customer Care and Billing (CC&B) and Oracle Utilities Application Framework (OUAF) deployment. When a gas, electric, or water utility runs its nightly billing cycle or month-end financial close, the database absorbs millions of row reads and writes across tables like CI_SA, CI_BSEG, CI_METER, and the financial summary structures. Poor query plans, missing index coverage, or bloated PL/SQL loops do not merely inconvenience developers; they break billing SLA windows and delay revenue recognition. This post focuses on the Oracle Database practices that matter most in a utility CIS context. ## Understand the CC&B Data Model Before Tuning Oracle’s OUAF schema is deeply normalised. A single bill segment calculation touches service agreements, rate schedules, read remarks, and contract riders across dozens of joined tables. Before adding an index or rewriting a query, map the join path the batch engine actually uses. The OUAF batch control framework logs job steps with timing; use those logs alongside Oracle Automatic Workload Repository (AWR) snapshots to identify which SQL IDs are costing the most time. A common mistake is tuning an ad-hoc report query while the real offender is a framework cursor used by every billing engine thread. AWR’s Top SQL section surfaces these accurately. ## Index Strategy for High-Volume Billing Tables CC&B installs ship with a baseline index set, but implementations often grow volumes far beyond the reference configuration. Key considerations: - Foreign key indexes on CI_BSEG and CI_BILL: Oracle does not automatically index foreign keys. Unindexed FKs cause table-level lock contention during batch deletes of old bill segments. - Function-based indexes for status filters: Billing queries frequently filter on status columns that are stored as single-character codes. A function-based index on `UPPER(BILL_STAT_FLG)` can help if application code mixes case. - Composite index column ordering: Put the most selective column first unless the query always supplies a leading equality predicate on a lower-selectivity column, such as account type. Avoid creating redundant indexes on the same leading columns. OUAF’s object maintenance routines rebuild statistics periodically; too many indexes slow that window. ## Partitioning for Billing History and Meter Reads Utility databases accumulate billing and meter-read history continuously. Range partitioning CI_BSEG on `READ_DT` or billing period columns lets the optimizer prune partitions during date-range billing re-runs and simplifies the purge of aged data without affecting active partitions. List-subpartitioning by operating company or rate class can further isolate batch threads that run per division. Before partitioning, validate that CC&B’s framework SQL uses the partition key in its WHERE clause. Partitioning on a column the billing engine never filters provides no pruning benefit and adds maintenance overhead. ## PL/SQL Batch Cursor Hygiene OUAF batch steps often invoke PL/SQL procedures that loop over large result sets. Two patterns that degrade performance at utility scale: - Row-by-row processing instead of bulk collect: Fetching one row at a time in a cursor loop at volumes of two million meter reads causes millions of context switches between the PL/SQL and SQL engines. Use `BULK COLLECT ... LIMIT` with a sensible array size, typically 500 to 1000 rows, and process via `FORALL` for DML. - Implicit commits inside tight loops: Committing after every row prevents undo space growth but fragments redo logs and causes excessive checkpoint activity. Commit after processing a full account batch or billing group instead. These patterns apply whether the PL/SQL is in custom extensions or in utility-specific procedures layered on the framework. ## Query Plan Stability Across Patching Cycles Oracle’s cost-based optimizer recalculates plans after statistics changes. CC&B patch bundles frequently refresh schema objects, which can shift plans unexpectedly. Use SQL Plan Management (SPM) baselines to lock critical billing job plans after acceptance testing. When a new plan is accepted during a patch, verify AWR metrics before retiring the old baseline. Avoid over-indexing as a shortcut to stability; a stable plan on a poor index structure is still a poor plan. ## Connection Pooling and Concurrency for Batch Threads OUAF billing runs parallel threads, each holding a database connection for its duration. Over-allocating threads relative to CPU and I/O capacity causes queue waits at the latch and buffer cache level. Use Oracle Database’s wait event data to distinguish CPU-bound bottlenecks from I/O waits before adding hardware. On engineered systems like Oracle Exadata, smart scan offloading changes the tuning calculus; consult the Exadata-specific AWR reports rather than generalising from non-engineered tuning guides. --- ## SAP ABAP for Utility IS-U: Performance, Clean Core, and BTP URL: https://www.utilitieslabs.com/mastering-sap-abap-essential-tips-for-efficient-programming/ # SAP ABAP for Utility IS-U: Performance, Clean Core, and BTP AvanSaber Research/November 29, 2023/Updated June 2, 2026/4 min read ABAP remains the primary development language for SAP IS-U on ECC and continues to be used in S/4HANA Utilities, though SAP’s clean-core direction shifts the balance toward BTP-based extensions over time. For utility developers and architects, the practical question is not whether to use ABAP at all but where it is still the right choice, how to write it well for the specific demands of utility processing, and when to route new work to SAP Business Technology Platform instead. ## Why Utility ABAP Has Distinct Performance Demands A generic ABAP performance tip, “avoid SELECT inside a loop,” matters everywhere. In SAP IS-U it is critical. Mass billing runs (transaction EA50 and its successor in S/4HANA Utilities) process service agreements in parallel work processes. Each work process executes rating, billing, and invoicing logic for its assigned batch. An ABAP routine with a database call per iteration that would take two seconds on 100 accounts takes 20 minutes on 60,000. That difference breaks the billing window. Key patterns to check in IS-U custom code: - Array buffering for rate component tables: Custom rate analysis code that queries rate schedule tables row-by-row during billing should buffer rate definitions into an internal table once per work process, not once per service agreement. - ABAP SELECT with FOR ALL ENTRIES: This construct is efficient when the input table is large and representative. It is inefficient for very small input sets or when the input table can be empty. Review custom code paths that pass dynamic account sets into FOR ALL ENTRIES. - Avoid functional modules with implicit commits inside a billing batch step: Mass processing steps that call function modules with built-in COMMIT WORK can cause data inconsistencies if the billing run is restarted after a partial failure. ## Clean Core Discipline in IS-U SAP IS-U on ECC has decades of Z-object accumulation in many installations. Utilities planning an S/4HANA Utilities migration will need to reclassify each Z-object: does it become a BTP extension, a custom ABAP object inside the adaptation layer, or something to retire entirely? SAP’s extensibility framework for S/4HANA provides key user exits and Business Add-Ins (BAdIs) that are explicitly supported across upgrades. Custom code that was previously implemented as direct table modifications or implicit enhancement points should migrate to these released APIs. The SAP Readiness Check and ABAP Test Cockpit (ATC) with the clean-core check variant both surface objects that need attention before a migration. Clean core is not about removing all custom logic. It is about placing custom logic in layers that SAP can upgrade without breaking. For utilities with complex rate structures, custom device category logic, or non-standard billing exceptions, some ABAP will always remain. The goal is to minimise the surface area that must be retested with every Support Package. ## When to Extend on BTP Instead SAP BTP is the right destination for utility extensions that: - Require external API integration, such as connecting IS-U to a third-party AMI headend, a demand response aggregator, or an outage notification service. - Have UX requirements better served by SAP Fiori or custom HTML5 than by ABAP reports or SAPGUI screens. - Need to run on a release cycle independent of IS-U core upgrades. - Benefit from SAP’s AI services, such as applying machine learning to billing exception triage or meter read anomaly detection without embedding that logic in ABAP. The SAP BTP guide on this site covers the platform’s integration and extension capabilities in detail. The ABAP Cloud development model, available on BTP and as the S/4HANA Utilities extension approach, uses only released ABAP APIs and prevents direct calls to SAP-internal database tables. ## Practical BAdI and User Exit Strategy for IS-U IS-U provides specific BAdIs for billing (ISU_BILLING_), rate calculation (ISU_RATE_), and invoicing. Using these released extension points rather than implicit enhancements or source code modifications means the custom logic survives SAP’s shipping of new IS-U functionality without manual reapplication. Document which BAdIs your implementation uses and their purpose; this documentation becomes critical during upgrade feasibility assessments. ## Testing and Regression Coverage IS-U billing logic is difficult to unit-test in isolation because it is deeply coupled to the rate schedule configuration and the state of service agreements in the test system. ABAP Unit tests are valuable for stateless helper classes and calculation routines but are insufficient for end-to-end billing testing. Complement unit tests with a regression billing run against a representative set of service agreements covering each rate class, and compare billing outputs before and after any ABAP change. For more on the SAP utilities landscape and integration practices, see the SAP IS-U pillar and SAP IDoc integration guide. For independent review of your IS-U ABAP landscape ahead of an S/4HANA Utilities migration, AvanSaber offers pre-upgrade code quality assessments. ## Frequently asked questions Why does ABAP performance matter specifically for SAP IS-U billing? IS-U mass billing processes, rate analysis runs, and invoicing runs process tens of thousands of service agreements per cycle. ABAP code that is inefficient at small scale becomes a billing window blocker at utility scale. What is clean core and why does it matter for utility ABAP? Clean core means keeping IS-U or S/4HANA Utilities modifications off the core SAP system and building extensions as BTP side-by-side apps or via SAP's extensibility framework instead. It reduces upgrade cost and compatibility risk when SAP ships new IS-U or S/4HANA Utilities releases. When should a utility developer use BTP instead of ABAP? When the extension is UI-heavy, integration-heavy, or likely to change frequently, BTP's low-code/no-code tools, Cloud Application Programming Model, or API Management are better fits than embedding logic in ABAP objects that will need re-testing on every SAP upgrade. --- ## VR for ERP Process Simulation and Design Review in Utilities URL: https://www.utilitieslabs.com/maximizing-efficiency-vr-benefits-for-erp-system-simulations/ # VR for ERP Process Simulation and Design Review in Utilities AvanSaber Research/June 10, 2024/Updated June 2, 2026/3 min read VR has two distinct roles in utility operations, and they are often conflated. One is field crew training, simulating switching procedures and emergency response, which is covered separately in the workforce training article. The other, covered here, is using VR for design review of capital infrastructure and for understanding complex ERP process flows during implementation projects. The second category is narrower and less mature, but it has specific legitimate applications. ## VR for Capital Infrastructure Design Review The strongest practical case for VR in utility operations, outside of safety training, is in the design review stage of capital projects. Before a new substation is built, before a gas regulator station is sited, before a water treatment expansion is engineered, project teams do design reviews from 2D drawings and 3D CAD models on flat screens. VR allows engineers, safety officers, and maintenance planners to walk through a proposed design at scale. This surfaces problems that flat-screen CAD review misses: a cable tray that creates a maintenance access conflict, a switchgear bay that leaves insufficient clearance for arc flash boundaries, or an equipment layout that puts a valve behind an obstacle. Catching these in design review costs far less than a change order during construction. The VR model is typically generated from the same CAD or BIM data used for engineering drawings, so there is no separate model-building effort if the design tools (Autodesk, Bentley, similar) support VR export. This is the key condition that makes the use case practical rather than expensive. This is a design tool for project teams, not an operational tool for grid management. The ADMS and ERP remain the operational systems. ## VR for ERP Process Flow Understanding The second application is less established but worth examining: using VR-style walkthroughs or simulation environments to help project teams understand how ERP processes connect during an implementation or upgrade project. SAP IS-U and S/4HANA Utilities have complex process chains: meter reading flows through device management into billing in FI-CA, service orders connect to PM work orders, and contract account postings feed the general ledger. Oracle CC&B has similarly interconnected flows through OUAF. For teams implementing or upgrading these systems, understanding how a configuration choice in one module affects downstream processes is a recurring challenge. A 3D visualization that maps these process flows, allowing a project manager to follow a transaction from meter reading through billing settlement through FI-CA posting, can help non-technical stakeholders understand system architecture decisions during an implementation. This is documentation and communication tooling, not operational software. It supplements, rather than replaces, the sandbox environment where configuration testing actually happens. This use case is still mostly in the territory of specialized consulting tooling and vendor demonstrations rather than broad deployment. ## Where VR Does Not Improve ERP Operations To be direct about the limits: VR does not make billing faster, improve CIS data quality, or accelerate FI-CA period-end processing in SAP IS-U. It does not improve the accuracy of Oracle CC&B rate schedules or speed up Cayenta-to-SmartWorks integration jobs. The ERP systems that run utility billing, SAP IS-U/FI-CA, Oracle CC&B on OUAF, Cayenta CIS, are transaction processing systems. Their performance depends on data quality, configuration, and process discipline. VR does not touch any of those factors. Claims that VR transforms ERP operations are overstated; the technology lives at the training and design layer, not the operational layer. ## Connecting VR Completions to the ERP Record For both design review sessions and training simulations, tracking participation and outcomes matters. VR platform completions should connect to the project record (in SAP PS or Oracle Project module) or to the training record (in SAP HCM/SuccessFactors or Oracle HCM). Without this integration, the VR work is invisible to the compliance and project tracking systems that utilities rely on for audits and project governance. This integration is achievable but requires deliberate configuration during deployment. Most VR platforms support xAPI or webhook exports that can feed an LMS or project management system. ## A Realistic Assessment VR for design review on capital infrastructure projects has a clear, bounded value proposition and is worth considering for projects where CAD walkthroughs already happen but miss spatial conflicts. VR for ERP process visualization is an emerging communication tool during large implementations. Neither replaces the ERP, the CIS, or the ADMS as an operational system. For the AR field operations side of this picture, see AR for utility asset management and work orders. For the SAP-specific AR integration angle, the SAP augmented reality hub covers that in more depth. ## Frequently asked questions How does VR for process simulation differ from VR for crew training? Crew training VR focuses on physical field procedures, switching, emergency response, confined space. Process simulation VR focuses on ERP workflows, infrastructure design walkthroughs, and operations planning scenarios. Both have value; they address different problems. Can VR replace a sandbox ERP environment for SAP IS-U or Oracle CC&B training? Not directly. ERP workflow training requires hands-on interaction with the actual software. VR can provide context, a virtual walkaround of a billing process flow, but sandbox system access remains the standard for CIS/ERP skills training. What capital project design review scenarios suit VR best? Substation layout and cable routing review, gas regulator station design, and water treatment plant layout are well-suited. The value is in identifying conflicts and clearance issues before construction, saving change-order costs. Is VR for ERP simulation widely deployed in utilities? Not widely. Crew training VR is more common. ERP process simulation in VR is still mostly in pilot or specialized use at larger utilities and their implementation partners. --- ## Utility Data Management with AI: MDM, Quality, and Integration URL: https://www.utilitieslabs.com/maximizing-utility-data-management-with-ai/ # Utility Data Management with AI: MDM, Quality, and Integration AvanSaber Research/March 2, 2024/Updated June 2, 2026/3 min read Every AI application in the utility sector depends on the same thing: data that is accurate, complete, and consistently structured. The systems that produce and manage that data are the meter data management (MDM) platform, the CIS, and the ERP asset register. When those systems have quality problems, AI applications amplify the problems rather than fix them. ## The Role of MDM in the Utility Data Stack Meter data management sits between the AMI network and the billing system. Raw interval reads from Itron or Landis+Gyr meters arrive at the MDM platform, where they go through validation, estimation, and editing (VEE) before being passed to the CIS as billing-ready consumption figures. Oracle Meter Data Management, built on the Oracle Utilities Application Framework (OUAF), handles this workflow for large and mid-size electric and gas utilities. SmartWorks, a Harris MDM product that Cayenta offers as an add-on, serves a similar function for utilities running Cayenta CIS, providing VEE, interval data storage, and data feeds to billing and outage management. AI adds value at the MDM layer in two ways. First, anomaly detection models can identify meters whose read patterns are statistically inconsistent with their account type, history, or neighboring meters, flagging likely meter errors, tampering, or installation problems that VEE rules alone may not catch. Second, imputation models can produce better missing-read estimates than rule-based estimation by learning the typical consumption profile of each premise type and applying that knowledge to the specific gap being filled. Neither AI function replaces the MDM VEE rules or the regulatory requirements for how estimated reads are handled. The AI layer surfaces candidates for review; MDM staff and billing analysts confirm and approve corrections. ## Data Quality Problems That AI Can Surface Several common data quality problems in utility environments are detectable through pattern analysis: Meter multiplier errors occur when a meter is billed at an incorrect demand multiplier. Comparing billed consumption against the modeled consumption for the account type and connected load can flag accounts where the multiplier is likely wrong. Clock drift and interval gaps in AMI communication can produce reads that are offset in time, leading to incorrect time-of-use calculations. Temporal alignment checks across the meter population can catch systematic drift before it affects billing. Account-to-meter mapping errors, where a meter is linked to the wrong service point in the CIS, produce consumption anomalies that are hard to spot in individual account review but visible in population-level analysis. Transformer loading imbalance detected through aggregated interval reads can identify data problems at the account level: if a transformer’s measured output does not reconcile with the sum of metered consumption on that transformer, there are either unmetered accounts, mapping errors, or meter failures in that segment. ## Integration with the Broader CIS and ERP Clean MDM data is the prerequisite for accurate billing in the CIS. It is also the prerequisite for useful demand forecasting, customer energy analytics from tools like Oracle Opower and Bidgely, and asset health monitoring in predictive maintenance programs. The data quality work is not glamorous, but it determines whether the AI applications that utilities invest in produce reliable outputs or unreliable ones. Utilities that have done the work to understand their MDM data quality issues before deploying analytics tend to reach useful insights faster. For Oracle Utilities shops, the OUAF integration model means MDM, CIS, and network management share a common framework, which simplifies data governance. For utilities on Cayenta CIS, the SmartWorks MDM integration provides a tighter coupling between interval data and the billing workflow. The broader context for AI applications across utility operations is covered in the impact of AI and machine learning on the utilities industry. For demand-side analytics and forecasting, see predictive analytics for efficient utility management. If you are assessing your MDM data quality or planning an analytics integration, the AvanSaber team works on CIS and MDM integration projects across Oracle and Cayenta environments. ## Frequently asked questions What is meter data management (MDM) and why does data quality matter for billing? MDM is the system that receives raw interval reads from AMI meters, validates and estimates missing reads, and produces the clean billing-ready consumption figures that the CIS uses to generate customer invoices. Poor MDM data quality leads to estimated bills, rebilling, and customer complaints, and it undermines any AI application that relies on consumption data. Can AI replace MDM validation rules? Not directly. MDM validation rules encode regulatory requirements and utility business rules that AI cannot override. AI augments validation by detecting patterns that standard rules miss, for example, a meter that is consistently passing individual read validations but whose aggregate profile is statistically inconsistent with the premise type. How do Oracle MDM and SmartWorks differ in their AI integration approach? Oracle Meter Data Management (OUAF-based) integrates with the broader Oracle Utilities Application Framework, and Oracle offers analytics add-ons within that ecosystem. Cayenta SmartWorks is a separate MDM product from Harris Computer that integrates with Cayenta CIS, offering tighter coupling between MDM and billing workflows for mid-size utilities. --- ## From Meter-Centric to Service-to-Cash: AI and Utility CIS Architecture URL: https://www.utilitieslabs.com/meter-centric-to-service-to-cash-ai-utility-cis-architecture/ # From Meter-Centric to Service-to-Cash: AI and Utility CIS Architecture AvanSaber Research/April 9, 2026/Updated June 2, 2026/5 min read For most of the twentieth century, a utility customer information system had one organizing principle: read the meter, rate the consumption, print the bill. The system sat at the center of a simple transaction. That simplicity was not a design flaw; it matched the business. A residential customer received one commodity, one bill, once a month. That model is cracking under the weight of distributed energy resources, electric vehicles, demand-response programs, community solar subscriptions, and dynamic tariffs. The CIS that was built to handle a meter-read-and-bill loop was never designed to manage a prosumer who exports solar in the morning, charges a vehicle overnight, and participates in a grid-flexibility event on Tuesday afternoon. ## What the Old Architecture Assumed Legacy CIS platforms were designed around three assumptions: the utility owns the generation, the flow of energy is one direction, and the customer relationship is a monthly billing event. SAP IS-U, Oracle Customer Care and Billing, and most of the platforms built before 2015 share this DNA. The meter is the anchor. Everything upstream of it belongs to operations, and everything downstream is finance. That design produced stable, auditable billing systems. It also produced systems with a very narrow surface area. New product types require new rate tables, new rate tables require development cycles, and development cycles slow down the business. When a utility wants to launch an EV charging incentive or a time-of-use pilot, the CIS is often the constraint. ## What Changed and Why 2025 Is the Inflection Point Two forces accelerated simultaneously. On the grid side, distributed energy resource penetration crossed the threshold where utilities could no longer treat it as an edge case. On the software side, cloud-native platforms and AI inference at scale became commercially available to companies outside the hyperscaler tier. Gartner formalized the market response in its Market Guide for Utility Customer Information Systems, published July 14, 2025 (document G00827595, also summarized at openintl.com). The guide named 19 representative vendors and framed modern CIS as going “beyond billing,” with composable architecture, real-time event-driven capabilities, and DER and EV integration as the defining characteristics of competitive platforms. Gartner also maintains a Utilities Customer Systems Peer Insights market where utility practitioners rate platforms on these dimensions directly. The “composable” framing matters because it describes a shift in how vendors structure their products. Rather than a monolithic billing engine that owns every adjacent function, a composable CIS exposes APIs that allow utilities to add, replace, or extend components without touching the core. That is architecturally significant because it allows the billing foundation to remain stable while the product and engagement layer changes at the pace the business demands. ## The Service-to-Cash Framing The term “service-to-cash” has gained traction among practitioners and vendors as a way of naming what the new architecture must cover. Methodia, which publishes extensively on CIS modernization, uses the term explicitly to contrast the older meter-to-cash model with a broader flow: from service enrollment through usage tracking, pricing, billing, collections, and customer engagement across every product type the utility offers, not just consumption billing. The distinction is not just semantic. A service-to-cash architecture has to support subscriptions, demand-response agreements, EV session fees, and DER compensation in the same system that handles a traditional residential bill. That requires a product catalog that is not hardcoded to commodity rate schedules, and it requires an event-driven processing model that can respond to grid signals in near real time rather than in a nightly batch window. ## Cloud-Native Entrants and the AI Layer The most watched new entrant in the CIS space is Kraken Technologies, which was spun out of Octopus Energy in December 2025 at an approximately $8.65 billion valuation and manages more than 70 million accounts globally. Kraken was designed from the start as a cloud-native, AI-native platform, which means the AI layer is not a bolt-on module sitting above a legacy core; it is woven into routing, scheduling, tariff recommendation, and customer interaction throughout the system. Kraken’s emergence matters not because every utility should move to it, but because it demonstrates that the architectural assumptions of legacy CIS are not laws of physics. A platform can be built differently, and buyers now have a concrete example of what that looks like at scale. The market estimates for CIS software vary meaningfully depending on scope. SNS Insider estimates the CIS market at roughly $1.26 billion in 2023 growing to approximately $3.54 billion by 2032 at a 12.23% CAGR. MarketsandMarkets puts the figure at approximately $1.79 billion in 2024 reaching $3.26 billion by 2030 at 12.8%. If you use a broader “utility billing software” scope, Mordor Intelligence arrives at roughly $6.23 billion in 2025, with cloud deployments at approximately 58% of installations. These estimates are not directly comparable; they reflect different definitions of the market boundary. The consistent signal across all three is sustained double-digit growth driven by modernization investment. ## What It Means for Buyers The architecture question for a utility in 2026 is not simply “which CIS vendor,” but “what capabilities do we need the CIS to own natively versus source through integration.” A composable, event-driven platform allows a utility to adopt best-of-breed DER management or EV charging settlement without replacing the billing core. A monolithic platform forces a tradeoff between stability and agility that compounds over time. Evaluating CIS platforms against the service-to-cash model, rather than the meter-centric model, surfaces the right questions early: can the product catalog handle non-commodity products, can the system process sub-hourly interval data natively, and can the customer engagement layer respond to real-time grid events without a batch delay. For a grounded starting point, the comprehensive guide to choosing utilities software lays out an evaluation framework applicable to this generation of platforms. If you are assessing SAP’s forward path specifically, the SAP Utilities overview and the Oracle vs SAP comparative analysis provide vendor-specific context. And for a broader view of what modern platforms are doing across the industry, modern software solutions transforming the utilities industry covers the wider landscape. The meter-centric CIS served utilities well for decades. The service-to-cash architecture is not a replacement for the billing function; billing still has to be accurate, auditable, and on time. It is a broader frame that treats billing as one output of a system that manages the full customer and asset relationship across every product the utility offers. Buyers who frame their evaluations that way will ask better questions and make better decisions. ## Frequently asked questions What does 'meter-centric' mean in a utility CIS context? A meter-centric CIS was designed primarily to read meters, rate consumption, and generate a monthly bill. The meter-to-cash transaction was the organizing principle. Modern utilities need the system to handle distributed energy resources, EV charging sessions, demand-response events, and flexible tariff products, none of which fit cleanly into a meter-read-and-bill loop. What does Gartner mean by 'composable' CIS architecture? In its July 2025 Market Guide for Utility Customer Information Systems, Gartner frames modern CIS platforms as composable, meaning they expose functionality through APIs that allow utilities to add or replace components (billing, engagement, MDM, DER management) without ripping out the entire core. That replaceability is increasingly important as the pace of product and tariff change accelerates. Is Kraken a replacement for SAP or Oracle in large utilities? Not automatically. Kraken Technologies, spun out of Octopus Energy in December 2025 at roughly an $8.65 billion valuation, serves 70 million accounts and is cloud-native and AI-first by design. Whether it fits a given utility depends on complexity, regulatory context, and how deeply the organization is already invested in a broader SAP or Oracle estate. Vendor-neutral evaluation is essential. How do I know if my current CIS can handle DER and EV integration? The fastest diagnostic is to ask your vendor three questions: does the system model DER assets at the service-agreement level natively, can it ingest sub-hourly interval data from EV charging sessions without a batch import, and does it support time-of-use and dynamic pricing without custom code. If any answer is 'only with a third-party add-on,' the architecture is still meter-centric at its core. --- ## Modern Utility Software, Mapped: CIS, ERP, MDM, OMS, and GIS URL: https://www.utilitieslabs.com/modern-software-solutions-transforming-the-utilities-industry/ # Modern Utility Software, Mapped: CIS, ERP, MDM, OMS, and GIS AvanSaber Research/August 23, 2023/Updated September 24, 2026/1 min read “Modern software solutions for utilities” only becomes useful when you map it by layer. A utility runs a stack, and each layer has a distinct job, a distinct vendor set, and a distinct integration into the system of record. ## The core: CIS and billing This is the system of record for customers, contracts, rates, and bills. The three serious options are SAP IS-U / S/4HANA Utilities, Oracle Customer Care and Billing (CC&B), and Cayenta CIS for the municipal and mid-market tier. Choosing among them is the highest-stakes decision in the stack; the Oracle vs SAP comparison and the Cayenta CIS review cover it. ## Meter data management (MDM) MDM validates and stores interval and AMI meter reads before they reach billing. Cayenta offers SmartWorks MDM as an add-on from a sister Harris product; SAP and Oracle offer their own MDM, and Itron and Landis+Gyr provide head-end and meter-data layers. Clean MDM is what keeps high-volume billing accurate. ## Outage management (OMS) OMS turns meter and call data into a picture of who is out and why. Oracle Utilities Network Management, DataVoice OMS (a sister Harris product Cayenta offers as an add-on), and the ADMS platforms from GE Vernova and Schneider Electric operate here. ## GIS and the spatial layer Utilities are spatial businesses: poles, wires, pipes, and service points live on a map. Esri ArcGIS dominates utility GIS, feeding asset locations into OMS, work management, and increasingly into the CIS for service-point accuracy. ## How the layers fit The pattern that works is a strong CIS as the system of record, with MDM, OMS, and GIS integrated around it through standard interfaces (in SAP environments, often via IDocs, covered in the SAP IDocs guide). Buying a “modern platform” without deciding the CIS first is the common, expensive mistake. Start from the core and integrate outward. ## Frequently asked questions What software does a utility run? At the core, a customer information system and billing platform (SAP IS-U, Oracle CC&B, or Cayenta CIS). Around it: meter data management (MDM) for interval and AMI reads, outage management (OMS), geographic information systems (GIS) like Esri ArcGIS, and increasingly DERMS for distributed energy. The CIS is the system of record the rest integrate with. --- ## Deregulated-Market ERP for Utilities: SAP IS-U and CC&B URL: https://www.utilitieslabs.com/navigating-the-deregulated-market-how-sap-utilities-can-thrive-in-a-changing-landscape/ # Deregulated-Market ERP for Utilities: SAP IS-U and CC&B AvanSaber Research/December 23, 2023/Updated June 1, 2026/1 min read Deregulation splits the company that sells energy from the company that owns the wires, and it demands speed: new rate plans, clean customer switching, and constant message exchange with the market. This guide covers what that requires of a utility ERP and CIS, and how the major platforms cope. ## What deregulation changes - Retail and distribution split. The CIS has to model the retailer role, the distribution role, or both, with the right data flowing between them. - Market messaging. Suppliers and the grid operator exchange standardized messages for enrollments, switches, and meter data. The CIS has to send and receive them reliably. - Rate agility. Competitive retail means new plans launch often. Slow rate configuration is a direct commercial disadvantage. - Clean switching. Customers move between suppliers. The CIS has to enroll and drop them without billing gaps or double-billing. ## How the platforms handle it SAP IS-U and S/4HANA Utilities support deregulated models with region-specific market communication add-ons. Oracle CC&B does the same through its market-message and rate frameworks. Cayenta CIS covers the common retail models well, and the question for a smaller retailer is usually whether the market-message depth matches the local regime, for example ERCOT in Texas. For the platform-level comparison, see Oracle vs SAP for utilities. ## The takeaway Deregulation rewards rate agility and reliable market messaging. Any of the major platforms can be configured for it, so the selection turns on how well a given system handles your region’s market rules and how fast your team can launch new rate plans. ## Frequently asked questions What does a deregulated market require of a utility CIS? Retail-distribution unbundling, market-message exchange with the grid operator and other suppliers, fast time-to-market for new rate plans, and clean customer switching between suppliers. A CIS built only for a regulated monopoly often struggles with these. Can SAP IS-U handle deregulated markets? Yes. SAP IS-U and S/4HANA Utilities support deregulated-market models, including retail and distribution roles and market communication, with configuration and region-specific add-ons. Oracle CC&B supports them similarly. The effort is in the market-message handling for your region. --- ## AI for Energy Consumption Analysis in Utilities: What Works URL: https://www.utilitieslabs.com/optimizing-energy-consumption-ai-analysis/ # AI for Energy Consumption Analysis in Utilities: What Works AvanSaber Research/July 21, 2024/Updated June 2, 2026/1 min read AI for energy-consumption analysis is one of the few utility AI areas with a clear data source and a clear payoff. The data is the smart-meter (AMI) interval reads a utility already collects, and the payoffs are lower call volume, better demand programs, and earlier detection of losses. The work is turning meter data into insight, not buying a generic AI platform. ## Load disaggregation Disaggregation splits a household’s total consumption into appliance-level categories (heating, cooling, water heating, EV charging) from the meter signal alone. Bidgely is the established name here. It powers personalized efficiency advice and helps target demand-response and electrification programs. ## Behavioral analytics Comparing a household against similar ones and reporting the result changes behavior. Oracle Opower built behavioral energy reports into a standard utility program, and the same comparisons feed app and portal insights. ## Loss and anomaly detection The same interval data flags non-technical losses (theft, tampering) and faulty meters. The AMI head-end analytics from Itron and Landis+Gyr, and the meter data management layer of the CIS, are where this runs. It protects revenue and prioritizes field visits. ## Demand response and load shaping Forecasting load and predicting which customers will respond lets a utility shape demand at peak, which matters more every year as electrification and distributed energy grow. This connects to the grid-side tools covered in smart grid optimization. ## Where it sits All of this reads from the meter data the CIS and MDM hold; none of it replaces the system of record. For how AI attaches to the core billing and operations stack without taking it over, see AI for utility operations on SAP IS-U and Oracle CC&B. ## Frequently asked questions How does AI analyze energy consumption? It runs on smart-meter (AMI) interval data already stored in the meter data management layer. Common techniques are load disaggregation (splitting a home's usage by appliance), behavioral analytics (comparing similar households), and anomaly detection for losses and faulty meters. Vendors include Bidgely, Oracle Opower, and the analytics in Itron and Landis+Gyr head-ends. --- ## AI for Energy Distribution: Cutting Losses and Balancing the Grid URL: https://www.utilitieslabs.com/optimizing-energy-distribution-with-ai-powered-utilities/ # AI for Energy Distribution: Cutting Losses and Balancing the Grid AvanSaber Research/November 5, 2024/Updated June 2, 2026/1 min read Energy distribution is where efficiency is won or lost on the grid: voltage held within limits, losses minimized, load balanced, and distributed energy absorbed safely. AI improves each of these, but it improves the decisions; the distribution management platform still executes them. Here is the practical map. ## Volt/VAR optimization Holding voltage within limits while minimizing it (conservation voltage reduction) cuts energy use and losses. AI sharpens Volt/VAR optimization by forecasting load and voltage across the feeder, a function delivered through the ADMS platforms (GE Vernova, Schneider Electric EcoStruxure, Oracle Utilities Network Management). ## Loss reduction Distribution loses energy technically (impedance) and non-technically (theft, faulty meters). AI on AMI interval data and feeder telemetry locates anomalies and prioritizes where to investigate, protecting revenue and efficiency. This ties to the meter-data layer covered in AI for energy consumption analysis. ## Load balancing and hosting capacity Rooftop solar, batteries, and EV charging make feeder loads volatile. AI forecasts net load and helps balance it across phases and feeders, and informs how much distributed energy a feeder can host safely. Coordinating those resources is the DERMS job, covered in smart grid optimization. ## Keep the platform in control The pattern holds: the ADMS and DERMS execute and protect the grid, while AI improves the forecasts and recommendations feeding them. AI does not autonomously run the distribution grid, and for safety reasons should not. For the broader operations view, see AI for utility operations. ## Frequently asked questions How does AI optimize energy distribution? It runs on top of the distribution control platforms. AI improves Volt/VAR optimization (holding voltage while cutting losses), load balancing across feeders, technical and non-technical loss detection, and DER hosting-capacity decisions. The ADMS executes; AI improves the decisions feeding it. What is the difference between this and smart grid optimization? Smart grid optimization is the broader platform view (ADMS, DERMS, the control room). This focuses on the distribution-efficiency outcomes specifically: voltage, losses, load balancing, and hosting capacity. --- ## SAP Utilities Finance: FI-CA, Billing, and the Utility Revenue Process URL: https://www.utilitieslabs.com/optimizing-financial-processes-in-the-utilities-sector-with-sap-utilities-finance/ # SAP Utilities Finance: FI-CA, Billing, and the Utility Revenue Process AvanSaber Research/June 13, 2023/Updated June 1, 2026/1 min read The finance side of a utility is not generic accounts receivable. Millions of small, recurring invoices, dunning at scale, and collections that have to respect regulation are a different problem, and that is what SAP Utilities Finance is built for. This guide explains FI-CA and how it ties into the meter-to-cash process. ## FI-CA: the engine under utility finance FI-CA (Contract Accounts Receivable and Payable) is the SAP subledger for high-volume industries. In a utility it manages: - Contract accounts. One account per customer contract, at a volume normal accounts receivable cannot handle. - Payments. High-volume payment processing and allocation. - Dunning and collections. Scheduled reminders and collection steps, with the controls regulators expect. ## How it ties into meter-to-cash The revenue process runs end to end: SAP IS-U reads the meter and produces the bill, the receivable posts to FI-CA, and payment, dunning, and collections run in the finance layer. Because billing and finance are integrated, the utility keeps one consistent view of what each customer owes. ## Where this fits SAP Utilities Finance is part of the wider SAP IS-U picture. For the platform overview, see the SAP IS-U guide, and for how SAP compares to Oracle on billing and finance, the Oracle vs SAP comparison. ## Frequently asked questions What is FI-CA in SAP utilities? FI-CA is Contract Accounts Receivable and Payable, the SAP subledger built for high-volume industries like utilities. It manages the large number of customer contract accounts, payments, dunning, and collections that standard accounts receivable cannot handle at utility scale. How does SAP Utilities Finance connect to billing? SAP IS-U produces the bill, and the receivable posts to FI-CA as a contract-account entry. Payments, dunning, and collections then run in FI-CA. The finance layer and the billing layer are tightly integrated so the revenue process stays consistent end to end. --- ## Oracle CC&B Data Model: How Utility Billing Resolves URL: https://www.utilitieslabs.com/oracle-ccb-data-model-for-utilities/ # Oracle CC&B Data Model: How Utility Billing Resolves AvanSaber Research/August 11, 2026/4 min read Most Oracle Customer Care and Billing problems that look like configuration problems are actually data model problems. A rate that will not apply, a bill that splits when it should not, a move-in that leaves stranded consumption: trace each one back and it usually ends at an entity that was modelled wrong before anyone touched a rate schedule. This is the core chain, what each entity is really for, and where implementations tend to go wrong. ## Key takeaways - The chain is person, account, premise, service point, service agreement. Billing resolves along it rather than against a single customer record. - Person and account are separate on purpose. The party and the payer are not always the same, and accounts change hands while premises do not. - Premise and service point are separate on purpose. One location can take several commodities, each with its own metering and contract. - The service agreement is where the rate lives. It is the contract layer that connects a payer to a delivery point under specific terms. - Data model errors are the expensive kind, because every rate, bill, and report downstream inherits them. ## The five entities Person. The party: an individual, a business, sometimes a government body. A person holds identifying details and can be related to several accounts. Keeping identity separate from billing is what allows one business to hold many accounts without duplicating the party record on each. Account. What gets billed. An account carries the financial relationship: balance, payment history, credit and collections status, billing cycle. Several people can be related to one account with different roles, which is how joint occupancy and business signatories are represented. Premise. The physical location, usually an address. Premises are durable in a way accounts are not. Occupants move in and out; the building stays. Modelling the premise separately is what makes a change of occupant a change to the account and the service agreement rather than to the infrastructure record. Service point. Where a specific commodity is delivered at a premise. This is the entity people underestimate. A single premise can hold an electric service point and a water service point, each metered separately, each with its own agreement. A combination utility that collapsed this into the premise would be unable to bill the two commodities on their own terms. Service agreement. The contract that ties an account to a service point. It carries the rate schedule, the effective dates, and the billing terms. It is the entity that answers “who is paying, for what delivery, at what rate, over what period.” ## How a bill actually resolves The chain matters because billing walks it rather than reading a single record. - The billing run starts from the account. - It finds the service agreements attached to that account. - Each agreement points at a service point, where consumption has been recorded. - The rate schedule on the agreement is applied to that consumption. - The results assemble into one bill for the account, covering every service. Which is why a combination utility can put electric and water on a single bill without merging the two services: the separation lives at the service point and the agreement, and the merge happens at the account. ## Where implementations go wrong Premises that are really service points. The most common structural error. Somebody models each metered service as its own premise, and now a single building appears as several addresses. Move-ins become inconsistent, service orders route oddly, and any report grouped by location double-counts. Accounts used as parties. Where the party record is skipped and everything hangs off the account, a customer with three accounts becomes three customers. Collections treat them separately, service history fragments, and nobody can answer how much this business owes in total. Service agreements without clean effective dating. A rate change or a move-out that is not dated correctly leaves consumption stranded between agreements. It usually surfaces as an unbillable period that nobody notices until a month-end reconciliation refuses to close. Migration that preserves the errors. A conversion from a legacy system that maps old records straight across carries every one of these problems into the new platform, and now they are expensive. Data quality assessment belongs before configuration, not next to it. ## Coming from SAP IS-U The concepts map, the names do not. Anyone moving between the two platforms should expect the ideas to transfer and the specifics not to. | Oracle CC&B | SAP IS-U | What it is | Person | Business partner | The party | Account | Contract account | The billing relationship | Premise | Connection object | The physical location | Service point | Point of delivery | Where a commodity is delivered | Service agreement | Contract | The terms binding payer to delivery The boundaries differ in detail, so a mapping exercise between the two is a real piece of work rather than a rename. Our SAP CIS overview covers the IS-U side, and the SAP, Oracle, and Cayenta comparison sets the platforms side by side. ## The verdict The CC&B data model is not complicated, but it is precise, and the precision is load-bearing. Five entities, each answering one question: who is the party, who pays, where is it, what is delivered there, and under what terms. Implementations that get those five right spend their configuration budget on rates. Implementations that do not spend it on rework. If you want a second read on a data model before configuration starts, the AvanSaber team does this work on utility CIS programmes case by case. ## Frequently asked questions What is the core entity chain in the CC&B data model? Person, account, premise, service point, and service agreement. A person is the party, an account is what gets billed, a premise is the physical location, a service point is where a specific commodity is delivered at that premise, and a service agreement is the contract that links an account to a service point. Billing resolves along that chain rather than against a single customer record. Why does CC&B separate person from account? Because the party and the payer are not always the same, and neither is stable over time. A landlord may hold the account for a premise a tenant occupies, a business may have one person of record and many accounts, and an account may change hands while the premise and service point stay put. Keeping them separate means a change of occupant does not disturb the physical infrastructure records. What is the difference between a premise and a service point? A premise is the location, typically an address. A service point is where a particular commodity is delivered at that location. One premise can hold several service points, for example electric and water, each with its own metering and its own service agreement. Modelling them separately is what lets a combination utility bill several commodities to one account without duplicating the address. How does a service agreement relate to billing? The service agreement is the contract layer. It ties an account to a service point, carries the rate schedule that applies, and defines the billing period. When a bill runs, the system walks from account to its service agreements, reads the consumption recorded at each linked service point, applies the rate on the agreement, and assembles the result into one bill for the account. Why does the data model matter for an implementation? Because most CC&B implementation overruns are data problems wearing a configuration costume. If premises and service points are wrong in the source system, every downstream rate, bill, and report inherits that error, and the fix costs more the later it is found. Data model alignment is worth doing before configuration, not alongside it. Is the CC&B data model similar to SAP IS-U? The concepts map closely, since both model a utility's meter-to-cash chain, but the names and the boundaries differ. SAP IS-U uses business partner, contract account, connection object, point of delivery, and contract for roughly the same roles. Anyone moving between the two should expect the ideas to transfer and the specifics not to. --- ## Oracle Opower and Energy & Water Intelligence: Utility AI in 2026 URL: https://www.utilitieslabs.com/oracle-opower-energy-water-data-intelligence-ai-utilities/ # Oracle Opower and Energy & Water Intelligence: Utility AI in 2026 AvanSaber Research/April 21, 2026/Updated June 2, 2026/4 min read Oracle has been building a visible AI story in utilities across two fronts: Opower, the long-running behavioral science and engagement platform, and newer infrastructure like Energy and Water Data Intelligence that aims to unify data across the Oracle Utilities portfolio. A press release dated April 13, 2026 gave the clearest public tally yet of Opower’s cumulative scale. This review looks at what Oracle is actually shipping, what is still on the roadmap, and where independent caution is warranted. ## Opower at scale: the Oracle-reported figures Oracle reported in April 2026 that Opower now benefits “nearly 45 million” North American households, with the precise enrolled figure given as 44.6 million residential households cumulative since 2009 through March 2026. Over that period, Oracle reported 3.5 billion personalized communications delivered across channels, with cumulative energy savings of 44.23 TWh and nearly $4.3 billion in cumulative bill savings. Of that bill savings total, Oracle attributed $369 million to calendar year 2025. These are Oracle-reported figures, not independently audited. Measurement and verification methodologies for behavioral energy programs vary across utilities and regulators, and attribution is difficult to standardize at scale. Utilities filing with an IRP or DSM proceeding should request program-specific measurement and verification documentation, not portfolio-level totals. ## What the AI and behavioral science actually do Opower’s core engine compares a household’s consumption against a peer group of similar homes, then produces a personalized Home Energy Report. The framing draws on behavioral science research around social norms: showing customers that comparable neighbors use less energy consistently produces modest but measurable reductions in a way that generic efficiency appeals do not. The AI layer sits on top of usage data from the utility’s meter data management system. It segments customers, selects which communications to send (print, email, in-app, digital), calibrates timing, and over time adjusts the program based on response. The system is not replacing the CIS as a system of record. It is consuming usage and account data from the CIS and MDM, running its models, and pushing outbound communications. That architecture matters: Opower adds engagement capability to an existing CIS, it does not substitute for one. For a broader view of what Oracle’s CIS portfolio looks like as a foundation for tools like this, see the Oracle Utilities overview and the Oracle vs SAP comparison. ## Energy and Water Data Intelligence: what is live vs. what is roadmap Oracle announced Energy and Water Data Intelligence on September 11, 2024 as a data unification and AI acceleration layer. At launch, it was pre-integrated with Oracle Utilities Customer Cloud Service and Customer To Meter NOW. The announcement also noted planned integrations with Work and Asset Cloud Service (WACS), DERMS, and Opower as roadmap items. Those integrations were not described as shipped at launch. Buyers evaluating Energy and Water Data Intelligence should verify current integration status directly with Oracle rather than treating all four integrations as generally available. The intent is practical: utilities on Oracle Utilities products accumulate usage data, meter events, work orders, and customer records across systems that have historically required bespoke integrations to join together. Energy and Water Data Intelligence is Oracle’s answer to that fragmentation, providing a pre-built data layer so AI models have a unified input. ## AI in the contact center and in meter data management Two additional Oracle AI announcements from 2025 are worth noting for utilities on the Oracle Utilities Customer Platform. In May 2025, Oracle added AI call summarization and tagging to Oracle Utilities Customer Platform at no extra cost. After each customer service call, the AI produces a structured summary and applies tags, reducing the time agents spend on after-call work and creating a more consistent record for quality review. This is an internal-facing feature, not a customer-facing one. In June 2025, Oracle announced AI anomaly detection inside Meter Data Management targeted at cutting billing exceptions and truck rolls. The proposition is straightforward: meter reads that fall outside normal patterns get flagged automatically before billing runs, reducing the manual review workload on the CIS billing team and avoiding estimated bills that generate customer calls. Taken together, these two capabilities address two well-known cost drivers in utility customer operations: after-call handle time in the contact center, and exception-driven billing work in MDM. Neither requires a new CIS implementation; both are incremental additions to an existing Oracle Utilities footprint. ## An honest independent view Oracle’s Opower footprint is real and large. The cumulative household counts, TWh saved, and bill savings figures in the April 2026 press release represent a genuine track record across dozens of North American utilities over 17 years. The caveat stands: these are Oracle-reported, program-level aggregates, and the underlying measurement and verification methodology is not standardized across all utility programs in the portfolio. For utilities already on Oracle Utilities products, the newer AI additions (Energy and Water Data Intelligence, call summarization, MDM anomaly detection) are incremental expansions of capability on an existing platform rather than architectural shifts. They lower the integration cost of adopting AI because the data plumbing is already in place. For utilities not yet on Oracle, these AI capabilities are not a standalone reason to choose Oracle over alternatives. The CIS selection decision belongs at a different layer, as covered in the full Oracle Utilities platform review and the side-by-side SAP and Oracle comparison. The AI layer is most valuable when the foundational CIS data is clean, current, and well-integrated with MDM. ## Frequently asked questions What is Oracle Opower and how does it work for utilities? Oracle Opower is a utility customer engagement product that combines AI and behavioral science to deliver personalized energy reports, home energy analysis, and targeted efficiency programs. It integrates with Oracle Utilities CIS products and uses customer usage data to produce individualized communications designed to reduce consumption and improve satisfaction. What is Oracle Energy and Water Data Intelligence? Oracle Energy and Water Data Intelligence, launched in September 2024, is a data unification and AI acceleration layer pre-integrated with Oracle Utilities Customer Cloud Service and Customer To Meter NOW. Integrations with Work and Asset Cloud Service, DERMS, and Opower were on the announced roadmap as of launch and had not all shipped. Are Oracle Opower energy savings figures independently verified? No. The figures Oracle publishes for Opower, including households enrolled, energy saved, and bill savings, are Oracle-reported and self-audited. They have not been independently audited by a third party. Utilities evaluating the product should request program-specific measurement and verification documentation. How does Oracle AI in Customer Platform differ from Opower? They serve different workflows. Opower is a customer engagement and demand-side management product, delivering reports and behavioral nudges outbound. The Oracle Utilities Customer Platform AI features announced in May 2025 are internal-facing: call summarization and tagging that help contact-center agents after each call, at no added cost. --- ## Oracle vs SAP for Utilities: 2026 ERP Comparison (CC&B, IS-U, S/4HANA) URL: https://www.utilitieslabs.com/oracle-vs-sap-a-comparative-analysis-of-utilities-implementation/ # Oracle vs SAP for Utilities: 2026 ERP Comparison (CC&B, IS-U, S/4HANA) AvanSaber Research/June 11, 2023/Updated June 1, 2026/3 min read Oracle and SAP are the two enterprise heavyweights in utility billing and customer information. The generic “Oracle vs SAP” comparisons miss what utility buyers actually need, which is a comparison by the real products: Oracle Customer Care and Billing (CC&B) against SAP IS-U and S/4HANA Utilities, not the vendors in the abstract. ## The products you are actually choosing between On the SAP side, the utility solution is SAP IS-U on SAP ECC, with S/4HANA Utilities as the forward platform and Cloud for Utilities (C4U) as the newer cloud model. Billing runs through FI-CA contract accounts. On the Oracle side, the customer and billing system is Oracle CC&B, part of the Oracle Utilities application suite, surrounded by Oracle Meter Data Management and Oracle Network Management for MDM and OMS. Naming these is the point. A comparison that says “Oracle” and “SAP” without naming CC&B, IS-U, and S/4HANA Utilities cannot help you make a real decision. ## Feature comparison | Dimension | SAP IS-U / S/4HANA Utilities | Oracle CC&B | Billing core | FI-CA contract accounts | CC&B rating and billing | MDM | SAP MDUS or S/4HANA option | Oracle Meter Data Management | OMS | SAP partner or third-party | Oracle Network Management | Cloud path | S/4HANA Utilities, C4U | Oracle Energy and Water Cloud | Deregulated markets | Deep, configuration-heavy | Deep, configuration-heavy | Best fit | SAP-estate utilities | Oracle-estate utilities | Forward roadmap | IS-U to S/4HANA Utilities | CC&B to cloud ## Cost comparison Neither vendor publishes list pricing, and both land in the enterprise tier. In AvanSaber’s implementation experience, large Oracle utility programs have often run above comparable SAP ones, though the spread is wide and scope-dependent. The honest guidance is to treat that as context, not as your number, and to compare scoped quotes that hold service types, customer count, and integration scope constant across both vendors. ## Implementation reality Both are multi-year programs at scale. The difference that decides most selections is not a feature gap, it is fit with your existing estate and the availability of a capable implementation partner in your region. A utility already running SAP across finance and operations usually finds IS-U or S/4HANA Utilities lowers integration cost. A utility on Oracle finds the reverse. ## Where Cayenta fits For municipal and mid-market utilities, the third option is often the right one. Cayenta CIS delivers an integrated CIS, MDM, OMS, and mobile suite at lower cost and a smaller implementation footprint than either giant. If you are not a large investor-owned utility, read the Cayenta CIS review before you default to SAP or Oracle. ## The verdict There is no universal winner. SAP IS-U and S/4HANA Utilities fit SAP-estate utilities and those wanting one ERP and CIS platform. Oracle CC&B fits Oracle-estate utilities and billing-led selections. Cayenta fits the mid-market. Decide on estate fit, deregulated-market needs, and partner strength, and validate cost against scoped quotes rather than averages. ## Sources and disclosure Product details here come from the vendors’ own documentation: Oracle’s Customer Care and Billing docs and the SAP for Utilities portal. For the mid-market alternative, Cayenta CIS is reviewed on Gartner Peer Insights. Cost observations are from AvanSaber’s implementation experience, not a published price list, so validate them against scoped quotes. Disclosure: UtilitiesLabs is backed by AvanSaber, which operates a SAP utilities consulting practice. That is a commercial relationship that could bias us toward SAP. We manage it by naming it, citing independent sources, and reviewing SAP, Oracle, and Cayenta to the same bar. See our editorial policy. ## Frequently asked questions Is Oracle or SAP better for utilities? Neither wins outright. SAP IS-U and S/4HANA Utilities tend to fit utilities already standardized on SAP and those wanting one ERP and CIS estate. Oracle CC&B tends to fit utilities on an Oracle stack or wanting a billing-focused CIS that integrates with best-of-breed systems. The deciding factors are existing estate, deregulated-market needs, and partner availability. What is the difference between SAP IS-U and S/4HANA Utilities? IS-U is the classic utilities solution on SAP ECC. S/4HANA Utilities carries the IS-U functional scope onto S/4HANA and the HANA database with a simplified data model and Fiori apps. C4U (Cloud for Utilities) is the newer cloud-oriented model alongside it. How do Oracle CC&B and SAP IS-U pricing compare? Both are enterprise-tier and quoted, not list-priced. In AvanSaber's implementation experience, large Oracle utility programs have often run above comparable SAP ones, but the spread depends heavily on scope, service types, and the implementation partner. Validate against scoped quotes rather than industry averages. --- ## SAP Analytics Cloud and BTP AI: BI for Utilities on SAP IS-U URL: https://www.utilitieslabs.com/revolutionizing-business-intelligence-with-sap-ai-and-advanced-data-models/ # SAP Analytics Cloud and BTP AI: BI for Utilities on SAP IS-U AvanSaber Research/September 29, 2023/Updated June 2, 2026/3 min read Utilities running SAP IS-U or S/4HANA Utilities accumulate deep operational data across the full meter-to-cash cycle: meter reads, rate calculations, invoices, payments, and collections all pass through the system. The challenge is turning that transaction history into operational intelligence without burdening the core system or requiring every analyst to write ABAP. SAP Analytics Cloud (SAC), SAP HANA embedded analytics, and BTP AI services are the three layers that address this, and they serve different purposes. ## SAP HANA as the Analytics Foundation SAP HANA’s in-memory architecture allows SAC and other reporting tools to query large IS-U or FI-CA datasets without pre-aggregated cubes. The SAP-delivered Core Data Services (CDS) views for Utilities expose key entities like contract accounts, billing documents, and meter reading orders in a way that is query-optimized and upgrade-stable. For utility billing operations, HANA live queries against FI-CA mean that a collections manager can see yesterday’s dunning results at opening of business rather than waiting for an overnight batch report. The same infrastructure feeds the meter-to-cash performance dashboards that billing operations teams rely on. ## SAP Analytics Cloud: Planning and Storytelling on Top of IS-U Data SAC adds three capabilities that HANA embedded analytics does not natively provide: Collaborative planning. Revenue forecasting models in SAC can draw on IS-U billing history, rate change scenarios, and customer growth projections. Finance teams can run rate impact scenarios and compare them against prior periods without touching production IS-U tables. Predictive scenarios. SAC’s smart predict feature applies time-series forecasting to consumption and revenue data. For a utility planning capital recovery, the ability to model demand growth against rate scenarios in a governed BI environment is more auditable than an analyst-maintained spreadsheet. Storytelling for regulatory reporting. SAC stories combine KPI tiles, charts, and commentary in a format that is reproducible and version-controlled. Regulatory filings that require consistent presentation of billing metrics, collections performance, or demand-side management program results benefit from this structure. ## BTP AI Services: Where the Intelligent Layer Sits SAP Business Technology Platform (BTP) AI services extend into the IS-U and S/4HANA Utilities context in two practical ways. Document and exception classification. BTP Document Information Extraction can process inbound correspondence and route account adjustment requests without manual triage. Volume spikes during rate change events, when customer letters and dispute filings increase sharply, are a well-suited use case. SAP Joule as analyst assistant. Joule is embedded in SAC and S/4HANA interfaces and can answer natural-language questions against the live data model. A billing manager asking “which rate classes have the highest write-off rate this quarter” gets a result without writing a query. The answer is sourced from the same authorized IS-U data the manager would access directly; Joule does not bypass data governance. The key architectural principle is that BTP AI reads and summarizes data from IS-U or FI-CA; it does not write to billing or contract account records autonomously. A human analyst or controlled workflow confirms any adjustment. ## What SAP BI Does Not Replace SAP Analytics Cloud and BTP AI are not substitutes for a well-maintained IS-U data model or clean master data. Reporting on meter-to-cash performance is only as reliable as the business partner, contract account, and rate structure data in the underlying system. Before investing in BI tooling, utilities should assess IS-U master data quality. For the broader ERP analytics picture, the AI analytics on utility ERP/CIS data article covers what applies across platforms. SAP-specific implementation context lives in the SAP IS-U pillar and SAP HANA utilities deep dive. Advisory support on SAP Analytics Cloud implementation or BTP AI integration for IS-U environments is available through Avansaber. ## Frequently asked questions What is the difference between SAP Analytics Cloud and embedded HANA analytics for IS-U? Embedded SAP HANA analytics run directly against the IS-U or S/4HANA data model and are best for transactional KPIs like meter-to-cash cycle metrics. SAP Analytics Cloud (SAC) adds a governed BI layer with planning, predictive, and storytelling capabilities on top of live HANA connections or imported datasets. Where does SAP Joule fit in utility BI workflows? SAP Joule is the generative AI assistant embedded across the S/4HANA and BTP surface. For utility analysts it can interpret natural-language queries against SAC models or summarize FI-CA account exceptions, but all data access goes through standard SAP authorization checks on the underlying IS-U or S/4HANA system. What meter-to-cash KPIs are most commonly tracked in SAP Analytics Cloud for utilities? Common KPIs span the full meter-to-cash cycle: meter read success rates, billing exception volumes, days sales outstanding (DSO) in FI-CA, dunning progression rates, and payment channel mix. These are sourced from IS-U or S/4HANA Utilities tables via HANA virtual data models. --- ## Oracle AI and Data Models for Utility Analytics on CC&B Data URL: https://www.utilitieslabs.com/revolutionizing-data-analysis-with-oracle-ai-and-advanced-data-models/ # Oracle AI and Data Models for Utility Analytics on CC&B Data AvanSaber Research/September 19, 2023/Updated June 2, 2026/3 min read Utility companies sit on large volumes of structured operational data: billing segments, interval meter reads, service agreement history, and financial transactions. Most of that data lives in Oracle Customer Care and Billing (CC&B) or Oracle Utilities Application Framework (OUAF) products. Oracle provides a set of analytics and AI tools specifically designed to work with that data, but the value they deliver depends heavily on the quality of the data model underneath. This post examines what those tools actually are and where they fit in a utility analytics programme. ## Oracle Utilities Analytics Warehouse as the Foundation Raw CC&B tables are normalised for transactional efficiency, not reporting. Oracle Utilities Analytics Warehouse (OUAW) provides a pre-built dimensional model that extracts CC&B, Oracle Meter Data Management, and other OUAF source tables into a star schema. Fact tables cover billing history, payment history, meter reads, and field activity; dimension tables cover premises, accounts, rate classes, and geography. Starting analytics work from OUAW rather than direct CC&B tables avoids the complexity of joining OUAF’s deep normalisation and reduces the risk that a CC&B patch will silently break a report query. For utilities that have not yet deployed OUAW, building an in-house equivalent is feasible but requires sustained investment to keep the ETL current across CC&B patch cycles. ## Oracle Analytics Cloud for Billing and Revenue Reporting Oracle Analytics Cloud (OAC) connects to OUAW or to an Oracle Autonomous Data Warehouse that holds utility data and provides a governed semantic layer for self-service reporting. Finance teams can build revenue-by-rate-class or unbilled-revenue views without writing SQL. Operations teams can track billing exception rates or field activity cycle times. OAC’s prebuilt subject areas for Oracle Utilities cover common billing and customer metrics out of the box. Custom subject areas can extend those definitions for utility-specific rate structures or deregulated market reporting. ## Oracle Machine Learning Inside the Database Oracle Machine Learning (OML) runs machine learning algorithms directly inside Oracle Database, which means models can be built and scored against OUAW or CC&B replica tables without moving data to an external platform. Practical utility use cases include: - Churn and delinquency propensity: Classification models trained on payment history and service agreement attributes to identify accounts at risk before they reach collections. - Consumption anomaly detection: Unsupervised clustering or statistical outlier detection on interval reads to flag meter faults, theft, or unbilled usage before the billing cycle closes. - Rate class optimisation: Regression models that estimate revenue impact of rate redesign scenarios by simulating billing outputs across the customer base. These models require clean, consistent training data. Utilities with incomplete address standardisation, high rates of estimated reads, or manual billing adjustments will need data quality remediation before model outputs are reliable. ## Oracle Opower for Customer-Facing Energy Analytics Oracle Opower is a separate product within the Oracle Utilities suite. It consumes account and consumption data from CC&B or AMI systems and produces personalised home energy reports, behavioural demand response programmes, and customer engagement campaigns. The value proposition is that Opower’s models are pre-trained on broad consumption datasets across its utility customer base, reducing the cold-start problem for individual utilities. The integration between CC&B and Opower is well-defined, but it requires clean account-to-premise linkage in CC&B and reliable interval or monthly read data. Utilities migrating from legacy metering to AMI often find that Opower programme quality improves substantially once the interval data pipeline is stable. ## What AI Cannot Fix in a CC&B Environment Analytics and AI tools do not compensate for upstream data problems. If billing corrections, manual adjustments, or legacy migrations have left gaps in billing history, models trained on that data will reflect those gaps. Before investing in OAC dashboards or OML models, utilities should assess CC&B data completeness, particularly for accounts that were migrated from predecessor systems. Similarly, AI-assisted forecasting is only as good as the rate schedule configuration in CC&B. Incorrect rate component setup produces billing data that looks complete but is structurally wrong from a revenue modelling perspective. --- ## AI Analytics on Utility ERP and CIS Data: A Practical Guide URL: https://www.utilitieslabs.com/revolutionizing-erp-with-ai-driven-data-analytics/ # AI Analytics on Utility ERP and CIS Data: A Practical Guide AvanSaber Research/April 15, 2024/Updated September 24, 2026/3 min read Utility ERP and CIS platforms, whether SAP IS-U, Oracle CC&B, or Cayenta CIS, accumulate years of billing, payment, and customer interaction history. That history contains patterns that are difficult to extract through standard reporting but are well-suited to machine learning: customers likely to miss payment, meters showing read anomalies that precede billing disputes, rate codes applied inconsistently across similar premises. AI-driven analytics surfaces those patterns without requiring changes to the core transactional system. ## The Architecture That Actually Works The reliable pattern for AI analytics on utility ERP data has three components: - Data extraction. The CIS or ERP exposes data through standard interfaces. For SAP IS-U, IDocs and BTP integration services are the primary channels. Oracle CC&B uses its web service layer and database views. Cayenta CIS exposes data through its reporting database. In each case, the production system is not queried directly by the analytics model. - Analytics processing. A separate analytics environment, whether a cloud data warehouse, SAP Analytics Cloud, or an Oracle Utilities analytics module, holds the processed data. Models run here against historical snapshots plus recent transaction data. - Recommendation delivery. Outputs are surfaced back in the CIS interface or a connected portal as work queue items or flags. A billing analyst reviewing a dispute queue sees an AI-generated note indicating similar disputes were resolved by a specific adjustment type. The analyst decides; the CIS records the action. This is the correct model because the CIS remains the system of record for all customer and billing data. An AI recommendation that is not confirmed by a human or a pre-authorized automated rule does not change any account. ## High-Value Use Cases Across CIS Platforms Credit risk scoring on FI-CA / contract accounts (SAP IS-U). Payment behavior, dunning history, and consumption pattern changes can feed a credit score that routes new service requests or deposit waiver decisions. SAP BTP AI services can host the scoring model, with the result written to a custom field the collections team sees in the account view. Billing exception triage (Oracle CC&B, Cayenta CIS). Billing runs produce exception queues that billing analysts work manually. AI classifiers trained on resolved exception history can categorize new exceptions and route them to the correct team, reducing the time from bill run to invoice delivery. Oracle’s OUAF framework supports rule-based routing that can incorporate model outputs. Cayla AI in Cayenta CIS. Harris Computer’s Cayenta CIS includes Cayla, an AI assistant designed for the platform. Cayla can surface account insights and draft responses to customer inquiries, with the CSR confirming before any account action is taken. This is a native integration rather than a third-party overlay. Customer consumption anomaly detection (all platforms). Interval data from Itron or Landis+Gyr AMI meters flows through meter data management into the CIS. AI models can flag accounts where consumption deviates from expected patterns, which may indicate equipment faults, meter errors, or unauthorized use. The CIS work queue carries the flag; the field crew investigates. ## What to Build vs. What to Buy Utilities frequently overestimate the custom build required. SAP Analytics Cloud with BTP AI handles much of the SAP IS-U analytics use case without custom model development. Oracle Utilities Analytics (OUA) does the same for CC&B. On Cayenta CIS, the Cayla AI assistant addresses the customer-facing side. Custom model development makes sense for organization-specific credit risk scoring using local payment behavior data, or for asset failure prediction on non-standard equipment types. General-purpose billing analytics rarely justify a build-from-scratch approach. For the SAP-specific BI layer, the SAP Analytics Cloud and BTP AI article covers the platform details. The Cayenta CIS review explains where Cayenta’s native AI features fit for mid-market utilities. The ERP and AR integration guide covers the accounts receivable side of the data flow. For a vendor-neutral assessment of analytics tooling against your specific CIS platform, Avansaber’s team can help scope the integration architecture. ## Frequently asked questions Can AI analytics tools write back to SAP IS-U or Oracle CC&B? Properly architected integrations keep AI analytics in a read-and-recommend role. The AI layer reads data from the CIS or ERP, surfaces recommendations, and a human operator or a controlled workflow in the CIS executes any account action. Direct AI write-back to production CIS records is not a recommended pattern. What ERP data is most useful for AI analytics in utilities? The highest-value data sets are billing history, payment and collection event timelines, meter read validation outcomes, dunning progression records, and rate structure change logs. Together they describe the full meter-to-cash cycle and expose the patterns AI models learn from. How does AI analytics on the ERP differ from the SAP Analytics Cloud approach? SAP Analytics Cloud is a governed BI platform built on top of the SAP data model. AI analytics can run on any CIS platform. The common thread is that the CIS or ERP is always the system of record; the analytics layer reads from it and returns insights, but does not replace the transactional system. --- ## AI Across the Oracle Utilities Suite: CC&B, Opower, and NMS URL: https://www.utilitieslabs.com/revolutionizing-the-utility-industry-harnessing-the-power-of-oracle-utilities-ai/ # AI Across the Oracle Utilities Suite: CC&B, Opower, and NMS AvanSaber Research/November 4, 2023/Updated June 2, 2026/3 min read The Oracle Utilities product family covers billing and customer information in CC&B, meter data in Oracle Utilities MDM, customer engagement in Oracle Opower, and grid operations in Oracle Network Management System (NMS). AI capabilities enter each of these layers differently, and understanding those differences matters for utilities evaluating where to invest. This post describes where AI is practically applicable today across the Oracle Utilities suite, without inflating what any single product can do out of the box. ## CC&B: AI on the Data, Not in the Application Oracle Customer Care and Billing is a transactional system. It stores service agreements, billing segments, payment history, and meter reads. The application itself does not embed AI; the intelligence comes from applying Oracle Analytics Cloud, Oracle Machine Learning, or third-party tools to the data CC&B generates, typically after extracting it to Oracle Utilities Analytics Warehouse. Practical AI applications on CC&B data include collections risk scoring, billing anomaly detection, and customer lifetime value estimation. Each of these requires a clean, consistent extract from CC&B and enough historical billing cycles to train useful models. Utilities that have recently migrated from a legacy system often need 12 to 24 months of post-migration data before model quality is sufficient for production use. ## Oracle Opower: Behavioural AI for Demand Management Oracle Opower is the clearest example of deployed AI across the Oracle Utilities suite. Opower’s platform uses machine learning to group customers by consumption archetype and then produces home energy reports and demand response offers personalised to each group. Because Opower’s models are trained across multiple utility deployments, a smaller utility can benefit from patterns observed in a larger peer dataset. The key integration point is a reliable feed of account and interval read data from CC&B or the AMI headend. Opower’s programme effectiveness directly tracks data quality: utilities with high rates of estimated reads or incomplete premise-account linkage see lower engagement rates in their Opower programmes. Opower also supports rate analysis tools that use consumption data to project customer bill impacts under alternative rate structures, which feeds directly into rate case preparation and customer communication. ## Oracle Utilities MDM: Interval Data and Anomaly Detection Oracle Utilities Meter Data Management sits between the AMI headend and CC&B. At scale, a utility with smart meters generates interval reads at 15- or 30-minute granularity for hundreds of thousands of accounts. Oracle MDM includes validation, estimation, and editing (VEE) rules that catch implausible reads before they reach billing. Machine learning models can extend these rules by learning normal consumption signatures per premise type and flagging deviations that rule-based VEE misses. Tamper detection and non-technical loss identification benefit from the same approach: a classification model trained on labelled examples of theft or meter fault patterns can triage accounts for field investigation more accurately than threshold-based alerts alone. ## Oracle Network Management System: Grid Operations AI Oracle NMS supports outage management and network switching for electric utilities. AI-assisted features in this layer focus on predictive reliability: using historical outage records, asset age, load patterns, and weather data to score circuit segments by outage probability. Utilities can use these scores to prioritise infrastructure replacement and pre-position crews before storm events. NMS also supports mobile workforce integration, and AI-based route optimisation for field crews is an adjacent capability. The integration with CC&B’s service agreement data allows NMS to cross-reference active service agreements with outage polygons for real-time customer impact counts. ## Where Honest Caution Applies Oracle’s AI capabilities are real, but several conditions limit their impact in practice. First, they require clean foundational data in CC&B and MDM; no AI layer compensates for systematic billing corrections, read gaps, or migration debt. Second, the Oracle Utilities product suite has significant licensing and integration complexity, and adding Opower or advanced analytics requires its own implementation project. Third, utilities on older CC&B releases may find that some AI-adjacent features are only available in more recent OUAF versions. For a full picture of where the Oracle Utilities suite fits relative to SAP and other alternatives, see the comparative analysis of Oracle versus SAP and the Oracle Utilities pillar. The Oracle Utilities implementation guide covers prerequisite data and integration work that shapes whether AI investments deliver value. For independent advice on your Oracle Utilities AI roadmap, AvanSaber’s utility practice can assess your data readiness and product version landscape. ## Frequently asked questions Does Oracle CC&B have built-in AI features? CC&B itself is a transactional billing and CIS system; AI is applied to CC&B data through Oracle Utilities Analytics Warehouse, Oracle Analytics Cloud, or Oracle Machine Learning rather than inside the CC&B application directly. What does Oracle Opower actually do with AI? Opower uses machine learning models to segment customers by consumption behaviour and generate personalised home energy reports. Its models are pre-trained on anonymised consumption data across Oracle's utility customer base. How does AI fit into Oracle Network Management System? Oracle NMS can surface predictive outage risk indicators and optimise crew dispatch using historical outage, asset condition, and weather data. AI models feed into the switching and restoration workflows rather than operating as standalone tools. --- ## AI and XR Digital Twins: Operator Visualization in Utilities URL: https://www.utilitieslabs.com/revolutionizing-utilities-with-ai-and-xr-powered-digital-twins/ # AI and XR Digital Twins: Operator Visualization in Utilities AvanSaber Research/November 3, 2024/Updated September 24, 2026/4 min read The phrase “AI and XR-powered digital twin” appears in many utility technology presentations. It is worth parsing what the combination actually delivers versus what is projected benefit. This article focuses on the operator visualization angle, who benefits from XR overlays on digital twin data, and what AI provides underneath the display. ## The Operator’s Actual View A control room operator at a large distribution utility spends a shift working from an ADMS console. The ADMS, whether GE Vernova’s product, Schneider EcoStruxure, or Oracle Network Management System, provides the real-time switching state, alarm feeds, SCADA telemetry, and FLISR recommendations. This is the operator’s primary interface. A digital twin in this context is a detailed network model that the ADMS uses internally for analysis, load flow, fault current calculations, switching validation. The operator sees results from the model through the ADMS display, not through an AR/VR interface. Suggesting that operators will wear headsets to manage live switching is not accurate to how utilities run today, and it is not where the practical XR value lies. The XR value is in different roles: engineers reviewing a planned network change in a 3D environment, training coordinators running simulation exercises, or planners evaluating substation expansion options before committing to a layout. ## What AI Actually Does in This Stack AI in a utility digital twin context operates at several layers, most of which are invisible to a visualization interface: Predictive failure scoring runs on asset sensor data, maintenance history, and operating conditions to rank equipment by near-term failure probability. The output feeds the maintenance planning workflow in the ERP, a list of assets to prioritize for inspection or replacement. Load forecasting models consume AMI data from Itron or Landis+Gyr networks, weather forecasts, and historical patterns to predict hourly and daily load profiles. This feeds ADMS operational planning and DERMS dispatch. Fault location analysis uses SCADA telemetry patterns to narrow down the probable location of a distribution fault before crews are dispatched. This reduces the driving time and search time in outage response. These AI functions improve outcomes in the ADMS and ERP workflows. The visualization layer, whether a conventional ADMS display or an XR interface, is how results are surfaced, not where the AI runs. ## Where XR Visualization on a Digital Twin Adds Genuine Value The specific XR applications on digital twin data that have demonstrated value in utilities: Infrastructure design review: Engineers can load the digital twin model into a VR environment and walk through a proposed substation layout, feeder routing, or substation protection scheme at scale. This surfaces physical conflicts and design issues that 2D drawing review misses. It is a project-phase tool, not an operations tool. Outage war-room visualization: During a large storm outage with multiple simultaneous faults, an XR visualization of the network model, color-coded by fault state, customer count, and crew location, can support coordination meetings more effectively than a printout or shared screen. This is a collaboration support tool for infrequent, high-complexity events. Training simulation: VR scenarios built on the digital twin model allow operators to practice outage response procedures on a model that reflects their actual network topology. This is the strongest operational case for XR on a digital twin. For a deeper look at digital twin data architecture, how GIS, ADMS, AMI, and ERP data combine to form the network model, see the grid management and digital twins article. ## The Hype to Avoid Several vendor and analyst narratives overstate where XR digital twins are today in utility operations: “Operators will manage the grid in AR.” Current ADMS displays are purpose-built for the precision and speed that switching decisions require. AR headset overlays on a distribution network model are not replacing those interfaces in any live utility operation this decade. “Digital twins eliminate unplanned outages.” No. They reduce them, at asset types with good sensor coverage and clean maintenance records, when the predictive maintenance workflow is actually executed. That is a meaningful but bounded claim. “The metaverse will transform customer engagement.” Customer interactions in utilities run through CIS platforms, SAP C4U, Oracle CC&B, Cayenta CIS, through web portals and call centers. An immersive metaverse customer interface is not in any CIS roadmap with a near-term delivery date. ## System of Record Clarity Regardless of the visualization and AI layer, the system of record architecture stays the same: ADMS for real-time network state and switching control, ERP (SAP IS-U, S/4HANA Utilities, or Oracle CC&B) for asset records, work orders, and financials, and CIS for customer and billing data. Digital twin models and XR visualizations sit alongside these systems, not above them. For AI-driven operations improvements, see AI for utility operations. ## Frequently asked questions Who are the actual users of XR visualization on a digital twin? Primarily engineering and planning staff during design review and capital project evaluation, and field crew trainers. Control room operators use ADMS consoles, not AR/VR headsets, for live grid operations. What AI capabilities are most useful in a utility digital twin context? Predictive maintenance scoring, load flow analysis, outage scenario simulation, and DER dispatch optimization. These run in the background and surface results to the ADMS or ERP, not always through an XR interface. Does the digital twin replace the ADMS or ERP? No. The ADMS (GE Vernova, Schneider EcoStruxure, Oracle NMS) is the real-time network model and switching control system. The ERP manages asset records, work orders, and financials. The digital twin is a modeling and analysis layer, not a replacement. Is XR visualization of digital twins mature technology in utilities? For design review and training applications, yes. For real-time operational control, no. Most utilities with digital twin programs use conventional ADMS visualization for operations. --- ## AI Data Analytics in Utilities: Loss, Demand, and Asset Outcomes URL: https://www.utilitieslabs.com/revolutionizing-utilities-with-ai-driven-data-analytics/ # AI Data Analytics in Utilities: Loss, Demand, and Asset Outcomes AvanSaber Research/January 10, 2024/Updated June 2, 2026/3 min read Utility analytics conversations often stay generic: “AI will improve efficiency” without specifying which efficiency metric, whose data, and what the workflow looks like when a result is produced. This article takes three concrete outcome areas and traces the data and system connections behind each. ## Non-Technical Loss Detection Non-technical loss (NTL) sits at the intersection of AMI data and CIS billing records. The signal is a gap: a meter showing reduced or erratic consumption relative to its historical pattern, while the billing system shows no legitimate explanation such as a rate change, vacancy, or seasonal adjustment. AI models trained on AMI interval data from Itron or Landis+Gyr meters can score every account in a service territory for NTL probability without requiring individual analyst review. The score is written back to the CIS as a work queue item. A billing investigator reviews accounts above a threshold, orders a field check, and the CIS records the outcome. What AI does not do here is change the billing record autonomously. The CIS is the system of record for all account data; the analytics model is advisory. This matters for audit trail and regulatory compliance purposes. ## Short-Term Load Forecasting and Demand Response Accurate short-term load forecasting directly affects generation scheduling, capacity market positions, and the dispatch of demand response programs. Traditional regression models work reasonably well under normal conditions. They degrade when conditions depart from historical norms, which is precisely when forecast accuracy matters most. Machine learning models that combine AMI load shapes, weather data, and calendar features handle non-linear interactions between variables more effectively. For distribution operations, feeder-level load forecasts built from AMI data give the ADMS better inputs for Volt/VAR optimization setpoints and predictive switching plans. Oracle Opower and Bidgely both build customer-level consumption models from AMI data. Opower’s behavioral energy reports use those models to drive demand reduction. Bidgely’s disaggregation approach identifies which appliances are contributing to load, enabling targeted efficiency recommendations. Both platforms return insights through the customer portal or CIS, maintaining the utility as the interface with the customer. The AI consumption optimization article covers the demand-side detail, and the smart grid and AI optimization post covers how forecast data feeds grid operations. ## Asset Health Analytics Distribution asset failure is expensive in two ways: emergency repair costs and the customer-minutes interrupted during an unplanned outage. Predictive asset analytics attempts to shift maintenance from calendar-based to condition-based, prioritizing the assets most likely to fail before the next planned cycle. The data inputs vary by asset type: - Overhead lines and poles: storm damage history, inspection records, LiDAR point cloud data (for vegetation proximity), and age from the GIS and asset register in the ERP. - Substation transformers: oil analysis results, dissolved gas analysis trends, thermal model outputs from continuous monitoring, and loading history from SCADA. - Underground cables: insulation testing history, fault history by cable segment, and soil temperature data where available. AI models on this data produce risk scores by asset. The output feeds the capital planning process and the annual maintenance schedule. A planner reviews the ranked list, adjusts for budget and crew constraints, and issues work orders through the ERP or work management system. The model recommends; the planner decides and the ERP records. For utilities on SAP, this connects directly to the SAP IS-U plant maintenance integration and the broader utility billing ERP data environment. For Oracle environments, see the Oracle Utilities pillar for how Oracle’s asset management capabilities fit this picture. ## The Common Thread Across All Three Areas All three use cases share the same architectural principle: AI models read from the system of record, whether the CIS, AMI data management system, or ERP asset register, produce a scored or ranked output, and return it to a human workflow for confirmation. The data systems retain authority over billing, account status, and work order issuance. Utilities that have not yet connected AMI data to the CIS analytics layer should prioritize that integration before building prediction models. Clean, reconciled interval data is the prerequisite for all three outcomes described here. Avansaber can help assess the readiness of your current data architecture. ## Frequently asked questions What is non-technical loss and how does AI help detect it? Non-technical loss (NTL) refers to energy billed for but not received as revenue, caused by meter tampering, billing errors, or theft. AI models compare AMI interval consumption data against billing records in the CIS to flag accounts where patterns suggest NTL. The CIS remains the record of billing; field verification confirms the flag. Which analytics tools are purpose-built for utility customer energy data? Oracle Opower, Bidgely, and Uplight are the three most widely deployed customer energy analytics platforms. All three ingest AMI interval data and return personalized insights or demand flexibility signals back through the CIS or customer portal. Does AI improve demand forecasting compared to traditional statistical models? For short-term load forecasting, machine learning models that incorporate weather variables, AMI load shapes, and economic indicators tend to outperform classical time-series methods, particularly during weather extremes. For long-range planning, the accuracy difference is smaller and governance of model assumptions matters more than model type. --- ## AI and AR for Utility Asset Inspection and Field Management URL: https://www.utilitieslabs.com/revolutionizing-utility-asset-management-with-ai-and-xr/ # AI and AR for Utility Asset Inspection and Field Management AvanSaber Research/December 11, 2024/Updated June 2, 2026/4 min read AI-driven predictive maintenance and AR field tools are the two most practical technology overlays on utility asset management today. They address different problems, one predicts which assets need attention before failure, the other helps the technician who shows up to do the work. Both depend on the ERP as the system of record. ## AI for Predictive Asset Maintenance The operating model for AI-driven predictive maintenance in utilities is well-established at the concept level, but results vary significantly based on data quality. The AI system consumes sensor data (thermal, vibration, partial discharge, pressure), maintenance history from the ERP, asset age and specification data from GIS and the asset master, and sometimes AMI-derived load data to infer thermal stress on distribution transformers. Machine learning models then score each asset by estimated failure probability within a time window. This scoring is useful when it feeds directly into the maintenance planning workflow in the ERP. In SAP IS-U or S/4HANA Utilities, that means creating maintenance notifications or planned orders in the PM module, prioritized by the AI score. In Oracle CC&B environments with integrated EAM, it means populating work order queues. Without that ERP integration, the AI output sits in a separate dashboard and does not change how maintenance is actually scheduled. The cases where this works well: distribution transformers with sensor-equipped bushings, underground cable circuits with partial discharge monitoring, and gas mains in systems that have already deployed acoustic leak sensors. The cases where it works poorly: assets with sparse maintenance history, aging infrastructure where the ERP records were never consistently populated, and asset classes where sensor deployment is limited. ## AR for Field Asset Inspection AR field tools for asset inspection have a different value proposition from predictive maintenance: they help the technician who is already at an asset do the job more accurately and capture better data back to the ERP. The core workflow: a technician arrives at an asset, scans a barcode or QR code (or, in more sophisticated systems, uses visual recognition), and the AR application surfaces the asset record. This includes installation date, nameplate specs, last maintenance date, any open work orders, and prior inspection notes. The technician works through a structured checklist tied to the asset class. Findings, measurements, photos, condition ratings, write back to the ERP through the same API connection. This matters for data quality. Inspection findings captured in a structured AR workflow are more consistently recorded and more reliably tied to the correct asset record than findings entered from memory after returning to the truck or the office. For the specific case of SAP IS-U and S/4HANA Utilities, how AR platforms integrate through SAP APIs and what the PM module data model looks like, see AR integration with SAP utilities. ## Remote Expert Assist A scenario that has proven genuinely useful in utility asset management is remote expert assist via AR. A field technician encountering an unfamiliar fault condition or a non-standard equipment configuration connects to a specialist via AR video. The specialist sees the technician’s camera feed and can annotate the physical environment, highlighting a specific terminal, drawing a connection path, flagging a clearance concern. This reduces specialist dispatch for non-routine situations without removing expert knowledge from the equation. It also creates a video record of the interaction that can be attached to the work order. Enel Group’s deployment of AR for field maintenance is a publicized example of this pattern at utility scale. ## Drones in the Inspection Stack Drone inspection is a distinct but complementary technology to AR field tools. Drones with thermal cameras and visual inspection payloads are used for: - Transmission line patrol and hotspot identification - Substation yard surveys for thermal anomalies on high-voltage equipment - Solar array inspection - Pipeline right-of-way surveillance for encroachment and vegetation The AI analysis of drone imagery, identifying thermal hotspots, classifying conductor condition, flagging vegetation proximity, is maturing and reducing the manual review burden. The findings need to flow to the ERP asset record and work order system, same as AR-captured inspection data. Drones do not replace technician-level AR for equipment that requires close-in inspection and hands-on assessment. The two tools address different points in the inspection workflow. ## The ERP Dependency Every AI and AR tool in this stack depends on ERP data quality to deliver its claimed value. Predictive maintenance needs accurate maintenance history and asset specifications. AR field tools need a populated asset master with correct equipment records. Remote assist records need work orders to attach to. Utilities that get value from these tools have typically invested in ERP data quality improvement as a prerequisite, cleaning up asset records in SAP PM or Oracle EAM, standardizing equipment classifications in GIS, and ensuring maintenance records are captured consistently. The technology layers amplify whatever data foundation exists beneath them. For the broader AI-in-operations picture, see AI for utility operations. ## Frequently asked questions Which asset types benefit most from AI-driven predictive maintenance? Distribution transformers, underground cable circuits, gas mains in aging systems, and water mains with corrosion history. Assets with consistent sensor coverage and clean maintenance history produce the most reliable failure predictions. How does AR field tooling connect to the ERP asset record? Via API to the ERP asset module, SAP PM within IS-U or S/4HANA, Oracle EAM, or the asset module in Cayenta CIS. The AR interface reads asset data and can write inspection notes or work order updates back to the ERP record. What is the difference between AR for inspection and AR for remote expert assist? Inspection AR surfaces asset data and structured checklists to a technician on site. Remote expert AR connects a field worker (with a headset or tablet camera) to a specialist who can see their view and annotate it with guidance. Both are useful; they address different scenarios. Do drones replace AR headsets for asset inspection? They address different scenarios. Drones are better for aerial inspection of transmission lines, large substation yard surveys, and hard-to-access terrain. AR headsets and tablets are better for equipment-level inspection where a technician is already on site. --- ## AI and XR for Utility Customer Service: Proactive and Field Support URL: https://www.utilitieslabs.com/revolutionizing-utility-customer-service-with-ai-and-xr/ # AI and XR for Utility Customer Service: Proactive and Field Support AvanSaber Research/November 5, 2024/Updated June 2, 2026/2 min read Most coverage of AI in utility customer service stops at the chatbot. The more useful frontier is proactive communication and field support, where AI and a narrow slice of AR change the shape of the interaction rather than just automating the front door. This piece covers those two angles; for the chatbot and agent-assist basics, see the companion articles linked below. ## Proactive beats reactive The cheapest support interaction is the one that never becomes a call. Driven by events from the CIS (SAP IS-U, Oracle CC&B, or Cayenta CIS) and the outage management system, utilities push: - Outage alerts and restoration updates before the customer calls to report or ask. - High-bill and usage alerts that explain a spike before it becomes a billing dispute. - Payment reminders and arrangement offers that reduce both calls and write-offs. AI improves the targeting (who to notify, when, on which channel) and drafts the messaging, but the triggering data lives in the systems of record. Cayenta’s Silverblaze, Oracle’s digital self-service, and SAP’s IS-U self-service are the delivery surfaces. ## Omnichannel self-service Customers expect to start on the app and finish on the phone without repeating themselves. AI helps by carrying context across channels and resolving routine intents (balance, payment, outage status) in any of them, again reading from the CIS. The detail on chatbots and agent-assist is in AI and XR in utility customer service and AI in utility customer service. ## Where XR actually fits XR on the customer side is mostly experimental, with one practical exception: AR remote-expert support. A back-office specialist sees the technician’s (or sometimes the customer’s) camera view and annotates it in real time, which can resolve a meter or connection issue faster or avoid a second visit. That is a field-support function tied to the work order, not a self-service channel. The field-work detail is in SAP augmented reality for utilities. ## The takeaway Invest first in CIS-triggered proactive notification and omnichannel self-service, because they remove calls against real account data. Treat AR remote-expert support as a targeted field-support tool, and keep customer-facing XR as a pilot. The wider picture is in customer experience in utilities. ## Frequently asked questions What is proactive utility customer service? Getting ahead of the call: outage alerts, high-bill alerts, and payment reminders pushed by SMS, email, and app, triggered by events from the CIS and outage management system. It converts inbound calls into outbound notifications. Does AR help with in-home utility service? In a narrow way. AR remote-expert tools let a back-office specialist see what an in-home technician or even a customer sees and annotate it, which can resolve some service visits faster or avoid a truck roll. It is a support aid, not a customer self-service channel. --- ## AI in Utility Customer Service: Chatbots and Agent Assist on the CIS URL: https://www.utilitieslabs.com/revolutionizing-utility-customer-service-with-ai/ # AI in Utility Customer Service: Chatbots and Agent Assist on the CIS AvanSaber Research/April 19, 2024/Updated September 24, 2026/3 min read Utility customer service AI falls into two distinct deployment modes: customer-facing self-service (chatbots and IVR automation) and agent-assist tools that support human CSRs during live interactions. Both modes depend on the CIS as the system of record. Neither should execute account changes without explicit authorization through the CIS’s own workflow controls. ## What the CIS Must Provide for AI to Function Well Before evaluating any AI customer service product, a utility should assess the quality and accessibility of its CIS data. AI recommendations are only as useful as the account context they draw from. The relevant CIS data for customer interactions includes: current account balance and payment due date, recent bill history and any billing exceptions, active service orders or pending premise changes, outage records linked to the customer’s premises, and any open disputes or payment arrangements. In Oracle CC&B, this data is accessible through the OUAF web service layer. In Cayenta CIS, this data sits in the core CIS and reaches customers through channels such as the Silverblaze portal add-on. In SAP IS-U, FI-CA contract account data is available through BTP integration services or standard BAPI/IDoc interfaces. Any AI layer that cannot reliably query these fields in real time will produce inaccurate responses that damage customer trust. ## Customer-Facing Self-Service: Where Chatbots Work and Where They Fail Chatbots handle high-volume, low-complexity inquiries reliably: balance inquiries, payment confirmations, scheduled outage notifications, and start/stop/transfer request intake. These interactions follow predictable paths and do not require judgment. Where chatbots fail is in exception handling: a customer whose bill looks unusually high because of a meter read estimate that was later corrected, or one who needs a payment arrangement that falls outside standard policy parameters. These interactions require access to billing exception detail and, usually, CSR judgment. A chatbot that attempts to resolve these without escalating generates frustration. The right design routes the customer to a CSR with context when the chatbot reaches the boundary of its authorized actions. That handoff transfers the session data to the agent so the customer does not repeat themselves. Customer self-service extends naturally into customer portals. The benefits of customer portals in utility services covers how authenticated portal access reduces inbound call volume and provides the session data that AI tools need. ## Agent Assist: AI Alongside the CSR Agent assist AI is less visible to customers but often more impactful operationally. During a live call, the AI tool reads the active account from the CIS, identifies the likely reason for the contact based on recent account activity, and surfaces a recommended script or resolution path for the CSR. Cayenta CIS’s Cayla AI operates in this mode. A CSR handling a billing dispute sees Cayla’s summary of recent bill history, any validation exceptions on the meter read, and a suggested response, all without navigating through multiple CIS screens. The CSR reads the suggestion, applies judgment, and takes action in the CIS. Cayla does not write to the account. Oracle CC&B environments typically integrate third-party agent assist tools through the OUAF service layer, though Oracle’s own customer service cloud components provide similar functionality in newer deployments. SAP IS-U environments can connect to SAP’s Customer Experience (CX) portfolio or third-party contact center tools. SAP Joule can provide natural-language account summaries to an authenticated CSR within the S/4HANA interface. ## Quality Monitoring and Continuous Improvement AI tools in the contact center generate interaction data that can improve both the AI models and the underlying processes. Patterns in unresolved escalations may indicate a billing process gap rather than a chatbot design problem. A spike in contacts about a particular rate change notice may indicate a communication clarity issue. This is where the CIS analytics layer becomes relevant. Review AI analytics on utility CIS data for how to extract and act on those patterns. For the billing efficiency side, AI in utility billing covers how billing accuracy affects customer service volume. The utility billing ERP pillar provides the broader system context, and utilities evaluating Cayenta CIS as a CIS platform should review the Cayenta CIS analysis. For help evaluating AI customer service tools against your specific CIS platform and contact center environment, Avansaber offers vendor-neutral advisory. ## Frequently asked questions What CIS actions can a utility chatbot perform autonomously? Safely automated actions are limited to read-only interactions such as balance display, outage status, and payment history. Account changes like payment arrangements, premise transfers, or rate modifications require CIS authorization rules and, typically, a human CSR or explicit customer confirmation via an authenticated self-service session. What is Cayla in Cayenta CIS? Cayla is the AI assistant embedded in Cayenta CIS by Harris Computer. It surfaces account context and suggests next-best actions to the CSR during a call or chat session. Cayla does not execute account changes; the CSR reviews the suggestion and acts in the CIS. How does AI agent assist differ from a customer-facing chatbot? A customer-facing chatbot handles the customer's questions directly. Agent assist AI runs alongside the CSR during a live interaction, showing the agent relevant account data, suggested responses, or resolution steps pulled from the CIS. The CSR speaks with the customer and decides which suggestions to act on. --- ## AI and AR for Utility Safety and Leak Detection URL: https://www.utilitieslabs.com/revolutionizing-utility-safety-ai-and-xr-for-real-time-leak-detection/ # AI and AR for Utility Safety and Leak Detection AvanSaber Research/October 29, 2024/Updated June 2, 2026/4 min read Gas and water leak detection is one of the utility safety applications where AI and AR are demonstrating real operational value, with specific, honest limitations. This piece covers the technology stack behind modern leak detection and the role AR plays in field response, without overstating either. ## AI for Gas Leak Detection Gas distribution systems have several data sources that AI-assisted analysis can use for leak detection and pipeline health assessment. Acoustic sensing on pipeline networks uses sensors to detect the characteristic frequencies of gas escaping from a pipe under pressure. Established vendors in this space have deployed acoustic monitoring on urban gas mains, particularly in older cast iron and unprotected steel pipeline systems where corrosion risk is elevated. AI classifies acoustic signatures to distinguish leaks from background noise, traffic, construction, water flow in adjacent pipes, reducing false positive rates compared to rule-based threshold approaches. SCADA pressure monitoring provides another detection layer. Pressure drop patterns that deviate from normal operating envelopes, after accounting for weather, temperature, and demand variation, can indicate a developing leak or main break. AI pattern recognition on historical SCADA data can identify subtle anomalies earlier than static threshold alerts. Drone-mounted methane sensors are used for periodic pipeline survey in areas where ground-based access is difficult or where the survey frequency required outpaces available crew capacity. AI classification of methane concentration readings from drone flight paths identifies probable leak locations for ground crews to investigate. What AI does not do in gas leak detection: it does not reliably pinpoint a leak to a specific joint or fitting from remote sensing alone. It narrows the search area. Field crews with gas detection instruments still locate and confirm the specific leak point. ## AI for Water Distribution Leak Detection Water utilities have deployed AMI-enabled meters from Itron and Sensus that provide interval consumption data, often at 15-minute or hourly resolution. At the distribution zone level, comparing the total flow metered at a zone inlet with the sum of consumption recorded at customer meters identifies non-revenue water, the gap includes both leakage and metering error. AI-driven district metering area (DMA) analysis processes this data continuously to flag zones where non-revenue water is increasing, triggering investigation priorities for acoustic correlation or pressure testing. This is more systematic than the traditional approach of waiting for customer complaints or visible surface water. Pressure transient analysis, examining pressure wave patterns in water mains, is a more specialized technique used by some utilities for leak location in transmission mains. This requires dedicated sensors and specialized analysis software rather than general-purpose AMI data. ## AR in Field Leak Response When a potential gas or water leak is reported or detected, the field crew’s first tasks are isolation valve location, pipeline routing confirmation, and safe excavation planning. These depend on accurate GIS data. AR tools connected to the utility GIS, Esri ArcGIS is the most common platform, can overlay pipeline routing, valve locations, and pressure zone boundaries on the technician’s tablet or headset view as they approach the site. This is more useful than a printed map extract for dynamic field conditions, particularly in urban environments where street layout makes the physical route to a valve non-obvious. During the isolation process, AR-surfaced step sequences from the ERP work order (drawn from the same SAP PM or Cayenta CIS service order) ensure the technician follows the correct isolation procedure for the specific pipeline configuration. Photos and condition notes captured during the response attach to the work order record. Remote expert assist is particularly valuable for unusual failure modes, a damaged valve that will not operate normally, an unexpected pipe configuration at excavation depth, or a coupled failure across multiple assets. The field crew connects via AR video to a specialist who can see their conditions and guide the response without requiring a second crew to be dispatched. ## What AR Does Not Do in Safety Response AR is a field information and guidance tool. The critical isolation and control decisions in a gas or water emergency run through the control room: SCADA isolation commands, pressure system control, customer notification through the CIS. AR at the field level does not replace or bypass those systems. The ERP work order and GIS are the data sources for AR content. If the GIS pipeline record is inaccurate, a relocated valve that was never updated, an unofficial tie-in from a prior repair, AR will surface the incorrect location. GIS data quality is a prerequisite for AR field tools to add safety value rather than create confusion. ## Connecting Safety Findings to the ERP Leak response findings, confirmed leak location, repair type, affected pipeline section, post-repair pressure test results, need to flow back into the ERP and GIS as updates to asset condition records, maintenance history, and pipeline integrity data. This closes the loop: the AI models that predict future leak probability are only as good as the historical repair and condition data they train on. For utilities running SAP IS-U or S/4HANA Utilities, this means PM notification and order completion records linked to the linear asset (pipeline segment) in the asset master. For Oracle CC&B or Cayenta CIS environments, it means service order completion records with condition data attached. The safety improvement cycle is: AI detection prioritizes investigation, AR assists field response, ERP captures findings, AI retrains on updated data. Each step requires the previous one to generate good data. Skipping the ERP capture step breaks the cycle. For the broader AR and ERP integration picture, see AR for utility asset management. ## Frequently asked questions What AI methods are used for gas leak detection today? Acoustic sensors on pipelines detect characteristic leak frequencies. Pressure drop pattern analysis on SCADA data identifies anomalies. Some utilities use drone-mounted methane sensors with AI classification for pipeline surveys. AMI pressure data from smart regulators adds granularity. How does AR assist during an active gas or water leak response? AR overlays on tablets or headsets can surface isolation valve locations, pipeline routing maps from GIS, pressure zone boundaries, and step-by-step isolation procedures from the ERP work order. It reduces dependency on paper maps and radio-based information relay. Does AR replace the control room in a leak emergency? No. SCADA and the ADMS or gas control system remain the operational center for isolation decisions. AR is a field support tool that surfaces relevant information to the crew on-site. What is the role of smart meters in water leak detection? AMI-enabled water meters (Itron, Sensus) provide interval consumption data that supports district metering area (DMA) analysis. AI identifies consumption anomalies at the zone level that indicate distribution losses between meter reads. --- ## SAP CIS for Utilities: What IS-U and FI-CA Actually Cover URL: https://www.utilitieslabs.com/sap-cis-for-utilities-what-it-covers-isu-fica-s4hana/ # SAP CIS for Utilities: What IS-U and FI-CA Actually Cover AvanSaber Research/June 30, 2026/Updated August 8, 2026/7 min read If you are writing a request for proposal or comparing vendors, you have probably seen “SAP CIS” used as if it were a single product you could buy. It is not. SAP does not ship anything called SAP CIS. The phrase is shorthand that utility buyers and integrators use for the customer information system capability that SAP delivers, and knowing what actually sits behind the label changes how you evaluate it, budget for it, and plan its future. This is a vendor-neutral walkthrough of what “SAP CIS for utilities” means in practice: the components involved, how they fit the meter-to-cash flow, where customer engagement now lives, and why the ECC maintenance calendar makes the topic urgent even for utilities that are otherwise happy with what they run. ## What “SAP CIS” Actually Refers To The customer information system in the SAP world is delivered through SAP IS-U, the Industry Solution for Utilities. Per cobicon’s SAP IS-U overview, it dates back to 1998 and was built for the specific requirements of the utilities industry, covering the meter-to-cash chain: device management, meter reading, billing, invoicing, and customer service. IS-U was never a standalone island. It uses the standard SAP financials, controlling, materials, and plant-maintenance components underneath it, which is why a SAP CIS decision is really a decision about a broader SAP estate rather than a single billing tool. That coupling is a strength for utilities already standardized on SAP and a consideration for those that are not. On current systems the picture shifts. As SAP’s own S/4HANA Utilities documentation describes, the IS-U functionality is embedded in the S/4HANA platform rather than sold as a separate industry solution. So a modern “SAP CIS” is technically SAP S/4HANA Utilities. The functional heritage is the same; the packaging, database, and roadmap are different. ## The Meter-to-Cash Core: Device Management, Billing, and FI-CA Strip away the terminology and a utility CIS has one job at its center: turn consumption into an accurate, auditable bill and collect the money. SAP structures that as a chain. Device management records the physical metering estate: installations at service points, serial-number history, and the link between a meter and the account it serves. Meter reading and rating convert reads into billable quantities. Billing and invoicing apply the tariffs and produce the document. Then the financial side takes over, and this is the part buyers most often underestimate. SAP posts utility receivables in FI-CA (Contract Accounts Receivable and Payable), a dedicated subledger rather than the general accounts-receivable module. As SAP explains in its S/4HANA Utilities contract-accounting guidance, FI-CA implements the usual receivables functions (posting, payments, dunning) but is engineered to process very large document volumes, which is exactly what a utility with millions of accounts generates every billing cycle. If you are scoping a SAP CIS project, FI-CA is where the finance and IT conversations have to meet, because it is the boundary between the billing engine and the general ledger. The operational reality of running this chain, and how it differs from the platform overview, is covered in our look at what an SAP IS-U implementation changes operationally. ## Where Customer Engagement Now Lives Classic IS-U installations paired billing with SAP CRM for Utilities to handle the front office: interaction center, sales, and service processes. That is one of the biggest areas of change on the modern platform, and it catches buyers off guard. On S/4HANA, SAP replaced standalone CRM with two directions. One is SAP S/4HANA Utilities for Customer Engagement, described by SAP as a simplified version of CRM for Utilities fully embedded in S/4HANA, which as an SAP community walkthrough notes shares a common database with the IS-U entities so no middleware is required between them. The other is a cloud service option that can run against either an ECC or S/4HANA back end and is favored by utilities leading with customer-experience investment. The choice is not cosmetic: it determines integration architecture, upgrade impact, and where your agents actually work. ## BPEM: How the System Handles What Goes Wrong No high-volume billing run is exception-free, and how a CIS surfaces and clears exceptions is a practical differentiator that rarely makes it into a feature matrix. SAP handles this through Business Process Exception Management (BPEM). As SAP’s BPEM documentation describes, BPEM turns each error thrown during a business process into a structured clarification case that carries the triggering message, links to the affected objects, and a list of remediation actions. Cases land in prioritized worklists that processors use as an inbox. In billing terms, when an invoice cannot be created cleanly, the exception does not silently disappear; it becomes a tracked case with an owner and an audit trail. For a utility measuring billing timeliness and unbilled revenue, that exception-handling framework is as important as the billing calculation itself. ## The ECC-to-S/4HANA Question Is a Deadline, Not a Preference The reason SAP CIS is a live topic in 2026 even for stable installations is the maintenance calendar. Most classic IS-U runs on SAP ECC 6.0, and SAP has held firm on its end dates. Per Rimini Street’s summary of SAP’s position, mainstream maintenance for SAP ERP 6.0 (enhancement packages 6 through 8) ends December 31, 2027, with optional paid extended maintenance available through 2030. That extended window is not free. As SAP licensing analysts describe, the extended-maintenance phase carries a surcharge of roughly two percentage points on top of the standard maintenance base fee, and beyond 2030 SAP has floated a time-bound private-edition transition option reaching 2033 for continuity rather than a true extension. Practically, staying on ECC past 2027 means paying more to run a platform SAP is no longer enhancing, which is why most SAP utilities are planning their S/4HANA Utilities path now rather than at the deadline. The scale of what is moving is not trivial. SAP is among the most widely deployed utility CIS providers in the world, with a large global utilities footprint per SAP-focused industry references, so the 2027 window is prompting one of the larger coordinated CIS-modernization waves the sector has seen. ## SAP CIS against the alternatives Buyers searching this term are usually mid-decision, so the honest comparison matters more than another feature list. Three platforms carry most utility CIS deployments in North America. | | SAP (IS-U / S/4HANA Utilities) | Oracle (Customer Care and Billing) | Cayenta CIS | Typical fit | Utilities already running SAP for finance and assets | Large investor-owned utilities, complex rate structures | Municipal and cooperative utilities | Billing engine | IS-U billing with FI-CA contract accounting | CC&B rate engine, highly configurable | Configurable, narrower scope by design | The real constraint | The ECC deadline forces a decision on a schedule you did not choose | Configuration depth becomes implementation length | Smaller ecosystem, fewer implementation partners | Where it hurts | Cost and duration of the S/4HANA path | Rate configuration expertise is scarce and expensive | Scale limits at the top end The choice is rarely about which billing engine is better in the abstract. It is about which one your finance system, your rate structure, and your available implementation partners already point toward. Our SAP, Oracle, and Cayenta comparison works through that in more detail, and the Cayenta review covers the municipal case specifically. ## What It Means for Buyers If you take one thing from the terminology, take this: evaluate the components, not the label. A “SAP CIS” scope is really IS-U or S/4HANA Utilities billing, FI-CA contract accounting, a customer-engagement layer, and the exception framework that keeps billing on time, all riding on a broader SAP financial core. Each of those is a distinct decision with its own cost, integration, and roadmap implications. Two questions separate a grounded SAP CIS plan from a vague one. First, are you scoping a fresh S/4HANA Utilities deployment or a conversion of an existing IS-U system, because the data, testing, and customer-engagement choices differ sharply between them. Second, how does the SAP option compare against alternatives for your specific size and market, which is where a structured evaluation matters more than any single vendor’s pitch. Our Oracle vs SAP comparative analysis and our broader view of how utility CIS is shifting from meter-centric to service-to-cash architecture both help frame that comparison, and the SAP Utilities platform overview covers the wider module landscape. For a vendor-neutral starting point that puts SAP in context rather than at the center, the comprehensive guide to choosing the right utilities software lays out an evaluation framework you can apply before you commit to any platform, SAP included. ## Frequently asked questions What is SAP CIS for utilities? SAP does not sell a product literally named 'SAP CIS.' In utility procurement, the phrase refers to SAP's customer information system capability, which is delivered through SAP IS-U (Industry Solution for Utilities) on older ECC systems and through the embedded utilities functions of SAP S/4HANA Utilities on current ones. It covers the meter-to-cash chain: device management, meter reading, billing, invoicing, contract accounting, and customer service. Is SAP CIS the same as SAP IS-U? IS-U is the SAP component most people mean when they say 'SAP CIS.' IS-U launched in 1998 as SAP's dedicated utilities solution and became the industry's most widely deployed CIS. On SAP S/4HANA, that same functionality is embedded in the platform rather than sold as a separate industry add-on, so a modern SAP CIS deployment is technically S/4HANA Utilities, not standalone IS-U. What is FI-CA and why does it matter in a SAP utility billing system? FI-CA (Contract Accounts Receivable and Payable) is the SAP subledger built for high-volume, contract-based receivables. It handles posting, incoming and outgoing payments, and dunning for millions of utility contract accounts, which the standard general-ledger accounts receivable module was never designed to process at that scale. It is the financial backbone that sits under IS-U and S/4HANA Utilities billing. Do I need to migrate off SAP IS-U by 2027? Mainstream maintenance for SAP ECC 6.0 (the platform most classic IS-U installations run on) ends December 31, 2027, with paid extended maintenance available through 2030 and a time-bound transition option reaching 2033. Utilities do not have to move by 2027, but staying past mainstream maintenance means paying a surcharge and freezing on a platform SAP is no longer enhancing, so most are planning the S/4HANA Utilities path now. --- ## SAP IS-U: Modules, S/4HANA and Migration URL: https://www.utilitieslabs.com/sap-isu/ # SAP IS-U: Modules, S/4HANA and Migration SAP IS-U is the utilities industry solution for meter-to-cash, device management, and customer service. This guide maps the modules, explains the S/4HANA Utilities path, and gives buyers and program teams a practical way to frame migration. ## What SAP IS-U is SAP IS-U is the utilities component historically deployed with SAP ERP. It models the relationships that generic order-to-cash software does not: business partner, contract account, premise, connection object, point of delivery, installation, device, meter read, rate, bill, and payment. SAP’s current Utilities product documentation groups the meter-to-cash capabilities under SAP Utilities Core and documents SAP S/4HANA Utilities as the forward platform. The simplest way to understand the system is to follow one bill. A device records consumption. Device management validates the read. Billing applies the rate and creates the bill. Invoicing posts the receivable to Contract Accounts Receivable and Payable, usually called FI-CA. Customer service then uses the same account, premise, and contract relationships to explain or correct the result. ## SAP IS-U module map | Capability | Main SAP IS-U objects or component | What the utility uses it for | Customer and account | Business partner, contract account, contract | Who receives service, who pays, and under which terms | Premise and service | Connection object, premise, point of delivery, installation | Where service is delivered and how the location relates to the account | Device management | Device, register, installation, meter-reading document | Meter inventory, installation, exchange, reads, and validation | Billing and invoicing | Rate category, rate, billing document, invoicing document | Turning validated consumption into a charge and customer bill | Receivables and collections | FI-CA documents, payments, dunning, returns | High-volume utility receivables, payment allocation, and collections | Market processes | Point of delivery and market communication processes | Supplier switching and regulated or deregulated market exchanges These boundaries matter during implementation. A meter is not a customer. A premise is not a contract account. When migration teams flatten those relationships, the errors later appear as unbillable consumption, duplicate customers, broken move-in and move-out processing, or receivables that cannot be reconciled. ## SAP IS-U, S/4HANA Utilities, and the cloud question IS-U describes the utilities process model. S/4HANA Utilities is the current SAP platform carrying that process scope on S/4HANA. The move is therefore more than a database upgrade, but it is not a request to discard the utility data model. Programs still have to preserve rates, device history, contracts, open items, and the integrations around meter data, customer channels, payment processors, and the general ledger. The deadline belongs in the business case. SAP states that mainstream maintenance for Business Suite 7 core applications runs through the end of 2027, with optional extended maintenance through the end of 2030 in its maintenance strategy. That does not make every utility’s migration path identical. It does make unsupported assumptions about the existing estate expensive. For customer engagement, analytics, and AI, treat cloud services as adjacent layers unless the product contract says otherwise. The billing engine and FI-CA remain deterministic systems of record. Joule, BTP services, and customer-facing assistants should read approved context and propose or execute controlled actions with an audit trail. The detailed product view is in SAP Joule for utilities. ## The migration decisions that change the program Before selecting tools or a systems integrator, answer five questions: - Which rates and billing variants still earn their complexity? A technical conversion that carries every obsolete rate forward preserves cost without preserving value. - How clean are the partner, premise, installation, and device relationships? Profile them before design. Conversion rules cannot repair a data model nobody has measured. - Which interfaces are contracts and which are accidents? Inventory IDocs, APIs, files, and point-to-point jobs. Keep the interfaces with an accountable consumer; retire the rest. - How will parallel billing prove equivalence? Define tolerances and reconciliation ownership before test cycles begin. - Which innovations are required for go-live? Separate migration-critical scope from AI, experience, and analytics work that can be layered after the core is stable. The result should be a release path, not a product slogan: data remediation, core conversion, rate and bill validation, integration cutover, operational readiness, then optional intelligence layers. ## When SAP IS-U is the right fit SAP is strongest when the utility already operates a substantial SAP estate, needs deep FI-CA and finance integration, and can support a configuration-heavy enterprise program. That fit is not universal. Oracle CC&B is a serious alternative in an Oracle-centered estate, while Cayenta CIS can fit municipal and mid-market utilities that value an integrated suite and a smaller operating footprint. Use the Oracle vs SAP utilities comparison for the enterprise decision. Use the Cayenta CIS review when the question is whether the enterprise platforms are more system than the utility needs. “Most utilities do not have an IS-U problem, they have a data problem that IS-U exposes. The move to S/4HANA Utilities is worth doing when it forces a clean meter-to-cash data model, not when it is treated as a like-for-like technical upgrade.” Nikhil Jathar, founder of AvanSaber and former SAP consultant ## In this topic - Oracle vs SAP for utilities: the comparison - What a SAP utilities CIS actually covers - Joule, ECC and the on-premise AI path - SAP IDocs in utility operations ## Everything else in this topic 17 more articles, A to Z. - AI in SAP Utilities: Joule, BTP, and Billing Exceptions - Building the Business Case for SAP Utilities Finance and FI-CA - Customer Service and Process Transformation via SAP IS-U - FI-CA Day-to-Day: How SAP Utilities Finance Runs Financial Operations - Real-Time Interval Data Processing with SAP HANA Utilities - S/4HANA Utilities: The Forward View of HANA in Utility Management - SAP ABAP for Utility IS-U: Performance, Clean Core, and BTP - SAP Analytics Cloud and BTP AI: BI for Utilities on SAP IS-U - SAP BTP AI and Joule for Utility Operations on IS-U and S/4HANA - SAP BTP for Utilities: Extend IS-U, Keep the Core Clean - SAP HANA for Utilities: Real-Time Data and S/4HANA - SAP HANA for Utility Operations: Faster Billing and Management - SAP IS-U for Utilities: Modules, S/4HANA Utilities Migration, and Fit - SAP Joule for Utilities: Agentic AI on IS-U and S/4HANA - SAP Utilities Finance: FI-CA, Billing, and the Utility Revenue Process - What an SAP IS-U Implementation Changes Operationally at a Utility - XR Visualization of Real-Time SAP IS-U and S/4HANA Data --- ## SAP Joule for Utilities: Agentic AI on IS-U and S/4HANA URL: https://www.utilitieslabs.com/sap-joule-for-utilities-agentic-ai-is-u-s4hana/ # SAP Joule for Utilities: Agentic AI on IS-U and S/4HANA AvanSaber Research/June 8, 2026/5 min read SAP’s language model assistant Joule has been part of the SAP narrative since 2023, but the framing shifted substantially in the twelve months leading into mid-2026. The headline word is now “agentic.” Understanding what that means in practice, and what it does not mean yet, is useful before a utility team commits time and budget to an AI roadmap built on the SAP stack. ## What “Agentic” Actually Means A traditional AI assistant responds to a prompt. You ask a question; it returns text. An agentic system goes further: it can plan a multi-step task, call tools or APIs, read from one part of a system, write to another, and complete the task with minimal human involvement at each individual step. For a utility, the practical difference is scope. A non-agentic Joule might surface a customer’s payment history when a call-center agent asks. An agentic Joule can receive a customer query, look up the account, identify a billing exception, initiate a correction workflow, and send a resolution message, all without the agent manually touching each step. The IS-U or S/4HANA Utilities system remains the system of record; Joule orchestrates the sequence around it. For more on how AI sits alongside the ERP rather than replacing it, see our guide to integrating AI with SAP and Oracle for utility operations. ## Joule’s Scale Across the SAP Portfolio The agent and skill counts SAP publishes have moved fast and the numbers carry a date stamp because they change. At SAP TechEd in November 2025, SAP set a Q4 2025 target of roughly 40 agents and 2,100 Joule Skills. By the Q1 2026 release summary published in April 2026, SAP reported 30 or more specialized agents and more than 2,500 Joule Skills. At SAP Sapphire in May 2026, SAP reframed the count under a new Autonomous Suite umbrella, citing more than 200 agents and more than 50 assistants. Not all of those apply to utilities. Utilities-specific coverage is a subset of the portfolio, and teams should evaluate which agents map to their actual processes in IS-U or S/4HANA Utilities before drawing conclusions from headline totals. ## The Utilities Customer Self-Service Agent One concrete utilities deliverable is the Utilities Customer Self-Service Agent. SAP reported it reached general availability in Q4 2025. It is designed to handle inbound customer inquiries, common account tasks, and routine service requests without routing every interaction to a human agent. For a utility running SAP IS-U or S/4HANA Utilities as its customer information system, this agent pulls account data from the CIS and responds through a conversational interface. The CIS remains authoritative; the agent operates in front of it. The practical question for any utility evaluating this is whether the customer journeys the agent covers align with the service types that generate the most call volume in their operation. ## Joule Embedded in Asset Management Separately from the customer-facing agent, SAP embedded Joule in the Asset Management module in the S/4HANA 2025 on-premises release in October 2025. The embedded capabilities include natural-language data retrieval and work-order creation, meaning a maintenance technician or planner can describe a task in plain language and have Joule structure the work order in the system. For asset-intensive utilities managing transmission and distribution infrastructure, this addresses a real friction point: the time and training required to navigate SAP Plant Maintenance screens to create and update work orders. Natural-language input lowers that barrier, though the underlying data model and configuration quality in PM still determine whether the results are accurate. The SAP S/4HANA Utilities and BTP platform context is useful background for teams thinking about where these embedded capabilities sit in the architecture. ## The RWE Offshore Wind Example At SAP Sapphire in May 2026, SAP cited RWE’s offshore wind operations as an example of agentic AI reducing unplanned downtime. SAP framed this as exploring or introducing agentic AI in the context of their broader Industry AI coverage across 26 industries, not a fully deployed production system. RWE is an early reference customer in a program SAP is actively developing, and utilities evaluating similar use cases should treat it as a directional signal rather than a proven deployment template. The underlying concept, using agentic AI to detect anomalies and route maintenance actions in grid or generation assets, is a logical fit for utilities that have already standardized asset data in SAP PM and have reliable sensor feeds. The quality of the underlying data in the ERP is the main gating factor. ## E.ON as a Real Production Adopter A more quantified reference point is E.ON. Reported in June 2026, E.ON has been deploying SAP S/4HANA-based AI for predictive maintenance and customer service and reported a 77% reduction in IT downtime over five years. E.ON is one of the larger European energy distributors, and the result points to what a mature, sustained SAP AI program can produce over a multi-year horizon rather than a short pilot. The distinction from the RWE example is that E.ON’s reported outcomes are from an operational deployment at scale, not a capability preview. Utilities in mid-migration from IS-U on ECC to S/4HANA Utilities can use E.ON as a concrete benchmark for what the destination state can deliver. For a comparison of where SAP and Oracle currently stand on this kind of capability, see our Oracle vs SAP utilities comparison. ## What Joule Does Not Do Yet Several capabilities remain forward-looking. The SAP AI Agent Hub and a new managed Joule Studio are both targeted for general availability in Q3 2026, which is when utilities teams would be able to build and govern their own custom agents in a managed environment. The original Joule Studio agent builder reached GA on December 19, 2025, giving developers an early path, but the managed enterprise-grade tooling is still ahead. The Anthropic partnership announced May 12, 2026 will embed Claude in the SAP Business AI Platform via the Model Context Protocol. Utilities are named among the target industries in that announcement, but no utilities-specific agent under the Anthropic partnership has shipped yet. It expands the model choice available within the SAP AI Platform rather than delivering a ready-made utilities workflow today. For utilities teams doing roadmap planning, the pattern is: the customer-facing and asset-management capabilities are available now, the custom-agent tooling arrives in Q3 2026, and the multi-model capabilities that follow from the Anthropic partnership are later still. Building the AI roadmap in that sequence, starting with the GA agents and working toward custom builds, is a more grounded approach than planning around capabilities that are still targets. The SAP IS-U overview covers the platform context for teams earlier in their evaluation. ## Frequently asked questions What is the difference between a Joule chatbot and an agentic Joule workflow? A Joule chatbot answers a question and stops. An agentic workflow can take a sequence of actions across SAP modules, for example retrieve a work order, check a schedule, update a field, and notify a team, without a person approving each step. The utility keeps human oversight at the workflow level rather than at every click. Is the SAP Utilities Customer Self-Service Agent generally available? Yes. SAP reported it reached general availability in Q4 2025. Does Joule work on on-premises IS-U or only in the cloud? Joule's agentic features for S/4HANA Utilities are primarily cloud-first. SAP announced in May 2026 that Joule agents can work in hybrid landscapes connecting on-prem S/4HANA and ECC, but only via RISE with SAP and the paid Max Success Plan. Standalone on-prem IS-U on ECC is not in scope. What did E.ON report about SAP AI in their infrastructure? E.ON reported a 77% reduction in IT downtime over five years while deploying SAP S/4HANA-based AI for predictive maintenance and customer service, as reported in June 2026. --- ## SAP On-Prem AI U-Turn: Joule on ECC and S/4HANA for Utilities URL: https://www.utilitieslabs.com/sap-on-prem-ai-u-turn-joule-ecc-s4hana-utilities/ # SAP On-Prem AI U-Turn: Joule on ECC and S/4HANA for Utilities AvanSaber Research/May 19, 2026/Updated June 2, 2026/4 min read For years the standard advice to utilities running SAP IS-U on ECC was straightforward: Joule and the newer agentic AI capabilities were cloud-native features of S/4HANA, and on-premises customers would need to plan their migration to access them. In May 2026, SAP changed that position. The shift matters for utilities mid-migration or still evaluating their path forward, but the conditions attached to it matter equally. ## What SAP Changed Around May 13, 2026, SAP announced that Joule assistants and agents would work in hybrid landscapes connecting on-premises S/4HANA and ECC systems to the SAP AI infrastructure. This reversed a previously clear boundary: the most capable Joule features had been positioned as exclusively cloud, meaning any utility wanting them would need to complete its transition to S/4HANA in the cloud first. The reversal allows a utility still running IS-U on ECC to access Joule agents without completing a full system migration first. That is a meaningful change in timeline flexibility for organizations with complex, multi-year migration programs. For background on the IS-U platform and its modules, the SAP IS-U guide covers the architecture in detail. ## The Conditions: RISE with SAP and the Max Success Plan The announcement came with conditions that significantly narrow who benefits. According to SAP’s own framing and reporting by CIO.com, the hybrid Joule capability requires two things: enrollment in RISE with SAP, which is SAP’s subscription-based managed cloud offering, and subscription to the Max Success Plan, which is a paid premium support and services tier on top of RISE. A utility on a standard RISE contract without Max Success Plan does not qualify. A utility on an on-premises license agreement without RISE does not qualify. The announcement reaches a narrower audience than the headline implies: organizations already on RISE or prepared to move to it, and willing to pay for the Max Success Plan tier, can connect their on-prem systems to Joule agents. For utilities deep in multi-year IS-U migrations, this could be meaningful if the cost of the Max Success Plan is lower than accelerating the full cloud migration. For those not yet on RISE, the on-prem AI offer is, in practical terms, a cloud commitment with an on-prem connectivity feature attached. The SAP Business Technology Platform layer that underpins this connectivity is worth understanding for any utility evaluating the architecture. The SAP BTP overview explains how BTP functions as the integration and extension layer that makes hybrid scenarios like this technically possible. ## Why It Matters for IS-U on ECC Utilities SAP IS-U on ECC is running on a defined support clock. Mainstream maintenance runs through 2027; extended maintenance is available through 2030. The forward path SAP supports is S/4HANA Utilities. For utilities that have not yet migrated, the 2030 extended maintenance boundary is the practical hard deadline for planning their transition. The May 2026 announcement does not change those dates. What it changes is the calculus about when a utility needs to complete the cloud migration in order to access modern AI capabilities. Previously, any utility wanting Joule agents had to move to S/4HANA in the cloud first. Now, a utility can get Joule agents on its existing ECC landscape if it is on RISE with Max Success Plan, potentially buying time for a more deliberate migration while still accessing AI tooling. For utilities still comparing their options, the Oracle vs SAP utilities comparison provides useful context on where each vendor’s AI roadmap stands relative to on-prem and cloud deployment models. ## The Analyst Pushback The analyst community’s response was cautious. Reporting by The Register in May 2026 summarized Gartner and Forrester concerns that the RISE with SAP plus Max Success Plan structure effectively functions as a lock-in mechanism. Analysts described it as a path that draws customers in with the promise of AI on existing systems but requires a cloud commitment as the price of entry, making it harder to exit the SAP cloud ecosystem later. The criticism is structurally consistent with how enterprise software vendors have historically bundled new capabilities into subscription tiers. Utilities should pressure-test whether the Max Success Plan pricing, combined with RISE subscription costs, represents genuine value compared to accelerating the migration to S/4HANA Utilities on their own timeline. ## A Practical Read for Utilities Mid-Migration For a utility in active IS-U migration, the May 2026 announcement is most useful as a planning input rather than a direction change. The core migration to S/4HANA Utilities remains the right long-term path; the 2027 and 2030 maintenance dates have not moved. What is new is that a utility can now access Joule agents on a hybrid ECC landscape as an interim state, provided it is prepared to take on the RISE with Max Success Plan commitment. The questions worth asking are: Does the AI capability accessible through the Max Success Plan justify its cost relative to the migration timeline already planned? Which specific Joule agents are available in the hybrid ECC scenario, and do they address the highest-value processes in our operation? What does the contract structure look like if we decide to exit RISE later? The SAP IS-U pillar page and the SAP BTP guide together provide the platform context that informs those questions. The on-prem AI offer from SAP is real, but so are the conditions attached to it. Utilities that enter it with clear-eyed expectations about cost and lock-in will be better positioned than those who treat it as a no-strings bridge to AI on legacy systems. ## Frequently asked questions What did SAP change in May 2026 regarding on-premises AI? SAP announced that Joule assistants and agents would work in hybrid landscapes that connect on-premises S/4HANA and ECC systems. Previously, Joule's agentic capabilities were cloud-first. The reversal came with conditions: utilities must be on RISE with SAP and subscribe to the paid Max Success Plan. Can a utility on SAP IS-U and ECC access Joule agents without moving to the cloud? No. The hybrid connectivity announced in May 2026 requires RISE with SAP plus the Max Success Plan. A utility running IS-U on ECC without those subscriptions is not in scope for the on-prem Joule capability. What is the support timeline for SAP IS-U on ECC? SAP IS-U on ECC has mainstream maintenance through 2027, with extended maintenance available through 2030. The forward path SAP supports beyond that is S/4HANA Utilities. Why are analysts concerned about the on-prem Joule offer? Analysts at Gartner and Forrester flagged that the RISE with SAP plus Max Success Plan conditions effectively push customers toward a cloud commitment as the price of accessing AI on their existing on-prem systems. The 'catch' is that accessing AI for legacy infrastructure requires signing on to SAP's cloud journey. --- ## SAP vs Oracle vs Cayenta: How They Are Adding AI in 2026 URL: https://www.utilitieslabs.com/sap-oracle-cayenta-utility-cis-ai-2026-comparison/ # SAP vs Oracle vs Cayenta: How They Are Adding AI in 2026 AvanSaber Research/June 14, 2026/5 min read Every major utility software vendor now leads its marketing with AI. That makes comparison harder, not easier, because “AI” covers everything from a help-text chatbot to an autonomous agent that posts transactions. This post cuts through the announcements to show what SAP, Oracle, and Cayenta have actually shipped, what is still on a roadmap, and which capabilities map to which utility sizes. “Every utility CIS vendor now leads with AI, and most of the demos are a chatbot on top of the same billing engine. The question that separates them is architectural: can the AI actually post a transaction or adjust a bill under audit, or does it just summarize what a human already did?” Nikhil Jathar, founder of AvanSaber and former SAP consultant ## Why the AI comparison matters now The 2026 procurement cycle is the first where AI capabilities are a genuine differentiator rather than a checkbox. Utilities replacing aging CIS platforms are evaluating ten-plus-year commitments, and the AI architecture a vendor chooses today shapes what the platform can do in 2030. That is reason enough to compare on AI specifically, separate from the deeper platform reviews in our Oracle vs SAP utilities comparison and the Cayenta CIS review. ## SAP: Joule, Industry AI, and the autonomous enterprise push SAP’s AI story runs through Joule, its AI copilot and agent layer embedded in S/4HANA and the SAP Business Technology Platform. The scale of the buildout is significant: as of Q1 2026, SAP reported more than 30 Joule agents and 2,500-plus skills. By the Sapphire conference in May 2026 that count had grown to 200-plus agents and 50-plus assistants, spanning finance, supply chain, and industry verticals. For utilities specifically, SAP announced the Utilities Customer Self-Service Agent as generally available in Q4 2025. That agent handles inbound customer interactions through natural-language channels, pulling data from the underlying SAP CIS and FI-CA contract accounts. Model breadth also expanded in May 2026 when SAP and Anthropic announced that Claude is available through the SAP Business AI platform via the Model Context Protocol, giving customers a choice of foundation model alongside SAP’s own and other provider options. At Sapphire, SAP named RWE as a utility customer in the context of its autonomous enterprise showcase, signaling that at least one large investor-owned utility is engaged with the agentic automation roadmap. What RWE has in full production versus pilot is worth clarifying in any vendor conversation. SAP’s AI is strongest when the utility already runs a deep SAP estate (S/4HANA, FI-CA, SAP MDM). Joule agents read and act within that estate; they are not designed to orchestrate outside it. ## Oracle: Opower household scale and targeted CIS AI Oracle’s AI approach is more surgical. The largest deployed footprint is in customer engagement through Opower, which Oracle reported at approximately 44.6 million households enrolled as of April 2026. That figure, which Oracle reported and should be validated against current disclosure, represents the enrolled base for personalized energy insights, behavioral efficiency nudges, and demand-response communications. It is the most scaled AI-driven customer engagement asset any of the three vendors can point to. On the CIS operations side, Oracle shipped AI-powered call summarization for utility customer service in May 2025, reducing after-call work for agents. Separately, Oracle introduced AI anomaly detection in its Meter Data Management product in June 2025, targeting loss detection and faulty-meter identification in AMI interval data. These are bounded, production-shipped capabilities. Oracle is not yet leading on the broad agentic automation narrative that SAP is pushing, but it is shipping specific AI features into parts of the stack that matter to utility operations. ## Cayenta: Cayla agent in production at mid-market utilities Cayenta’s AI layer is called Cayla, which covers both a voice and digital customer-facing agent and an in-app assistant for staff. The scope is narrower than Oracle’s or SAP’s, but Cayenta has a documented production deployment. The City of Lakeland, Florida went live with the Cayla agent in approximately May 2026, implemented with partner Dial AI. A municipal utility running Cayla in production is meaningful evidence of real-world readiness at the mid-market scale Cayenta targets. ## Comparison by capability and utility size | Capability | SAP (Joule / Industry AI) | Oracle (Opower + CC&B AI) | Cayenta (Cayla) | Customer self-service agent | GA, Q4 2025 (utility-specific) | Opower engagement scale (44.6M households, Oracle-reported) | GA, production at Lakeland FL ~May 2026 | Call / interaction summarization | Joule CX layer | Shipped May 2025 | In-app assistant | Meter / MDM anomaly detection | SAP MDUS + Joule | Shipped June 2025 (Oracle MDM) | SmartWorks MDM (AI features not separately dated) | Agent / agentic automation breadth | 200+ agents, 50+ assistants (Sapphire May 2026) | Targeted; not a broad agent platform | Narrow; Cayla scope | Best fit by utility size | Large IOU on SAP estate | Large IOU on Oracle estate, Opower cross-stack | Municipal and mid-market | Model choice | SAP AI + Claude (Anthropic, May 2026) | Oracle LLMs | Not publicly disclosed ## An honest verdict by utility size Large investor-owned utilities on SAP get the broadest AI surface area in 2026, but the value is proportional to how deeply the SAP estate is already deployed. Joule does not add much value to a utility that runs IS-U standalone without S/4HANA Finance, BTP, and the surrounding stack. Large investor-owned utilities on Oracle get a mature, scaled customer engagement platform in Opower and targeted AI in billing and MDM. The agentic automation story is less developed than SAP’s, but the shipped capabilities are real and production-proven. Municipal and mid-market utilities evaluating Cayenta get a narrower AI layer that is already in production at a comparable peer, without the implementation complexity of the enterprise platforms. For many municipals, Cayla doing a focused job reliably is worth more than access to 200 agents that require significant configuration and integration. The honest framing is this: AI is not yet a reason to change CIS platforms. It is a reason to ask harder questions about your current vendor’s roadmap, the integration architecture AI will require, and how human-in-the-loop controls will work for regulated billing decisions. Those questions connect to the broader platform evaluation covered in our guide to choosing the right utilities software. ## Frequently asked questions Which utility CIS platform has the most mature AI in 2026? SAP has the broadest published AI surface area in 2026, with hundreds of agents and assistants announced at Sapphire in May 2026. Oracle leads on customer engagement AI through Opower's household enrollment scale and has shipped targeted AI in CC&B call summarization and MDM anomaly detection. Cayenta's Cayla agent is narrower in scope but is already in production at mid-market utilities. Maturity depends on the task, not the vendor. Is the SAP Joule customer self-service agent available for utilities today? SAP announced the Utilities Customer Self-Service Agent as generally available in Q4 2025. The broader Joule agent and skills expansion (200+ agents announced at Sapphire in May 2026) is rolling out across the SAP Business AI platform. What is Cayenta's Cayla AI? Cayla is Cayenta's AI layer for utilities. It covers a voice and digital agent for customer interactions and an in-app assistant for staff. The City of Lakeland, FL went live with the Cayla agent in approximately May 2026 using Dial AI as the implementation partner. Does Oracle Opower work with non-Oracle CIS platforms? Oracle Opower is a customer engagement platform that can integrate with various CIS environments, though Oracle naturally pairs it most closely with its own stack. Buyers should validate integration scope against their specific CIS during procurement. --- ## Smart Grid Optimization: ADMS, DERMS, and AI for the Distribution Grid URL: https://www.utilitieslabs.com/smart-grid-optimization-ai-powered-sustainable-energy-solutions-2/ # Smart Grid Optimization: ADMS, DERMS, and AI for the Distribution Grid AvanSaber Research/August 22, 2025/Updated June 2, 2026/1 min read Smart grid optimization is operational technology, distinct from the customer and billing systems UtilitiesLabs covers most. It is worth understanding alongside them because the same utility runs both, and the distribution grid increasingly drives what the CIS and OMS have to handle. Here is the practical map. ## ADMS: running the distribution grid An advanced distribution management system is the control platform for the grid: real-time monitoring, fault location, isolation and restoration, and Volt/VAR optimization to hold voltage and cut losses. The established platforms are GE Vernova ADMS / GridOS, Schneider Electric EcoStruxure ADMS, and Oracle Utilities Network Management. This is where most concrete “grid optimization” actually happens. ## DERMS: coordinating distributed energy Rooftop solar, home batteries, and EV charging turn a one-way grid into a two-way one. A distributed-energy resource management system forecasts and dispatches these resources so they support the grid instead of destabilizing it. AutoGrid (Schneider) and the DERMS modules from the major grid vendors operate here. ## Where AI adds value AI sits on top of ADMS and DERMS for forecasting: load forecasting, DER output prediction, and anomaly detection on grid telemetry. The value is better decisions for the operators and dispatch systems, not autonomous control of the grid. ## How it connects back Grid events become customer events. Outages flow from ADMS into the outage management system and then to customer communication driven by the CIS, and distributed-energy customers create billing complexity (net metering, time-of-use) that lands in SAP IS-U, Oracle CC&B, or Cayenta CIS. The grid and the customer systems are increasingly one problem. For the customer side, start with the Oracle vs SAP comparison; for the consumption analytics that pair with grid data, see AI for energy consumption analysis. ## Frequently asked questions What software optimizes the smart grid? Advanced distribution management systems (ADMS) from GE Vernova, Schneider Electric EcoStruxure, and Oracle Utilities run the distribution grid, including Volt/VAR optimization and fault location. Distributed-energy resource management systems (DERMS) coordinate rooftop solar, batteries, and EVs. AI adds load and DER forecasting on top of both. --- ## SAP IDocs for Utilities: Integrating IS-U and Meter Data URL: https://www.utilitieslabs.com/streamlining-utility-operations-with-sap-idocs-a-comprehensive-guide/ # SAP IDocs for Utilities: Integrating IS-U and Meter Data AvanSaber Research/December 23, 2023/Updated June 1, 2026/1 min read SAP IS-U does not run alone. Meter data comes in, bills go out, and in a deregulated market messages flow constantly with other suppliers. IDocs are the standard SAP format that carries most of that traffic. This guide explains how IDocs work in a utility and how to keep the integration clean. ## What IDocs do in a utility An IDoc is a structured message. In SAP IS-U, IDocs typically carry: - Inbound meter readings from a meter data management system into IS-U for billing. - Device and installation updates keeping the technical view in sync. - Billing and invoice output to downstream finance or print and mail systems. - Market communication messages in deregulated markets, for enrollments, switches, and meter data exchange. ## Integration tips - Monitor the IDoc queue. Stuck IDocs mean unbilled reads or missed market messages. Alerting on failed and delayed IDocs is not optional at utility volume. - Handle errors as a process. Build a reprocessing workflow, because some failures are data issues that need correction, not just retries. - Keep mappings documented. Meter and billing message mappings drift over time; documentation saves the next upgrade. ## Where this fits IDocs are part of running SAP IS-U well. For the platform overview, see the SAP IS-U guide. ## Frequently asked questions What is an SAP IDoc? An IDoc (Intermediate Document) is SAP's standard structured message format for exchanging data between SAP and other systems. In utilities, IDocs move meter readings, billing data, device events, and market messages in and out of SAP IS-U. Where are IDocs used in SAP IS-U? Common uses are inbound meter readings from a meter data management system, device and installation updates, billing and invoicing output to downstream systems, and market communication in deregulated environments. They are the workhorse interface for high-volume utility data. --- ## AI in Utility Billing: Where It Helps on IS-U, CC&B, and Cayenta URL: https://www.utilitieslabs.com/the-advantages-of-ai-in-utility-billing/ # AI in Utility Billing: Where It Helps on IS-U, CC&B, and Cayenta AvanSaber Research/March 31, 2024/Updated June 2, 2026/1 min read Utility billing is high volume and exception-heavy, which is exactly where AI can help. The useful framing is narrow: AI improves a few specific parts of the billing process on SAP IS-U, Oracle CC&B, and Cayenta CIS, while the billing engine stays the system of record. ## Where AI improves billing - Exception triage. Large utilities generate thousands of billing exceptions per cycle. A model ranks and routes them so analysts clear the highest-impact ones first. - Smarter estimated reads. When a meter read is missing, consumption history and weather can produce a better estimate than a flat average, which means fewer corrected bills later. - Fraud and loss detection. Unusual consumption patterns flag tampering, leaks, and faulty meters before they become large write-offs. ## How to keep it safe The billing engine stays in control. AI reads consumption and exception data, produces a ranking or an estimate, and a person or a controlled workflow confirms the action. No model should post an unreviewed change to a customer bill. ## What changed in 2026 The vendors moved from talking about AI in billing to shipping it. Oracle added AI anomaly detection in its Meter Data Management to cut billing exceptions and truck rolls (Oracle, June 2025), and AI call summarization in its customer platform (Oracle, May 2025). SAP brought a Utilities Customer Self-Service Agent to general availability (SAP, Q4 2025 release highlights) and agreed to bring Joule AI to on-prem S/4HANA and ECC for utilities still mid-migration (covered in SAP’s on-prem AI U-turn). For how the three big platforms compare on AI specifically, see SAP vs Oracle vs Cayenta on AI; for the finance-team side, see agentic AI for the utility back office. ## Where this fits For the wider system view, see the Oracle vs SAP comparison and the Cayenta CIS review. ## Frequently asked questions How is AI used in utility billing? The strongest uses are billing-exception triage, smarter estimated reads when a meter read is missing, and fraud and loss detection in consumption patterns. Each works on data the billing system already holds and produces a recommendation a person or controlled workflow confirms. Does AI replace the billing engine? No. The billing engine in SAP IS-U, Oracle CC&B, or Cayenta CIS stays the system of record. AI sits alongside it, improving exception handling and estimates, but the bill is still produced and posted by the billing platform. --- ## Utility Customer Portals: Self-Service That Cuts Calls URL: https://www.utilitieslabs.com/the-benefits-of-customer-portals-in-utility-services/ # Utility Customer Portals: Self-Service That Cuts Calls AvanSaber Research/June 11, 2023/Updated June 1, 2026/1 min read A utility customer portal earns its keep by deflecting calls. Every bill paid, outage reported, or service started online is a call the contact center did not take. This guide covers what a portal should do and how it connects to the CIS behind it. ## The features that actually reduce call volume - Bill view and payment. The single highest-volume reason customers call. - Usage history. Interval and monthly consumption, which also supports demand programs. - Outage reporting and status. Customers report and track outages without calling. - Service start, stop, and transfer. Move-in and move-out handled online. - Paperless billing and payment arrangements. Routine account changes self-served. ## How it ties into the CIS The portal is a customer-facing surface over the CIS. It reads account, bill, and usage data from SAP IS-U, Oracle CC&B, or Cayenta CIS, and writes back payments and service requests. With Cayenta, the engagement layer is Silverblaze. The important rule is that the CIS stays the system of record and the portal never holds the only copy of a transaction. ## The payoff A portal that covers billing, outages, and service changes moves the routine work out of the call center. For where the portal sits in the wider system, see the Cayenta CIS review. ## Frequently asked questions What should a utility customer portal include? Bill view and payment, usage history, outage reporting and status, service start and stop, paperless billing, and payment arrangements. Each of these deflects a call when it works, and each depends on a live connection to the CIS. How does a customer portal connect to the CIS? The portal reads account, bill, and usage data from the CIS (SAP IS-U, Oracle CC&B, or Cayenta CIS through a layer like Silverblaze) and writes back transactions like payments and service requests. The CIS stays the system of record; the portal is the customer-facing surface. --- ## The Future of Utilities: Business Model Shifts to Watch URL: https://www.utilitieslabs.com/the-future-of-utilities-predictions-and-trends-to-watch/ # The Future of Utilities: Business Model Shifts to Watch AvanSaber Research/June 7, 2023/Updated September 24, 2026/3 min read Utility industry predictions often focus on specific technologies, solar costs, battery storage density, EV adoption rates. Those are important, but the more consequential shifts are structural: how utilities are organized, how they earn revenue, and what their core software systems must do as a result. This piece focuses on those structural changes. Note: for a technology-specific watch-list covering AMI, DERMS, AI, and cloud CIS, see emerging utility technologies to watch. This article is the business model complement to that piece. ## The Vertically Integrated Utility Model Is Eroding The traditional model, a single regulated entity owning generation, transmission, and distribution within a service territory, is under pressure from multiple directions. Community choice aggregation programs (active in California, Illinois, and other states) allow municipalities to source supply independently while using the incumbent wires. Rooftop solar and battery storage allow customers to reduce grid dependence. ERCOT’s fully deregulated market in Texas, one of the most studied examples, demonstrates both the efficiency gains and the risk exposure that restructuring introduces. For the software implications of deregulated markets, our deregulated market pillar and the SAP utilities in deregulated markets guide cover the CIS and ERP configuration requirements in detail. ## Revenue Model Evolution and the CIS Bottleneck Regulated utilities earn revenue through volumetric rates approved by public utility commissions. As customer self-generation grows, volumetric revenues decline. The regulatory response, fixed charges, demand charges, non-bypassable charges, requires CIS platforms capable of calculating and presenting complex bills. This is not a hypothetical future state. Utilities in Hawaii, California, and Arizona are already billing under rate structures that most legacy CIS installations were not designed to handle without significant customization. Oracle CC&B, SAP IS-U, and Cayenta CIS all support complex rate structures, but configuration depth varies. A utility planning to introduce demand charges for residential customers needs to audit its current CIS configuration before designing the rate, not after. See our Oracle Utilities and SAP IS-U pillar pages for platform capability comparisons. ## The Platform Consolidation Question The mid-market CIS landscape has been consolidating through acquisition. Harris Computer’s portfolio, Cayenta CIS plus the ServiceLink, Silverblaze and SmartWorks add-ons Cayenta offers, represents one model: a single vendor family providing CIS, field service, customer portal, and meter data management. The incumbent platforms (Oracle CC&B, SAP IS-U) are moving toward cloud hosting but remain architecturally complex. Our detailed comparison of Oracle and SAP for utilities covers where each platform wins and where each creates friction. For utilities outside the large IOU tier, the Cayenta CIS review provides a substantive evaluation of the alternative. ## Data as Infrastructure AMI deployments have made utilities among the largest generators of time-series operational data of any industry. The utilities that treat this data as infrastructure, investing in governance, MDM quality, and analytics platforms, are building a capability that compounds over time. Those that treat it as a byproduct of billing are accumulating technical debt. The practical implication: the choice of MDM platform and how it integrates with the CIS has long-term strategic consequences that extend well beyond billing accuracy. Oracle Utilities MDM and SAP HANA-based analytics are both viable foundations, but neither performs well on data that is not clean. ## What Honest Forecasting Looks Like The utilities most likely to navigate structural change successfully share certain characteristics: they have reliable billing data, their CIS can handle rate complexity, their integration architecture between MDM and CIS is maintainable, and their leadership treats platform decisions as infrastructure investments rather than IT projects. The utilities most at risk are those running heavily customized legacy CIS platforms that are expensive to maintain and difficult to update, and that lack the internal data quality to support either regulatory reporting or operational AI. For an independent view of where your organization sits on this spectrum, AvanSaber’s advisory team is available for a no-obligation assessment. ## Frequently asked questions Will investor-owned utilities remain the dominant model? The IOU model is under pressure from community choice aggregation, municipalization efforts, and DER self-supply. The rate structures and CIS platforms built for the IOU model require significant rework to support these alternatives. What does deregulation mean for utility CIS platforms? Deregulated markets separate the commodity (energy supply) from the delivery (wires and pipes). CIS platforms in deregulated markets must handle switching, competitive supplier billing, and settlement with the ISO/RTO, capabilities that are distinct from regulated utility billing. How does the data economy affect utility business models? Interval AMI data, grid sensor data, and customer usage data have value beyond billing. Utilities that build the data infrastructure to aggregate and govern this data can develop new revenue streams, though regulatory frameworks governing data commercialization vary by jurisdiction. --- ## Smart Water Utility Software: SCADA, NRW, and CIS for Water URL: https://www.utilitieslabs.com/the-future-of-water-management-smart-technologies-and-innovations/ # Smart Water Utility Software: SCADA, NRW, and CIS for Water AvanSaber Research/July 1, 2023/Updated September 24, 2026/3 min read Water utilities face a distinct set of software and operational challenges compared to electric utilities. The physical asset, water in pipes under pressure, behaves differently from electrons on a wire, and the business model differs too: water is rarely deregulated, rates are typically tiered by consumption volume, and non-revenue water (NRW) is a persistent efficiency problem that spans operations and billing simultaneously. ## SCADA as the Operational Backbone Supervisory control and data acquisition (SCADA) systems are the real-time nervous system of a water utility. They monitor pressure, flow, tank levels, pump station status, and treatment process parameters. Modern SCADA platforms from vendors active in the water sector integrate with GIS systems, typically Esri ArcGIS, to give operators a network map view alongside telemetry data. The integration between SCADA and the CIS is less mature in water than in electric utilities. In electric, AMI data flows through MDM into billing as a defined process. In water, SCADA data often stays in the operational silo unless a specific integration project has connected it to billing. This gap creates reconciliation problems: the SCADA system may show distribution volumes that do not match the sum of billed consumption, the NRW delta. ## Non-Revenue Water: An Operations and Billing Problem NRW reduction is the top operational priority for most water utilities globally. The components of NRW are real losses (leaks, main breaks, storage overflow), apparent losses (meter under-registration, data errors, unauthorized use), and unbilled authorized consumption (firefighting, flushing). Each component requires a different remediation approach. On the apparent losses side, AMI deployment using Itron or Landis+Gyr water meters gives utilities read intervals comparable to electric AMI, 15-minute or hourly reads, which makes meter tampering and low-flow registration failures detectable in the data. The catch is the same as in electric: the data is only actionable if the MDM-to-CIS handoff is reliable. A billing system that processes monthly meter reads cannot generate the exception reports needed to catch apparent losses at scale. Pressure management within defined DMAs is the primary tool for real loss reduction. Lower operating pressure reduces leak flow rates and extends pipe life. SCADA-integrated pressure reducing valve control is the implementation mechanism. This is an operational technology discipline rather than a billing one, but it affects billed consumption indirectly by reducing the volume lost before it reaches a meter. ## CIS and Billing for Water Utilities Water billing has its own complexity. Tiered consumption rates (low lifeline tiers, high-use penalty tiers), seasonal drought surcharges, combined water-wastewater billing on a single account, and budget billing programs all require a CIS with a flexible rate engine. Cayenta CIS from Harris Computer serves a substantial number of water and wastewater utilities in North America. It handles multi-commodity billing (water, wastewater, solid waste) on a single account, supports AMI data ingestion, and is offered with add-ons from sister Harris products, such as the Silverblaze customer portal for self-service and SmartWorks for meter data management. For utilities that bill for both water and electric service, a unified CIS can reduce the account reconciliation burden significantly. SAP IS-U can be configured for water billing but is not widely deployed in the water sector outside utilities that also operate electric service and want a single ERP backbone. Oracle CC&B is similarly capable but sector penetration in water is lower than in electric. ## Smart Metering and Customer Programs Water AMI deployments are less mature than electric, but the trajectory is the same. Itron’s water metering platform and Landis+Gyr’s water products both support automated meter reading and leak alert notifications to customers. Customer-facing leak alerts, notifying a homeowner of unusual nighttime flow that suggests a running toilet or slow leak, are a straightforward AMI use case that demonstrably reduces NRW and improves customer satisfaction. Customer portal adoption for water utilities is lower than electric, partly because billing is less complex and partly because the engagement touchpoints are fewer. Cayenta’s Silverblaze portal supports usage dashboards, paperless billing, and payment options for water accounts, the standard toolkit for driving digital self-service adoption. ## The Integration Priority For water utilities, the highest-value integration work is connecting SCADA/pressure data with meter read data in a way that enables DMA-level water balance analysis. This does not require replacing the CIS, it requires an analytics layer that can join operational and billing data. Once that data is available, the path to measurable NRW reduction is clear. Our utility billing ERP pillar covers platform options for multi-commodity utilities. For help assessing whether your current metering and CIS infrastructure supports an NRW reduction program, contact AvanSaber. ## Frequently asked questions What is non-revenue water and why does it matter for utility software? Non-revenue water (NRW) is treated water that is produced but not billed, due to leaks, meter inaccuracies, or unbilled accounts. Reducing NRW requires operational data from SCADA and metering systems to feed into the CIS for billing reconciliation. Can electric utility CIS platforms handle water billing? Some can, with configuration. Cayenta CIS from Harris Computer, for example, serves both electric and water utilities. SAP IS-U can be configured for water but is less commonly deployed in that sector. Water-specific billing includes consumption tiers, drought rate surcharges, and wastewater components. What is a district metered area (DMA)? A DMA is a defined zone in the distribution network with controlled entry and exit points and continuous flow monitoring. DMAs are the operational foundation of pressure management and active leakage control programs. --- ## AI and Machine Learning in Utilities: Hub Overview URL: https://www.utilitieslabs.com/the-impact-of-ai-and-machine-learning-on-the-utilities-industry/ # AI and Machine Learning in Utilities: Hub Overview AvanSaber Research/June 7, 2023/Updated June 2, 2026/3 min read This article provides a hub overview of AI and machine learning applications across the utilities industry, with honest assessments of maturity and links to detailed articles on each domain. The goal is to distinguish what is operating in production today from what remains in pilot or research stages. ## The Core Principle: CIS and ERP Stay in Control Before going through applications, one framing point is worth stating clearly. Every credible AI deployment in utilities is designed so that the CIS and ERP remain the authoritative systems of record. AI tools sit alongside those systems, reading data and returning recommendations. Billing transactions, work orders, and customer account changes are confirmed by a person or a controlled workflow before they execute. This is not a limitation to route around. It is the correct architecture for regulated infrastructure operations where errors in billing or grid management have direct consequences for customers and compliance obligations. ## Operational Technology Applications Grid operations and ADMS. AI-assisted fault detection, isolation, and restoration on ADMS platforms from GE Vernova GridOS and Schneider Electric EcoStruxure ADMS is operating at scale. Volt/VAR optimization and DER dispatch coordination are mature production applications. The detailed view is in AI for smart grid management. Predictive maintenance. Machine learning models applied to SCADA telemetry, AMI event data, and work order history to score distribution transformer, pump, and feeder risk are running in production at many utilities. The approach and data requirements are covered in AI predictive maintenance for utilities and the companion tools and pipeline article. Water utility operations. Leak detection through pressure and flow pattern analysis, pressure zone optimization, and water quality anomaly detection are all in use at water utilities today. The details are in AI in water utility management. ## Information Technology Applications Meter data management and data quality. AI anomaly detection applied to MDM data from platforms like Oracle Meter Data Management and SmartWorks (the Harris MDM product Cayenta offers as an add-on) improves read validation and meter error detection before issues affect billing. The MDM and data quality context is in maximizing utility data management with AI. Customer energy analytics. Oracle Opower delivers behavioral energy reports and demand response targeting using CIS billing data and AMI reads. Bidgely provides load disaggregation that estimates appliance-level consumption from interval reads. Both tools operate as analytics layers on top of the CIS, not as replacements for it. Predictive analytics for planning. Demand forecasting, asset risk scoring, and revenue protection analysis each have specific data requirements and production patterns covered in predictive analytics for efficient utility management. CIS and ERP assistant tools. SAP Joule and Oracle’s AI assistant capabilities are beginning to appear in utility ERP workflows for knowledge retrieval and guided configuration. These are useful for accelerating analyst work but require the same governance as any tool that touches transaction data. ## Where Utilities Should Be Cautious Claims that AI can operate utility billing autonomously, handle regulatory-sensitive customer interactions without human review, or manage grid switching without operator confirmation are not supported by current production experience. These remain areas where human oversight is both appropriate and required. The most common reason utility AI pilots do not reach production is data quality, not model sophistication. Utilities that invest in MDM data completeness, work order coding discipline, and asset register accuracy before deploying analytics tend to produce results that are repeatable and defensible. For the trends overview that puts these applications in strategic context, see AI transforming the energy sector: key trends. For platform-specific guidance on how SAP and Oracle approaches differ, see integrating AI with SAP and Oracle. The AvanSaber team advises utilities on AI program design, data readiness, and integration architecture across SAP IS-U, Oracle Utilities, and Cayenta CIS environments. ## Frequently asked questions Which utility AI application has the clearest return on investment today? Predictive maintenance on distribution transformers and pumps, Volt/VAR optimization on mature ADMS platforms, and short-term load forecasting have the most consistently documented results in production deployments. Customer energy analytics programs through Oracle Opower and similar platforms have a long track record in demand response and efficiency programs. Does implementing AI require replacing SAP IS-U or Oracle CC&B? No. In every current production deployment pattern, the CIS and ERP remain the systems of record. AI tools read data from them through APIs or integration layers, produce recommendations, and write back draft transactions for human review. Replacing core CIS or ERP platforms to adopt AI is not justified by any current AI use case. What organizational capabilities does a utility need before AI delivers value? The foundational requirements are clean data in the ERP asset register and AMI/MDM systems, qualified staff to review model outputs before acting on them, and a working connection between the analytics output and the operational workflow where action is taken, usually a work order or dispatch system. --- ## Buying Modern Utility Software: A Practical Evaluation Guide URL: https://www.utilitieslabs.com/the-impact-of-modern-software-on-utilities-a-comprehensive-guide/ # Buying Modern Utility Software: A Practical Evaluation Guide AvanSaber Research/August 23, 2023/Updated September 24, 2026/3 min read Selecting a utility CIS or ERP is one of the largest capital commitments a utility makes in its software portfolio. The platforms in use today, SAP IS-U, Oracle CC&B, and Cayenta CIS, will shape billing operations, customer service, and data architecture for a decade or more. This guide is written for the procurement and IT leadership team that needs a structured evaluation framework, not a vendor brochure. For a broader view of how these platforms fit the technology landscape, see modern software solutions transforming the utilities industry. This guide focuses on the buyer’s decision process. ## Define the Functional Scope Before Evaluating Platforms The scope of a utility software replacement typically spans: customer information management (accounts, premises, service points), rate and tariff configuration, billing and revenue cycle, customer payments and FI-CA collections, customer self-service (portal, mobile app), AMI integration and interval data handling, and work and field service management. Each of these domains can be met by different combinations of products. SAP IS-U covers billing and FI-CA natively; it integrates with SAP PM for field work and SAP’s customer portal for self-service. Oracle CC&B (C2M) includes field service and portal capabilities through the Oracle Utilities suite. Cayenta CIS from Harris Computer comes with Cayenta’s own work management module and offers Silverblaze (portal), ServiceLink (field service), and SmartWorks (meter data management) as add-ons from sister Harris products, keeping the stack within a single vendor family. Defining which modules you need from a single vendor versus which you will integrate from best-of-breed sources shapes the short-list immediately. ## The Three Major Platforms SAP IS-U / S/4HANA Utilities is the dominant choice for large investor-owned utilities, multi-commodity retailers, and deregulated market participants. The FI-CA engine handles complex contract account scenarios including ERCOT retail settlement. The platform is deep and difficult to implement quickly. Custom ABAP code accumulated over years of an IS-U deployment is the most common source of cost and risk in migration projects. Our SAP IS-U guide covers functional depth and migration considerations. Oracle CC&B / Oracle Customer to Meter is the leading alternative to SAP at large utilities. Oracle’s OUAF framework provides extensibility without custom code at the core layer. The Oracle Utilities suite (MDM, Field Service, Analytics) integrates natively, which reduces the integration surface area compared to assembling a best-of-breed stack. The Oracle Utilities overview addresses the CC&B-to-C2M migration path. Cayenta CIS from Harris Computer serves small to mid-size electric, water, gas, and telecom utilities. It is a genuinely modern codebase with a faster implementation cycle. Cayla AI for customer service automation and Silverblaze for self-service are production capabilities, not roadmap items. The trade-off is less depth in complex deregulated or multi-commodity billing. See our Cayenta CIS comparison for a direct evaluation against the major platforms. ## Key Evaluation Criteria Rate engine flexibility. The billing engine must handle your current rate structures and the rate structures you expect AMI to enable (time-of-use, demand, EV charging rates). Test this with your actual complex tariffs during the evaluation, not simplified scenarios. AMI and MDM integration. Confirm how the CIS consumes validated interval data from your MDM (or the vendor’s bundled MDM). Ask specifically about the VEE status tracking: can the CIS distinguish between estimated and validated reads for billing purposes? FI-CA / revenue cycle depth. For utilities with complex collections, deposit management, and budget billing programs, the financial sub-ledger capabilities of FI-CA in SAP or the equivalent in Oracle OUAF matter more than the UI. Implementation timeline and total cost of ownership. Large platform implementations run long. Build in realistic contingency. Evaluate the vendor’s implementation partner ecosystem and ask for reference contacts at utilities of comparable size and complexity. Upgrade and support commitments. How often does the vendor release updates? What is the forced upgrade timeline? Oracle’s quarterly SaaS update cadence removes flexibility that on-premises customers have; that is a governance question for your organization. ## The Procurement Process A structured RFP process for a utility CIS should include: a documented set of functional requirements (typically 300 to 500 requirements across all modules), a scripted demo scenario using your own rate structures and data, reference checks with utilities currently on the platform, and a total cost of ownership model covering license, implementation, integration, and 10-year support. For independent support in building the requirements catalog or evaluating vendor responses, AvanSaber’s advisory practice works with utilities through the selection process. See also the comprehensive software selection guide for a step-by-step procurement framework. ## Frequently asked questions What is the most important factor when selecting a utility CIS? Rate configurability and billing engine scalability for your current and planned rate structures. A CIS that cannot handle time-of-use or demand rates without custom code becomes a liability as AMI data drives new tariff designs. How long does a utility CIS replacement typically take? Large investor-owned utilities replacing SAP IS-U or Oracle CC&B typically allow four to six years from contract to production go-live. Smaller utilities on Cayenta CIS can complete implementations in 12 to 24 months. Should smaller utilities consider Cayenta CIS over SAP or Oracle? Often yes. Cayenta CIS has a lower implementation cost and faster time-to-value for utilities with under 200,000 accounts and straightforward rate structures. SAP IS-U and Oracle CC&B provide more depth for complex deregulated market or multi-commodity billing. --- ## Renewables and Utility Software: Billing, DERMS, and Forecasting URL: https://www.utilitieslabs.com/the-impact-of-renewable-energy-technologies-on-utility-providers/ # Renewables and Utility Software: Billing, DERMS, and Forecasting AvanSaber Research/June 8, 2023/Updated June 2, 2026/3 min read Renewable energy growth is often discussed as a generation technology question. For utility operations and IT teams, it is primarily a systems integration question. The billing, dispatch, and forecasting systems that were built for one-directional power flow and flat residential rates were not designed for a grid with millions of distributed generation endpoints. This article focuses on what that mismatch requires. ## Net Metering: Where Renewables Hit the CIS First Net metering is the policy mechanism by which customers with rooftop solar receive credit for export to the grid. It sounds straightforward and the physics are simple, but the billing implementation is not. The meter must record both import and export in the same interval. The billing system must retrieve both registers, apply the netting logic for the billing period, calculate the export credit (which may be at retail rate, avoided cost, or a separately approved value-of-solar tariff), and produce a bill that is both accurate and legible. Oracle CC&B handles this through its rates configuration framework. SAP IS-U manages it through the billing master data and rate determination process in FI-CA. Cayenta CIS supports net metering billing for the mid-market utilities it serves. In each case, the capability is real, but the configuration is non-trivial. Utilities that have not invested in updating their rate configuration since their CIS was first deployed often discover that their net metering billing is generating manual exceptions at volume. For utilities evaluating platform options, our comparison of Oracle and SAP for utilities covers rate engine depth in both platforms. ## DERMS and the Grid-Side Integration Distributed energy resource management systems handle what happens upstream of the meter on the grid side: forecasting DER output, sending dispatch signals to aggregated resources, settling with grid operators, and coordinating with the ADMS for real-time grid stability. GE Vernova’s DERMS product and Schneider EcoStruxure Grid are active in this space. Both must exchange data with the CIS for settlement and billing purposes. The integration architecture between DERMS and CIS is not standardized, and utilities deploying DERMS for the first time typically underestimate the integration scope. The DERMS calculates what a customer’s DER resource provided during a demand response event; the CIS must translate that into a billing credit and post it to the account correctly. For a broader view of how smart grid optimization works across these systems, see our smart grid optimization overview. ## Load Forecasting Under High DER Penetration Traditional load forecasting uses historical consumption data, weather data, and economic indicators to predict demand. High DER penetration changes the forecasting problem materially: behind-the-meter solar reduces net load in a way that is correlated with irradiance, not with the consumption patterns that traditional models were trained on. Load forecasting errors increase grid balancing costs and, in deregulated markets like ERCOT, can create procurement imbalances. Utilities are addressing this through distribution system state estimation (DSSE) that attempts to infer behind-the-meter generation from smart meter residuals. AMI interval data from Itron or Landis+Gyr meters, fed into the MDM and then into a load research platform, provides the input. But the data pipeline quality constraints apply here as they do everywhere: forecasting model accuracy degrades quickly if AMI read completion rates are low or if MDM processing exceptions are high. ## Community Solar and Subscription Billing Community solar programs allow customers who cannot install rooftop solar to subscribe to a share of an off-site solar array and receive a bill credit. The billing complexity rivals net metering: the utility must track each customer’s subscription share, calculate the generation credit for each billing period based on actual output from the array, and apply the credit against the customer’s consumption bill. This requires CIS features beyond standard residential billing. Oracle CC&B and SAP IS-U can support community solar billing through rate configuration, but implementation projects have been complex enough that some utilities have used separate subscriber management systems alongside their core CIS. Cayenta CIS has deployed community solar billing for customers in its installed base. ## The Honest Integration Challenge The common thread across net metering, DERMS settlement, and community solar is that renewables growth makes the CIS the bottleneck. Generation technology costs are falling; the constraint is whether the billing and customer management infrastructure can handle the resulting rate and program complexity. Utilities that address this proactively, through rate engine modernization and MDM data quality investment, will be better positioned than those that wait for a billing crisis to force the issue. For sustainability-related software capabilities connected to renewable programs, see incorporating sustainability practices in the utilities industry. For an independent assessment of your billing infrastructure readiness, contact AvanSaber. ## Frequently asked questions What billing changes does net metering require? Net metering requires bidirectional meter reads, a netting calculation for the billing period, the application of an export credit rate (which may differ from the retail rate), and a bill presentation that customers can understand. Most legacy CIS configurations require significant rework for this. What is the difference between DERMS and an ADMS? An advanced distribution management system (ADMS) handles real-time grid operations: switching, fault location, outage management. DERMS handles distributed resource dispatch, scheduling, and settlement. GE Vernova and Schneider EcoStruxure build both, and they increasingly interoperate. How does variable renewable generation affect billing system load? High DER penetration increases the number of billing edge cases: export credits, demand charge credits for demand response participants, time-of-use rate calculations, and community solar allocations. Each adds configuration and testing burden to the CIS. --- ## SAP HANA for Utilities: Real-Time Data and S/4HANA URL: https://www.utilitieslabs.com/the-impact-of-sap-hana-utilities-on-energy-efficiency/ # SAP HANA for Utilities: Real-Time Data and S/4HANA AvanSaber Research/July 30, 2023/Updated June 1, 2026/1 min read SAP HANA is the in-memory platform beneath S/4HANA Utilities. For a utility, the practical effect is speed: faster billing runs and real-time analytics on the flood of data that smart meters produce. This guide explains what HANA changes and where it actually helps. ## What HANA changes for a utility - Faster billing. In-memory processing shortens the high-volume billing runs that constrain many utility back offices. - Real-time analytics. Interval and AMI meter data can be analyzed as it arrives, which supports consumption insight, loss detection, and demand analysis. - A simpler data model. S/4HANA Utilities on HANA collapses some of the redundant tables that classic IS-U carried, which reduces reconciliation work. ## Where the efficiency comes from The energy-efficiency angle is mostly downstream of better data. When a utility can see consumption patterns and anomalies in near real time, it can target demand programs, spot meter and network losses, and give customers usage insight that changes behavior. HANA is the platform that makes that analysis fast enough to act on. ## Where this fits HANA matters as part of the move to S/4HANA Utilities. For the platform picture, see the SAP IS-U guide. ## Frequently asked questions What is SAP HANA in a utilities context? SAP HANA is the in-memory database that S/4HANA Utilities runs on. For utilities it means faster billing runs, real-time analytics on metering and consumption data, and a simplified data model compared with classic IS-U on a traditional database. Is SAP HANA the same as S/4HANA Utilities? No. SAP HANA is the database and platform. S/4HANA Utilities is the utility application that runs on it. You adopt HANA as part of moving to S/4HANA Utilities, not as a separate product. --- ## Disaster Recovery for Utility IT: CIS, OMS, and Billing Uptime URL: https://www.utilitieslabs.com/the-importance-of-disaster-recovery-planning-for-utility-providers/ # Disaster Recovery for Utility IT: CIS, OMS, and Billing Uptime AvanSaber Research/June 15, 2023/Updated June 2, 2026/4 min read Utility IT disaster recovery conversations often begin and end with the SCADA/OT environment. That focus is justified given the safety and NERC CIP implications of control system failures. But the business systems layer, customer information systems, billing engines, outage management systems, and ERP financials, carries its own continuity risk that is less discussed and often underfunded. ## What Breaks During a CIS Outage A CIS outage during an active billing cycle stops meter read processing, delays bill generation, prevents payment posting, and takes down the customer portal. For a utility billing several hundred thousand accounts on a monthly cycle, a two-day CIS outage can create a billing backlog that takes weeks to clear. Customer service volumes spike when bills are delayed or customers cannot access account information. Regulatory reporting deadlines may be missed. Outage management system (OMS) downtime has different consequences: field crews lose access to trouble ticket management, customer outage notifications fail, and storm response coordination degrades. Oracle Network Management System (NMS) and other OMS platforms are typically maintained with higher availability targets than CIS because the operational impact is more immediate. But the two systems are increasingly integrated, CIS customer contact data feeds OMS for outage notification, which means an OMS failure during a storm event can create cascading customer communication failures even if the CIS is functional. ## Recovery Time and Recovery Point Objectives for Utility Systems The starting point for any DR plan is defining RPO (recovery point objective, how much data loss is acceptable) and RTO (recovery time objective, how long the system can be down) for each critical system. For a billing system, an RPO of 24 hours is typically achievable through daily database backups to an off-site or cloud repository. An RTO of 4 hours requires active failover infrastructure, a warm standby database and application server that can be promoted without a full restore from backup. SAP IS-U and SAP S/4HANA Utilities deployments often use HANA System Replication for database-level HA and DR. Oracle CC&B deployments on the Oracle Utilities Application Framework can use Oracle Data Guard for database replication. Both approaches require configuration, testing, and ongoing validation, not just initial setup. Cloud-hosted deployments of these platforms shift some infrastructure responsibility to the vendor but do not eliminate the need to define and test RTO/RPO targets. ## Ransomware and the Utility Sector Ransomware has targeted utility companies specifically because of the exposure created by OT/IT convergence, attackers can threaten both operational systems and customer data simultaneously. The Colonial Pipeline incident in 2021 demonstrated that even indirect IT compromise can force operational shutdowns as a precautionary measure. NERC CIP standards (CIP-002 through CIP-014) govern cybersecurity for bulk electric system assets. CIS and billing systems that are air-gapped from OT may not fall within CIP scope directly, but they share the same network perimeter in many utility environments. Immutable backups, backup copies that cannot be overwritten or deleted by ransomware, stored off-site or in a write-once cloud tier, are the most practical DR control for ransomware scenarios. Testing recovery from these backups against realistic data volumes should be a scheduled annual exercise, not a theoretical checkbox. ## Business Continuity, Not Just Technical Recovery Technical recovery restores systems. Business continuity keeps operations running while systems are being restored. For utility billing, this means documented manual processes for high-priority functions: emergency payment processing, outage reporting via IVR without OMS integration, and billing holds for customers affected by declared disasters. Cayenta CIS, Oracle CC&B, and SAP IS-U all have provisions for billing holds, estimated reads, and deferred payment processing. These capabilities need to be configured and tested before a crisis, not discovered during one. Customer communication protocols, what to say when the portal is down, how to handle high call volumes, what self-service is available, should be part of the business continuity plan alongside the technical recovery documentation. ## Integrations Multiply DR Complexity Modern utility IT environments are heavily integrated: AMI head-end to MDM, MDM to CIS, CIS to OMS, CIS to ERP financials, ERP to NERC compliance reporting. A DR plan that recovers the CIS but not the MDM feed means billing cannot process new meter reads. A plan that recovers Oracle CC&B but not the Oracle Financials integration means billing revenue is not posting to the general ledger. Integration recovery sequencing should be mapped explicitly. The most common DR planning gap is not the failure to plan for individual system recovery but the failure to define the order in which interdependent systems must be brought back online and the manual bridges that must be maintained in the interim. For a broader view of how platform modernization affects operational resilience, see implementing digital transformation in the utilities sector. For implementation and vendor management best practices, see our implementation best practices guide. If you want a DR readiness review of your utility’s critical IT systems, contact AvanSaber. ## Frequently asked questions Does NERC CIP apply to CIS and billing systems? NERC CIP standards primarily target bulk electric system assets and high-impact control systems. However, utilities should assess whether their CIS has connectivity to operational technology systems that fall within CIP scope, particularly as AMI and OMS integrations deepen. What is an acceptable RTO for a utility CIS? Recovery time objectives vary by utility size and regulatory environment. Many large utilities target sub-4-hour RTO for their CIS during a billing cycle. Smaller utilities may accept longer windows but should have manual billing workarounds documented. What is the difference between DR and high availability? High availability (HA) prevents downtime through redundant components (database clustering, load balancing). Disaster recovery addresses scenarios where the primary environment is unavailable entirely, data center failure, ransomware, major infrastructure loss. Both are needed. --- ## Automation in Utility Meter-to-Cash: CIS, RPA, and AI-Driven Savings URL: https://www.utilitieslabs.com/the-role-of-automation-in-reducing-utility-costs/ # Automation in Utility Meter-to-Cash: CIS, RPA, and AI-Driven Savings AvanSaber Research/June 21, 2023/Updated September 24, 2026/4 min read Automation in the utility industry is discussed at two levels that rarely connect: the high-level vision (AI-powered everything) and the tactical reality (manual billing exceptions, paper checks, field crew dispatching on spreadsheets). The operational cost reductions available to most utilities sit squarely in the middle, in the meter-to-cash process, back-office collections, and field work order management, and they do not require AI. They require better use of the CIS and ERP platforms utilities already own. ## Meter-to-Cash: Where Manual Work Accumulates The meter-to-cash chain runs from AMI read collection through MDM processing, billing calculation, bill delivery, payment posting, and collections. Each transition point is a potential accumulation site for manual exceptions. Common manual touchpoints include: AMI reads that fail validation and require estimated reads, billing holds triggered by data discrepancies, returned mail from paper bills sent to wrong addresses, and payment files that do not match open billing documents. Oracle CC&B has workflow automation built into its billing cycle management. SAP IS-U handles this through the billing master data configuration and FI-CA dunning runs. Both platforms can automate the progression of accounts through estimated read, billing hold, and dunning states, but only if the configuration is current and the underlying meter data quality is high enough to keep exception volumes manageable. The practical starting point for automation ROI is measuring exception volumes at each process step. A utility generating 15,000 manual billing exceptions per month has an automation opportunity before it needs AI. Fixing the AMI read completion rate and MDM validation rules may eliminate most of those exceptions without new software. ## Collections and Dunning Automation Collections is one of the highest-cost back-office functions in a utility CIS environment. Account representatives spend significant time on outbound contact for delinquent accounts, payment arrangement setup, and disconnection order management. SAP IS-U’s FI-CA module automates dunning level progression, payment arrangement offer generation, and disconnection scheduling based on configurable rules. Oracle CC&B provides similar functionality through its collections configuration. Cayenta CIS, used across a broad mid-market utility base, includes collections workflow automation and integrates with DataVoice interactive voice response (IVR) for automated outbound delinquency notifications and inbound payment processing. IVR-based payment capture alone can deflect a meaningful percentage of inbound collections calls at lower cost than agent-handled calls. Robotic process automation (RPA) is useful where the CIS and the financial system do not have a native integration, for example, posting payment files from a third-party payment processor into Oracle CC&B when no API connection exists. RPA is a bridge technology; it should not substitute for building the proper integration, but in environments with budget-constrained IT, it can reduce manual rekeying immediately. ## Field Operations Automation Work order management for field service, connect and disconnect orders, meter exchanges, service investigation requests, is another high-cost manual area. Cayenta Work Management, the ServiceLink mobile workforce add-on Cayenta offers, and similar field service management tools route work orders to mobile field crews, capture completion status, and feed results back to the CIS for account updates. When a field crew completes a meter exchange, the new meter serial number and the old meter’s final read must flow back to billing correctly and promptly. Manual hand-offs here create final-read billing errors that are expensive to resolve. Esri ArcGIS integration with field work order systems gives dispatchers the spatial context needed to optimize crew routing and prioritize work within outage zones. This integration is most mature in electric utilities with OMS deployments, but water utilities are increasingly adopting similar models. ## AI-Assisted Automation: Realistic Applications AI is most credibly applied in utility automation for two specific use cases: anomaly detection in billing data (flagging accounts with usage patterns that suggest meter malfunction or account fraud) and predictive collections scoring (ranking delinquent accounts by likelihood of self-cure versus requiring agent contact). Both use cases require historical billing and payment data, which the CIS already holds, and can be implemented as analytics layers on top of Oracle CC&B, SAP IS-U, or Cayenta CIS without replacing the billing engine. Cayenta’s Cayla AI component specifically targets customer service automation, call deflection through self-service and AI-assisted agent response. This is a realistic near-term cost reduction for utilities with high inbound call volumes relative to their customer count. ## Where to Start The automation priority sequence for most utilities: first, reduce meter read exceptions through AMI data quality improvement; second, configure the collections and dunning workflows already available in the installed CIS; third, integrate field work order completion back to billing automatically; and fourth, layer in AI analytics once the underlying data is reliable. For the broader context of how these automation gains fit into a digital transformation program, see implementing digital transformation in the utilities sector. For platform-specific capabilities, see our utility billing ERP and SAP IS-U pillar pages. For an assessment of automation opportunity in your specific meter-to-cash environment, contact AvanSaber. ## Frequently asked questions What is meter-to-cash automation? Meter-to-cash refers to the full process from meter read collection through billing, payment processing, and collections. Automation targets the manual exception handling steps within that chain, estimated read correction, billing holds, payment plan setup, and delinquency workflows. Where does RPA fit in utility back-office automation? Robotic process automation is most useful for high-volume, rule-based tasks that span multiple systems without a direct API integration, for example, moving data between a legacy billing system and a CRM, or processing batch payment files from a bank. Does SAP IS-U support automated collections workflows? Yes. SAP IS-U's FI-CA module includes dunning and collections workflow automation. The depth of automation depends on how the collections process was configured during implementation and whether the dunning levels have been maintained as policy changes. --- ## Utility Data Platforms: AMI, SCADA, and CIS Data at Scale URL: https://www.utilitieslabs.com/the-role-of-big-data-in-improving-utilities-efficiency/ # Utility Data Platforms: AMI, SCADA, and CIS Data at Scale AvanSaber Research/June 7, 2023/Updated June 2, 2026/3 min read A large electric utility generating 15-minute interval reads from millions of smart meters produces data volumes that dwarf most enterprise systems. The challenge is not storage; cloud and on-premises warehouses handle the volume. The challenge is governance: ensuring that the right data, at the right quality level, reaches the right application without mixing validated billing data with raw, unvalidated telemetry. ## The Data Sources in a Utility Environment Understanding the data problem starts with mapping the sources: AMI interval data. Electric meters from Itron and Landis+Gyr report consumption every 15 minutes. Before this data is used for billing, it must pass through the MDM (Oracle MDM or SAP MDG-U) for VEE: validation, estimation, and editing. Downstream of VEE, the data is fit for the CIS. Upstream of VEE, it is raw and should not be used for revenue calculations. SCADA telemetry. Substations and transmission assets generate analog measurements (voltage, current, power flow) and digital status points (breaker positions, transformer tap positions) at one-to-four-second intervals. This data belongs in the historian (OSIsoft PI is the dominant platform) and the EMS/ADMS, not in the billing system. CIS transactional data. Oracle CC&B, SAP IS-U, and Cayenta CIS store customer accounts, rate schedules, billing cycles, payment history, and service orders. This is structured relational data with a well-defined schema. Analytical queries against production CIS databases create performance risk; a reporting replica or data warehouse extract is standard practice. GIS and network model data. Asset attributes, network topology, and service point locations are used in analytics for loss analysis, load forecasting by feeder, and capital planning. ## The Architecture: From Source to Analytics A typical utility data architecture routes each source to an appropriate staging layer before analytics: The MDM provides validated interval data to a time-series database or data warehouse. Cloud platforms on AWS and Azure both host large-scale time-series workloads. SAP HANA, used in SAP S/4HANA Utilities deployments, provides in-memory analytics against IS-U billing data and FI-CA financial records without requiring a separate warehouse for those domains. Historian data from SCADA remains in the OSIsoft PI ecosystem for operational analytics (frequency deviations, transformer loading trends) and can be replicated to a cloud data lake for longer-term capacity planning queries. CIS data typically replicates nightly to a data warehouse where it can join with AMI reads for customer analytics: identifying high-consumption outliers, flagging potential meter faults, and supporting demand response targeting. ## Practical Use Cases Non-technical loss detection. Comparing transformer load (from SCADA) against the sum of meter reads on that transformer’s secondary (from AMI) identifies distribution losses. Gaps beyond expected technical loss levels indicate metering problems or theft. This analysis runs at scale on a data platform, not inside the CIS. Rate design analytics. Utilities designing time-of-use or tiered rates need interval data by customer class. That analysis requires joined CIS and AMI data in a warehouse environment. SAP IS-U supports rate simulation tools, but the underlying consumption analytics typically happen in a separate layer. Demand forecasting. Accurate short-term load forecasting for the next 24 to 48 hours requires historical interval data, weather inputs, and calendar data. This feeds economic dispatch and wholesale market scheduling. In ERCOT, the Texas deregulated market, load forecast accuracy directly affects settlements. ## Data Governance: The Underinvested Layer Governance is the most frequently skipped investment in utility data programs. The specific failure mode is mixing unvalidated AMI reads with validated billing data in a shared analytics layer. An analyst who pulls “consumption data” without knowing its VEE status will produce incorrect results. A data catalog that tags data by source, VEE status, and intended use is a prerequisite for reliable analytics. Access control matters too. CIS data contains customer personal information and financial records subject to state privacy regulations and tariff confidentiality requirements. A single analytics platform handling both SCADA telemetry and CIS customer data needs role-based access that mirrors the separation required by policy. See the modern software solutions overview for how platform vendors are approaching this in their cloud offerings. For architecture guidance on your specific CIS or ERP environment, AvanSaber can help. ## Frequently asked questions What are the main data sources in a utility data platform? AMI interval reads (15-minute electricity consumption), SCADA telemetry from substations and pipelines, CIS customer and billing records, and GIS network model data. How does SAP HANA fit into utility data management? SAP S/4HANA Utilities stores and analyzes IS-U billing and FI-CA financial data in-memory. The HANA database also supports analytical queries across large interval data sets faster than older Oracle or SQL Server backends. What is the biggest governance challenge in utility data programs? Data lineage: knowing whether an interval read has been through VEE in the MDM before it reaches an analytics layer. Using unvalidated reads for revenue analysis or demand forecasting produces incorrect results. --- ## OT/IT Cybersecurity for Utilities: NERC CIP and Beyond URL: https://www.utilitieslabs.com/the-role-of-cybersecurity-in-protecting-utility-infrastructure/ # OT/IT Cybersecurity for Utilities: NERC CIP and Beyond AvanSaber Research/June 8, 2023/Updated June 2, 2026/3 min read Utility cybersecurity operates in two largely separate regulatory and technical domains: the operational technology (OT) world governed by NERC CIP for electric systems, and the enterprise IT world that covers CIS, ERP, and customer data. The challenge for security programs is that OT and IT are increasingly connected in ways the original compliance frameworks did not anticipate. ## NERC CIP: The Mandatory Floor for Electric OT NERC CIP standards apply to bulk electric system assets in North America. They require utilities to categorize cyber assets by impact level (high, medium, low), implement access management and multi-factor authentication for high and medium impact systems, maintain patch management programs with defined timelines, perform security event monitoring, and report incidents to E-ISAC within mandatory timeframes. Substations operating above defined voltage thresholds, control centers running EMS or ADMS, and associated communication networks are the assets in scope. IEC 62443 provides a complementary framework for industrial control system security that many utilities use alongside CIP for their OT security programs. NERC CIP compliance is a legal obligation with financial penalties for violations. It is not an aspirational framework. Utilities must document compliance evidence and submit to NERC regional entity audits on a defined cycle. This is operationally different from ISO 27001 or NIST CSF, which are voluntary. ## The CIS and ERP Security Posture Oracle CC&B, SAP IS-U, and Cayenta CIS are enterprise IT applications, not NERC CIP cyber assets in the typical deployment. Their security is governed by standard IT controls: role-based access, audit logging, patch management, and data encryption. State public utility commissions increasingly impose data protection requirements on customer information held in the CIS, particularly in states with strong consumer privacy laws. SAP IS-U implementations on SAP S/4HANA benefit from SAP’s Security Audit Log (transaction SM20) and access governance tools like SAP GRC Access Control for segregation-of-duties enforcement. Oracle CC&B deployments use Oracle Identity Governance and database vault features to restrict access to billing and payment data. Our SAP IS-U guide and Oracle Utilities overview cover access control configurations in more detail. FI-CA (Contract Accounts Receivable and Payable) in SAP IS-U handles payment processing. Any system touching payment card data also falls under PCI DSS, which adds its own controls on top of the IT security baseline. ## OT/IT Convergence: Where the Risk Is Concentrated AMI creates a direct data path from the field into the IT environment. The AMI head-end sits between the meter network (OT-adjacent) and the MDM (IT). If an attacker compromises the head-end, they can tamper with billing data flowing to the CIS and potentially reach distribution automation commands if network segmentation is inadequate. ADMS platforms like GE Vernova GridOS and Schneider EcoStruxure Grid increasingly run on virtualized infrastructure that shares compute environments with IT systems. The network boundary between the ADMS and the enterprise is a high-priority segmentation point. Defense-in-depth means the ADMS resides in a dedicated security zone with firewall inspection of all traffic crossing the IT boundary, regardless of whether CIP formally requires it. ## Incident Response for the Utility Context Utility incident response plans must account for both IT (CIS data breach, ransomware) and OT (SCADA compromise, FDIR system unavailability) scenarios. NERC CIP requires documented incident response plans and periodic testing for in-scope assets. For the IT side, most utilities align with NIST SP 800-61 (Computer Security Incident Handling Guide). The E-ISAC (Electricity Information Sharing and Analysis Center) and WaterISAC provide sector-specific threat intelligence that supplements commercial threat feeds. Both are worth subscribing to independent of whether a utility is NERC CIP-obligated. ## What a Practical Security Program Looks Like The baseline for any utility should include: network segmentation between OT zones (as defined in NERC CIP or IEC 62443), multi-factor authentication for all privileged access to CIS, ERP, and SCADA, patch management with documented timelines by system criticality, backup and recovery testing for core systems (CIS, ERP, MDM), and annual tabletop exercises for ransomware and OT intrusion scenarios. The modern software transformation that utilities are pursuing in cloud and AMI also expands the attack surface. Security architecture reviews should accompany every major platform migration. For a gap analysis of your current OT/IT security posture against NERC CIP requirements, contact AvanSaber. ## Frequently asked questions What is NERC CIP and which utilities must comply? NERC CIP (Critical Infrastructure Protection) is a mandatory reliability standard in North America requiring electric utilities to implement access management, patch management, physical security, and incident reporting for bulk electric system cyber assets. Is the CIS (billing system) a NERC CIP asset? Typically no. CIS platforms like Oracle CC&B and SAP IS-U are IT systems, not bulk electric system cyber assets under CIP definitions. They fall under standard IT security programs and state PUC data protection requirements instead. What is the biggest OT/IT convergence risk in utility security? Poorly segmented networks where an IT intrusion can pivot to OT. AMI head-ends and field device management platforms often sit in a middle zone between IT and OT, making them high-priority segmentation points. --- ## SAP IS-U for Utilities: Modules, S/4HANA Utilities Migration, and Fit URL: https://www.utilitieslabs.com/the-role-of-sap-utilities-in-streamlining-utility-management/ # SAP IS-U for Utilities: Modules, S/4HANA Utilities Migration, and Fit AvanSaber Research/June 8, 2023/Updated June 1, 2026/1 min read SAP IS-U is the utility-industry solution that turns generic SAP ERP into a system that can read meters, rate consumption, and bill customers. This guide explains the modules, how IS-U relates to S/4HANA Utilities and Cloud for Utilities (C4U), and when IS-U is the right system of record. ## What SAP IS-U actually covers IS-U adds the utility-specific processes that core ERP does not have: - Device management. Meter and device records, installations, and the technical view of the connection. - Billing and invoicing. Rate structures, billing runs, and invoicing for electric, gas, water, and district heating. - FI-CA (Contract Accounts Receivable and Payable). The high-volume subledger that handles utility payments, dunning, and collections at a scale generic accounts receivable cannot. - Customer service. The customer and contract view for service orders, moves, and disputes. ## IS-U, S/4HANA Utilities, and C4U These three names cause most of the confusion in utility SAP projects: | Name | What it is | SAP IS-U | The classic utilities solution on SAP ECC | S/4HANA Utilities | The IS-U functional scope re-platformed on S/4HANA and HANA | C4U (Cloud for Utilities) | The newer cloud-oriented utilities model The key point: S/4HANA Utilities is not a rip-and-replace of IS-U. It carries the same functional core onto a faster database and a simplified data model with Fiori interfaces. ## When IS-U is the right call IS-U and S/4HANA Utilities fit utilities that are already standardized on SAP, that want billing and finance in one estate, and that need the depth IS-U brings to high-volume metered billing. For a head-to-head against the main alternative, read the Oracle vs SAP for utilities comparison, and for the mid-market alternative, the Cayenta CIS review. ## Frequently asked questions What does SAP IS-U do? SAP IS-U is SAP's industry solution for utilities. It handles meter and device management, billing and invoicing through FI-CA contract accounts, and customer service for metered and non-metered utility services, sitting on SAP ECC. What replaces SAP IS-U? S/4HANA Utilities carries the IS-U functional scope forward onto the S/4HANA platform. It is a re-platforming with a simplified data model and Fiori apps, not a different product. Cloud for Utilities (C4U) is the newer cloud-oriented model. When should a utility migrate from IS-U to S/4HANA Utilities? Plan around SAP maintenance timelines (mainstream maintenance for the Business Suite runs to 2027 with extended maintenance to 2030) and your own upgrade cycle. The migration is a program, so most utilities scope it as a multi-year roadmap rather than a single cutover. --- ## AR and VR with Oracle Utilities ERP: Honest Scope and Use Cases URL: https://www.utilitieslabs.com/the-transformation-of-oracle-erp-systems-with-ar-and-vr/ # AR and VR with Oracle Utilities ERP: Honest Scope and Use Cases AvanSaber Research/January 30, 2024/Updated June 2, 2026/3 min read Augmented reality and virtual reality have attracted genuine interest across enterprise software, and Oracle ERP environments are no exception. For utilities specifically, the relevant Oracle products are Customer Care and Billing (CC&B), Oracle Utilities Network Management System (NMS), and Oracle Utilities Mobile Workforce Management (MWM). Understanding where AR and VR add realistic value in those contexts requires separating the demonstration use cases from the production-ready ones. This post is a narrow, honest assessment of that landscape. For a broader discussion of how AR intersects with ERP financial and AR modules, see Enhancing ERP: Integrating AR with SAP. For where VR genuinely fits in utility ERP and field-training workflows, see VR for utility workforce training. ## Where AR Has a Plausible Fit in Oracle Utilities Operations The strongest case for AR in an Oracle Utilities context is field operations. Technicians dispatched through Oracle MWM carry work orders, asset records, and switching instructions derived from Oracle NMS. AR glasses or mobile AR applications could surface that structured data in the technician’s field of view without requiring them to shift attention to a tablet. Specific scenarios that have been piloted at utilities include: - Substation switching: Overlaying switching sequence steps and isolation boundary markers on physical switchgear, reducing the risk of wrong-device operations. - Underground cable location: Displaying GIS asset records and cable routing from Oracle’s network data model in context with physical dig sites. - Meter inspection: Surfacing account and historical read data from CC&B against a scanned meter serial number during field investigation. None of these are available as packaged Oracle features. They require a custom AR application layer that authenticates against Oracle Utilities APIs, handles offline scenarios for areas with poor connectivity, and meets the safety-critical reliability standards that field operations demand. ## VR for Training and Simulation VR has a more contained use case in Oracle Utilities ERP contexts: training. Building a VR simulation of a control room that presents synthetic NMS switching scenarios allows operators to practice restoration procedures without risk to live assets. Similarly, billing process training in a VR environment could walk new staff through CC&B workflows in a guided, consequence-free setting. The honest constraint here is development cost relative to alternatives. A well-configured CC&B sandbox with realistic test data and guided exercises delivers comparable training outcomes at a fraction of the cost of a custom VR simulation. VR investment is more defensible where physical simulation has safety value, such as high-voltage switching or confined space procedures, than for office-based billing workflows. ## Data Visualization in 3D Oracle Utilities datasets, particularly network topology and geospatial outage data from NMS, are candidates for 3D visualization. Viewing a feeder segment and its downstream customers in a spatial model can aid restoration planning and infrastructure investment decisions. This falls more precisely into the XR data visualization space covered in a companion post on XR for ERP data visualization. The visualization layer does not require AR headsets; it can run in a browser-based 3D rendering environment connected to Oracle Analytics Cloud or OUAW, which lowers the barrier considerably compared to device-dependent AR deployments. ## Integration Complexity Is the Limiting Factor The fundamental constraint on AR and VR adoption in Oracle Utilities environments is integration complexity, not device availability. CC&B and OUAF expose data through REST APIs and web services, but those APIs were designed for transactional CIS use, not low-latency real-time data feeds that AR experiences require. Field connectivity gaps, security and compliance requirements for utility SCADA-adjacent systems, and the high cost of maintaining custom AR applications through CC&B patch cycles all weigh against broad deployment. For project teams evaluating AR and VR investments alongside an Oracle Utilities programme, the honest advice is to pilot a single, high-value field operations use case with a specific measurable outcome before committing to a platform-level investment. If you are assessing the technology landscape for an Oracle Utilities programme, AvanSaber’s utility ERP practice can help scope what is realistic given your product version and operational context. ## Frequently asked questions Can AR overlay Oracle Utilities data in a field service context? Yes, in principle. AR devices can surface work order, asset, and network topology data from Oracle Network Management System or Oracle Utilities Mobile Workforce Management during field operations, but this requires a purpose-built AR application that calls the Oracle APIs. Is VR useful for Oracle ERP training? VR-based simulations for billing process training or switching procedure walkthroughs are technically feasible but represent a significant development investment. Most utilities find that guided sandbox environments in CC&B are more cost-effective for the same training outcome. Does Oracle offer native AR/VR tools for its Utilities suite? Oracle does not ship native AR or VR interfaces with CC&B or OUAF. AR and VR integration requires custom development or third-party platforms that connect to Oracle Utilities APIs. --- ## AI Software for Utility Companies: 2026 Buyer Guide by Use Case URL: https://www.utilitieslabs.com/top-ai-software-for-utility-companies/ # AI Software for Utility Companies: 2026 Buyer Guide by Use Case AvanSaber Research/July 21, 2024/Updated September 11, 2026/2 min read Searches for “AI software for utility companies” return long vendor lists that blur very different tools together. The useful way to read the market is by the job the software does, because no single product covers them all and the buying decision differs for each. This page is the vendor view, organised by category. If the question behind your search is the broader one, where AI in utilities is actually in production rather than which product to buy, start with AI in utilities: where it actually lands and come back here when you are choosing between tools. ## Customer energy analytics This is the most mature category: software that analyzes consumption and nudges customers. Oracle Opower is the established name for behavioral energy analytics and home energy reports, Bidgely specializes in appliance-level load disaggregation from smart-meter data, and Uplight focuses on customer engagement and demand-side programs. These layer on top of the billing and meter data your CIS already holds. ## Grid and distributed-energy management As rooftop solar, batteries, and EVs grow, utilities run AI for grid operations and distributed-energy resource management (DERMS). GE Vernova’s GridOS and ADMS, Schneider Electric EcoStruxure ADMS, and AutoGrid (now part of Schneider) sit here, alongside Oracle Utilities Network Management for outage and distribution management. This is operational technology, a different buyer and budget from the customer-facing analytics above. ## AMI and meter-data analytics The meter vendors carry their own analytics. Itron and Landis+Gyr both offer AMI head-end and meter-data analytics for loss detection, outage insight, and consumption patterns. These feed, and often integrate with, the meter data management (MDM) layer of your CIS. ## AI inside the core systems The systems of record are adding their own AI. SAP offers Joule and BTP AI services that apply to IS-U and S/4HANA Utilities for assistance and automation; Oracle is embedding AI into CC&B and the Energy and Water suite; and Cayenta leads with Cayla AI inside its CIS, covered in the Cayenta CIS review. For where AI genuinely helps on the core billing and operations stack, see AI for utility operations on SAP IS-U and Oracle CC&B. ## How to choose Start from the job, not the vendor list. Decide whether the problem is customer engagement, grid operations, meter analytics, or automation inside the system of record, then shortlist within that category. Keep the rule that matters most: the CIS and billing platform stays the system of record, and AI software reads from it and proposes actions rather than replacing it. For platform selection itself, the Oracle vs SAP comparison and the utility software selection guide are the starting points. ## Frequently asked questions What AI software do utilities actually use? It splits by job: customer energy analytics (Oracle Opower, Bidgely, Uplight), grid and distributed-energy management (GE GridOS / ADMS, Schneider EcoStruxure ADMS, AutoGrid), AMI and meter analytics (Itron, Landis+Gyr), and AI inside the core systems (SAP Joule and BTP AI for IS-U, Oracle's AI in CC&B). There is no single 'utility AI' product; you assemble by use case. Should AI replace the utility CIS or billing system? No. SAP IS-U, Oracle CC&B, and Cayenta CIS remain the systems of record. AI software sits alongside them for analytics and automation, reading data out and writing back recommendations a person or controlled workflow approves. --- ## AI and ERP Implementation for Utilities: A 2026 Practitioner Checklist URL: https://www.utilitieslabs.com/ultimate-guide-ai-erp-implementation-tips-for-utilities-2025/ # AI and ERP Implementation for Utilities: A 2026 Practitioner Checklist AvanSaber Research/August 6, 2025/Updated June 1, 2026/1 min read Most “ultimate guide” ERP checklists could apply to any industry. A utility needs a different one, built around meter-to-cash and regulated billing. This is the utilities-specific checklist for an AI and ERP implementation in 2026. ## Scope around meter-to-cash The utility revenue process is the spine of the program: - Meter reading and data validation, including estimated reads. - Rating, including tiered, time-of-use, and seasonal rates. - Billing and invoicing at high volume. - Payment, collections, and dunning through a contract-account subledger (FI-CA in SAP). If the implementation plan does not center on these, it is a generic plan wearing a utility label. ## Build for rate cases and regulation Regulated utilities live with rate cases. The system has to let you model and launch new rates, and keep the audit trail regulators expect. Scope rate complexity early, because it multiplies test cases more than any other factor. ## Integrate the data layer Plan the meter data management and CIS integration as first-class work, not an afterthought. Inbound reads, device events, and market messages (often via SAP IDocs) are where high-volume billing succeeds or fails. ## Add AI after the core works Get the CIS, billing, and meter data stable first. Then add AI to specific tasks: exception triage, estimated-read quality, and forecasting. For where AI fits, see AI for utility operations on SAP IS-U and Oracle CC&B. ## Where this fits For platform selection before you implement, start with the Oracle vs SAP comparison. ## Frequently asked questions What is different about ERP implementation for a utility? Utility ERP centers on meter-to-cash: metered consumption, tiered and time-of-use rates, regulated rate cases, estimated reads, and high-volume billing through a contract-account subledger. A generic ERP rollout plan misses these, so the implementation has to be scoped around the utility revenue process. Where does AI fit in a utility ERP program? After the core works, not during the core build. Get the CIS, billing, and meter data right first, then add AI to specific tasks like exception triage and forecasting. Bolting AI onto an unstable core adds risk without value. --- ## Cloud for Utility Software: S/4HANA, Oracle, and CayentaCloud URL: https://www.utilitieslabs.com/understanding-the-benefits-of-cloud-computing-in-utilities-management/ # Cloud for Utility Software: S/4HANA, Oracle, and CayentaCloud AvanSaber Research/June 7, 2023/Updated September 24, 2026/3 min read Cloud deployment for utility CIS and ERP is available from all three major platform vendors, but the migration path is not simple for most utilities. Regulatory data residency requirements, the complexity of rate configuration and CIS data migration, and the long asset life of operational technology create constraints that don’t apply to most enterprise cloud migrations. ## SAP S/4HANA Utilities in the Cloud SAP’s RISE with SAP offering packages S/4HANA Utilities in a managed cloud model. SAP handles the underlying infrastructure, database, and basis administration. The customer retains responsibility for functional configuration, custom enhancements, and integration with external systems (AMI head-end, MDM, GIS, OMS). The migration from SAP IS-U on ECC to S/4HANA Utilities is a significant project regardless of whether it runs in the cloud or on-premises. Rate schedules, IDoc interfaces, custom ABAP, and FI-CA payment configurations all require assessment. SAP’s Transformation Navigator tool helps map IS-U functionality to S/4HANA equivalents, but there are functional gaps, particularly around complex tariff structures used in deregulated markets like ERCOT. Utilities in those markets need a detailed fit-gap analysis before committing to a cloud migration timeline. Our SAP IS-U guide covers the functional landscape in detail. ## Oracle Energy and Water Cloud Oracle’s cloud path for CC&B (now branded Oracle Customer to Meter or C2M) is hosted on Oracle Cloud Infrastructure. It includes the Oracle Utilities Application Framework (OUAF), which underpins CC&B, Oracle MDM, and Oracle Analytics for Utilities. Oracle publishes a quarterly update cadence for the cloud product, which removes the ability to defer patches the way on-premises customers can. For utilities running Oracle CC&B on-premises with heavily customized code, the cloud migration requires converting those customizations to configuration or rebuilding them within OUAF extension frameworks. This is the most common source of timeline and budget overruns in Oracle cloud migrations. Our Oracle Utilities review discusses the customization challenge in context. ## CayentaCloud Harris Computer’s Cayenta CIS offers a cloud-hosted option for smaller and mid-size utilities that cannot support dedicated on-premises infrastructure. That option, CayentaCloud, is Cayenta’s single-tenant SaaS deployment. The Cayla AI assistant, Cayenta’s own, handles customer service automation, while Silverblaze (customer self-service portal) and ServiceLink (field service) are separate sister Harris products that Cayenta offers as add-ons. The hosting model reduces the infrastructure management burden, which is significant for municipal utilities without large IT teams. For those evaluating Cayenta against the major platforms, the Cayenta CIS comparison provides a direct assessment of where it fits in the market. ## Data Residency and Regulatory Constraints This is the most significant caveat in utility cloud migrations that generic cloud ROI analyses ignore. State PUC tariffs in some jurisdictions include provisions about where customer usage data can be stored and processed. Some states have enacted consumer privacy laws that impose additional requirements on utilities holding energy consumption data, which can identify household occupancy patterns and behavior. Multi-tenant cloud regions that span geographic boundaries may not satisfy these requirements without explicit contractual data processing agreements. NERC CIP requirements affect OT-adjacent systems. An AMI head-end or MDM that is classified as a medium-impact cyber asset under CIP cannot move to a shared public cloud infrastructure without a formal security assessment. Most cloud deployments of these systems use dedicated infrastructure options or remain on-premises. Federal power utilities (TVA, BPA) and utilities with federal contracts face additional data handling constraints under FedRAMP, which requires cloud providers to hold specific authorizations before hosting covered federal data. ## Where Cloud Makes Sense Today Cloud deployment is most straightforward for the customer-facing and analytics layers: customer portals, mobile workforce management, and data warehouses for interval data analytics. These workloads are less constrained by OT security and data residency requirements than the core CIS and MDM. For the core CIS and ERP, cloud migration is viable but requires careful planning of the data residency, customization, and integration challenges. Utilities that are already on modern on-premises versions of SAP IS-U or Oracle CC&B should evaluate cloud primarily when they need to reduce infrastructure management costs or when approaching a major upgrade cycle anyway. For a realistic cloud readiness assessment for your platform, contact AvanSaber. ## Frequently asked questions Is SAP S/4HANA Utilities available as a cloud deployment? Yes. SAP RISE with SAP covers S/4HANA Utilities in a managed cloud model. It runs on hyperscaler infrastructure (AWS, Azure, or GCP) with SAP managing the basis layer. Customers retain functional configuration responsibility. What is Oracle Energy and Water Cloud? Oracle's SaaS path for Oracle CC&B (now Oracle Customer to Meter). It includes Oracle Utilities, CC&B, MDM, and associated modules hosted on Oracle Cloud Infrastructure. What data residency issues affect utility cloud migrations? State PUC tariffs and some state privacy laws require customer usage and billing data to remain within specific jurisdictions. Multi-tenant cloud regions may not satisfy these requirements without contractual data processing agreements or dedicated infrastructure options. --- ## Smart Grids Explained: AMI, ADMS, and DERMS in Utility Operations URL: https://www.utilitieslabs.com/understanding-the-role-of-smart-grids-in-modern-utility-management/ # Smart Grids Explained: AMI, ADMS, and DERMS in Utility Operations AvanSaber Research/June 8, 2023/Updated September 24, 2026/3 min read “Smart grid” functions as an umbrella label for a collection of specific systems that utilities have deployed in layers over the past two decades. Understanding which system does what prevents the kind of confusion where every vendor claims their product is the smart grid. This article maps the principal components and their relationships. ## The Foundation: Advanced Metering Infrastructure AMI is the most widely deployed smart grid component. It replaces manually read meters with communicating meters that report interval consumption data (typically 15-minute or hourly reads) through a communication network to a utility-operated head-end system. The two dominant AMI vendors in North American utilities are Itron and Landis+Gyr. Both supply the full stack: the meter hardware, the radio or cellular communication infrastructure, and the head-end software that manages meter configuration and data collection. The head-end feeds a Meter Data Management System (MDMS) which validates reads and makes clean interval data available to the CIS for billing and to analytics systems for load research. AMI by itself is not a smart grid. It is the data collection layer that makes the rest of the smart grid possible. Without interval data, load forecasting is imprecise, non-technical loss detection is limited, and customer energy management programs lack the granularity to be effective. ## Distribution Automation and the ADMS Distribution automation refers to switching devices, sensors, and communications equipment installed on feeders and substations. The Advanced Distribution Management System (ADMS) is the software that coordinates all of it. The ADMS ingests SCADA telemetry from substations and field devices, maintains a real-time network model, and runs automated control functions including Fault Location Isolation and Restoration (FLISR), Volt/VAR optimization, and restoration switching. GE Vernova GridOS, Schneider Electric EcoStruxure ADMS, and Oracle Utilities Network Management are the three platforms with the broadest North American and European utility deployments. The ADMS network model is built from the GIS, typically Esri ArcGIS, which holds the authoritative record of feeder topology, asset locations, and connectivity. Keeping the GIS model current is an operational discipline that determines how accurate the ADMS analysis functions are. AI models can assist ADMS operations by recommending pre-fault maintenance actions, optimizing switching plans, or forecasting feeder load to improve VVO setpoints. The ADMS retains control authority; AI outputs inform human decisions or pre-authorized automated sequences. ## DERMS: Coordinating Distributed Energy Resources Distributed energy resources, rooftop solar, behind-the-meter batteries, EV chargers, and enrolled demand response participants, are not owned or operated by the distribution utility. Coordinating them requires a Distributed Energy Resource Management System (DERMS). AutoGrid is among the most deployed DERMS platforms. A DERMS aggregates the available flexibility across enrolled DERs and dispatches them as needed to provide grid services: peak shaving, frequency response, or voltage support. The dispatch signals go to the DER owner’s device through an API or the utility customer portal; the ADMS sees the aggregate effect on the feeder. The relationship between the ADMS and a DERMS is a co-ordination architecture. The ADMS controls the utility’s own switching equipment. The DERMS influences customer-sited resources through commercial agreements. Neither replaces the other. ## The CIS and ERP: Where Smart Grid Data Meets the Business System Smart grid systems generate data that has to reach the billing and customer systems to create value. AMI interval data flows to the MDMS and then to the CIS for interval billing, time-of-use rate calculations, and high-bill alerts. ADMS outage records flow to the CIS to support outage restoration notifications and to validate billing adjustments for affected customers. On SAP IS-U, this integration runs through IDocs and BTP integration services. On Oracle CC&B, the OUAF integration framework handles the AMI and OMS data feeds. On Cayenta CIS, the MDM and OMS pieces are add-ons powered by sister Harris products: SmartWorks for meter data management and DataVoice for outage management. For utilities operating in deregulated markets, the smart grid data layer also feeds retail and wholesale market systems. The deregulated market pillar covers how settlement and interval data management work in that context. ## Where to Go Deeper This article is the foundational layer. For AI-assisted operations on the ADMS, see FLISR and self-healing grids. For the analytics that sit on top of AMI and grid data, see AI analytics in utility operations and smart grid optimization. The Oracle Utilities and SAP IS-U pillars cover how these grid systems integrate with the utility’s CIS and ERP. If you are evaluating ADMS platforms or DERMS options, Avansaber provides vendor-neutral advisory. ## Frequently asked questions What is the difference between SCADA and an ADMS? SCADA (Supervisory Control and Data Acquisition) monitors and controls field equipment in real time. An ADMS builds on SCADA telemetry and adds network analysis, load flow modeling, Volt/VAR optimization, and FLISR automation. SCADA is a component; the ADMS is the full distribution management application layer. Which AMI vendors are most common in North American utilities? Itron and Landis+Gyr are the two largest AMI system vendors in North America. Both supply the meters, the communication network (RF mesh or cellular), and the head-end system that delivers interval data to the meter data management system (MDMS) and downstream to the CIS. What does a DERMS do that the ADMS does not? An ADMS manages the utility's own distribution assets: switches, capacitors, voltage regulators. A DERMS coordinates distributed energy resources owned by third parties or customers: behind-the-meter batteries, solar inverters, demand response participants. DERMS platforms like AutoGrid aggregate DER flexibility and pass dispatch signals, while the ADMS maintains grid control authority. --- ## SAP BTP AI and Joule for Utility Operations on IS-U and S/4HANA URL: https://www.utilitieslabs.com/unlocking-business-potential-with-sap-btp-ai/ # SAP BTP AI and Joule for Utility Operations on IS-U and S/4HANA AvanSaber Research/January 10, 2025/Updated June 2, 2026/4 min read SAP Business Technology Platform provides the integration, extension, and AI layer for SAP’s ERP and industry products. For utilities running SAP IS-U on ECC or S/4HANA Utilities, BTP is the designated path for AI-based extensions, integrations with third-party metering and grid systems, and custom applications that need to stay off the core system. The SAP BTP guide on this site covers the platform architecture; this post focuses on its AI capabilities and where they apply in a utility operational context. ## SAP AI Core and AI Launchpad for Custom Utility Models SAP AI Core is the managed runtime on BTP for training, deploying, and serving machine learning models. For a utility, the models worth building on this infrastructure fall into a few categories: - Billing exception classification: Training a model on historical billing exception records from IS-U to classify new exceptions by likely cause (estimated read, rate configuration error, AMI transmission failure) and priority. This reduces the manual triage load on billing operations staff. - Consumption anomaly detection: A model trained on interval meter data that flags accounts with consumption signatures inconsistent with their historical pattern, pointing to meter faults, theft, or connection errors before they reach billing. - Collections propensity scoring: A classification model trained on FI-CA payment history, service agreement attributes, and contact history that scores accounts by likelihood of self-cure, informing collections strategy. These models use IS-U and FI-CA data as their input, which is accessed via BTP’s connectivity services and the SAP Integration Suite rather than through direct database calls. Using AI Core keeps the model lifecycle, retraining, and deployment separate from the SAP system itself, which is consistent with clean-core principles. ## Joule as an Operational Copilot in S/4HANA Utilities SAP Joule is an AI copilot that surfaces inside S/4HANA Fiori experiences. For utility billing and customer operations, the practical near-term use cases are: - Natural-language queries on IS-U data: A billing supervisor asking “show me all invoicing runs in the past 30 days with exception rates above 5 percent” can get an answer through Joule without navigating to a specific IS-U report transaction. - Guided process support: Joule can walk a new billing analyst through the steps to investigate a billing exception or reverse an incorrect posting, reducing the reliance on tribal knowledge documentation. - Drafting customer correspondence: Joule can generate initial drafts of billing dispute responses or service interruption notices based on account data, which an agent then reviews and sends. Joule in S/4HANA Utilities is still maturing. The depth of IS-U-specific coverage, especially for rate schedule maintenance, device management, and FI-CA contract account operations, is less complete than for standard S/4HANA Finance or MM processes. Utilities evaluating the Joule roadmap should request specifics from SAP on IS-U Fiori app coverage before including Joule-driven efficiency in business case calculations. ## Document Processing AI for Utility Use Cases SAP’s Document Information Extraction service on BTP applies machine learning to extract structured data from unstructured documents. For utilities, relevant applications include processing incoming customer correspondence, extracting meter read data from legacy paper or PDF formats during a migration, and parsing third-party contract documents for tariff terms. This service is available without building a custom model and can be integrated with SAP Workflow or BTP Integration Suite. ## Integration Patterns for AI Outputs Back into IS-U A model that scores billing exceptions or flags consumption anomalies has no value until its output drives action in IS-U. BTP Integration Suite handles the return path: writing anomaly flags back as IS-U billing notes, creating FI-CA collections activities for high-propensity delinquency accounts, or triggering field service orders in SAP Work Manager based on meter fault predictions. This integration design is where utility AI projects most commonly stall. The model development gets done; the operational workflow integration that makes model outputs actionable in the IS-U or FI-CA context gets underestimated. Budget and plan the integration work with the same rigour as the model development. ## What to Realistically Expect from BTP AI Today BTP AI infrastructure is production-ready. Joule’s IS-U coverage is still expanding. Custom models via AI Core are viable now but require ML engineering capacity that most utility IT teams do not have in-house. The document processing and pre-built business services are the most accessible entry points. For a broader view of BTP’s role in utility IS-U extensions and integrations, see Unlocking Business Potential with SAP BTP, the SAP IS-U pillar, and the discussion of AI integration across SAP and Oracle utility operations. For an independent assessment of where BTP AI fits in your specific utility SAP programme, AvanSaber works with both IS-U ECC and S/4HANA Utilities environments. ## Frequently asked questions What is Joule in the SAP context? Joule is SAP's generative AI copilot embedded across SAP products. In S/4HANA Utilities and related products it can assist with natural-language queries against business data, guided process support, and automated content generation within SAP Fiori workflows. Which SAP BTP AI services are relevant for utilities? SAP AI Core (for training and deploying custom ML models), SAP AI Launchpad (for managing model lifecycle), and the pre-built AI business services for document processing, translation, and recommendation are the primary BTP AI services utilities can apply to IS-U or billing data. Does Joule work with SAP IS-U on ECC? Joule is a S/4HANA and BTP-native capability. IS-U on ECC does not have native Joule integration; the copilot experience is available for utilities that have moved to S/4HANA Utilities or are accessing SAP functionality through BTP-based extensions. --- ## SAP BTP for Utilities: Extend IS-U, Keep the Core Clean URL: https://www.utilitieslabs.com/unlocking-business-potential-with-sap-btp/ # SAP BTP for Utilities: Extend IS-U, Keep the Core Clean AvanSaber Research/January 16, 2025/Updated June 2, 2026/1 min read SAP Business Technology Platform (BTP) is easy to wave at and hard to pin down. For a utility running SAP IS-U or S/4HANA Utilities, its value is specific: it is where you add everything around the core without modifying the core itself. That discipline is what keeps the platform upgradeable. ## The clean-core reason BTP exists Custom code inside the ERP core is the main reason utility SAP upgrades drag. The clean-core principle says keep the core standard and build extensions side-by-side. BTP is the side-by-side environment: custom apps and logic run on BTP and call IS-U or S/4HANA Utilities through stable APIs, so the core stays clean and the eventual IS-U to S/4HANA Utilities move stays feasible. ## What a utility actually builds on BTP - Extensions. Custom Fiori apps and logic for utility-specific processes that standard IS-U does not cover, without touching the core. - Integration. The SAP Integration Suite connects IS-U and S/4HANA Utilities to meter data management, customer portals, payment processors, and market-message exchange. - Data and analytics. SAP HANA Cloud and SAP Analytics Cloud run reporting and analytics on consumption and billing data without loading the operational system. - AI. BTP AI services and Joule add assistance and automation (for example billing-exception triage) that read from the core and propose actions. ## Where it fits the roadmap BTP is most valuable to utilities planning the move to S/4HANA Utilities, because the clean-core extensions built on it survive the migration. For the platform picture, see the SAP IS-U guide; for the underlying in-memory platform, see SAP HANA for utilities; and for where AI genuinely helps on the core, see AI for utility operations on SAP IS-U and Oracle CC&B. ## Frequently asked questions What is SAP BTP used for in utilities? SAP Business Technology Platform is where a utility builds extensions, integrations, analytics, and AI around SAP IS-U or S/4HANA Utilities without modifying the core. It covers app development, the Integration Suite, HANA Cloud and SAP Analytics Cloud, and AI services including Joule. Why extend on BTP instead of modifying IS-U directly? The clean-core principle: custom code in the ERP core makes upgrades (and the move from IS-U to S/4HANA Utilities) slow and risky. Building side-by-side on BTP keeps the core upgradeable while still letting you add utility-specific apps and integrations. --- ## Utilities ERP & Billing: SAP, Oracle and Cayenta URL: https://www.utilitieslabs.com/utility-billing-erp/ # Utilities ERP & Billing: SAP, Oracle and Cayenta Utilities ERP and billing software turns metered consumption into revenue. This guide maps the meter-to-cash stack, compares SAP IS-U, Oracle CC&B, and Cayenta CIS, and shows buyers which questions decide platform fit. ## What utilities ERP and billing software must do Utilities ERP is not generic invoicing with a meter attached. It must maintain the service location and device relationships, validate consumption, apply regulated rates, create bills at high volume, post receivables, allocate payments, run collections, and preserve an audit trail through every correction. The core sequence is meter-to-cash: - Collect and validate interval or register reads. - Bind consumption to the correct installation, service agreement, and customer account. - Apply the effective tariff, rider, tax, discount, and proration rules. - Create and deliver the bill. - Post receivables, allocate payments, and manage returns, dunning, and collections. - Reconcile the billing platform with finance and the general ledger. A product that cannot explain how it handles estimated reads, meter exchanges, backdated rate changes, move-in and move-out, multi-service accounts, and rebilling is not ready for a utility procurement. ## SAP IS-U vs Oracle CC&B vs Cayenta CIS | Platform | Best-fit context | Billing and CIS posture | Main implementation trade-off | SAP IS-U and S/4HANA Utilities | Large utility or established SAP estate | Deep meter-to-cash, FI-CA receivables, device management, and finance integration | Broad capability with a configuration-heavy program and demanding data conversion | Oracle Utilities CC&B | Large utility or established Oracle Utilities estate | Customer care and billing tied to Oracle MDM, network, and customer products | Strong utilities depth with a substantial implementation and integration footprint | Cayenta CIS | Municipal and mid-market utility seeking an integrated suite | CIS and billing alongside Harris utility products such as MDM, OMS, and customer channels | Smaller suite footprint, with less depth for the largest or most complex operating models The vendor product descriptions are useful starting points, not selection evidence. SAP documents its current utilities scope in the SAP for Utilities portal; Oracle publishes the CC&B documentation; and the Cayenta product catalog shows the modules in its suite. A selection still needs scripted demonstrations using the utility’s own rates, exceptions, and operating scenarios. ## Seven questions that decide platform fit How complex are the rates? Count active rate variants, riders, taxes, proration rules, and exceptions. The count matters less than the combinations the billing engine must evaluate and test. How many commodities and service models share the account? Electric, gas, water, wastewater, district energy, and non-metered charges create different device and billing relationships. Which surrounding systems are already strategic? MDM, OMS, CRM, payment, finance, data platform, and field service integration can decide the platform before a feature score does. What must remain available during conversion? A billing cutover requires parallel validation, operational fallback, and clear ownership for exceptions. The migration design should state how the utility proves a bill is correct, not simply how records are loaded. How much configuration can the operating team sustain? Enterprise breadth is useful only when the utility can test, release, and support the resulting rules after the implementation team leaves. What deployment and security controls are mandatory? Data residency, identity, audit, integration boundaries, recovery, and release governance belong in the shortlist criteria. Where does AI sit? AI can explain bills, triage exceptions, and assist agents, but the tariff calculation, posting, and control decisions must stay deterministic and auditable. The architecture view is in AI across service-to-cash. ## Build the shortlist from operating evidence Turn the seven questions into demonstration scripts. Give each vendor the same account history, meter event, rate change, payment exception, and rebill scenario. Score whether the product completes the workflow, which configuration or customization it needs, what lands in the audit log, and how the result reconciles to finance. Then test the data model. The most expensive failures come from ambiguous ownership of customer, account, premise, service point, contract, and device records. The Oracle CC&B data model guide shows the Oracle relationships; the SAP CIS guide maps the comparable SAP concepts. For the enterprise choice, continue with Oracle vs SAP for utilities. For a municipal or mid-market alternative, use the Cayenta CIS review. ## In this topic - Cayenta CIS review - Oracle vs SAP for utilities ## Everything else in this topic 25 more articles, A to Z. - 5G in the Utilities Sector: Real Uses, Real Limits - AI and ERP Implementation for Utilities: A 2026 Practitioner Checklist - AI in Utility Billing: Where It Helps on IS-U, CC&B, and Cayenta - Automation in Utility Meter-to-Cash: CIS, RPA, and AI-Driven Savings - Best Utility Technology in 2026: Cloud ERP, AI, and CIS - Buying Modern Utility Software: A Practical Evaluation Guide - Cloud for Utility Software: S/4HANA, Oracle, and CayentaCloud - Customer Experience in Utilities: The CIS-Led Digital Transformation - Digital Transformation in Utilities: A CIS and Data-First Guide - Disaster Recovery for Utility IT: CIS, OMS, and Billing Uptime - Emerging Utility Tech: AMI, DERMS, AI, and Cloud CIS - GIS in Utilities: Esri ArcGIS, Network Models, and Integration - How to Choose Utility CIS and Billing Software: A Selection Guide - Innovation in Utilities: What's Actually Changing in CIS and ERP - IoT and AMI Sensors: The Data Path Into MDM, CIS, and ADMS - Modern Utility Software, Mapped: CIS, ERP, MDM, OMS, and GIS - OT/IT Cybersecurity for Utilities: NERC CIP and Beyond - SAP vs Oracle vs Cayenta: How They Are Adding AI in 2026 - Smart Water Utility Software: SCADA, NRW, and CIS for Water - The Future of Utilities: Business Model Shifts to Watch - The Utility CIS Modernization Wave: Legacy to Cloud and AI - Utility Customer Experience: A CIS-Centric Guide - Utility Customer Portals: Self-Service That Cuts Calls - Utility Data Platforms: AMI, SCADA, and CIS Data at Scale - Utility ERP and CIS Implementation: Best Practices --- ## The Utility CIS Modernization Wave: Legacy to Cloud and AI URL: https://www.utilitieslabs.com/utility-cis-modernization-legacy-to-cloud-ai-migration/ # The Utility CIS Modernization Wave: Legacy to Cloud and AI AvanSaber Research/April 14, 2026/Updated June 2, 2026/5 min read The utility industry is in the middle of a sustained wave of customer information system replacements, and the drivers are less about ambition than about arithmetic. SAP IS-U on the ECC platform reaches the end of mainstream maintenance in 2027, with extended maintenance running through 2030. After 2030, utilities that have not migrated to S/4HANA Utilities or an alternative platform will be running unsupported software against live billing and collections. That is not a theoretical risk; it is a scheduled event. At the same time, cloud deployment has crossed the majority threshold in the broader utility billing software segment, and a new generation of AI-native platforms has emerged with large reference customers. Utilities that were already considering a CIS refresh now have both a deadline and a new set of options. ## Why the Timing Pressure Is Real in 2026 The SAP maintenance timeline is the most legible forcing function, but it is not the only one. Legacy IS-U systems, even when they are technically supported, struggle with real-time IoT data ingestion and require specialized expertise that is increasingly difficult and expensive to source. A system that was configured in the 1990s or early 2000s carries years of customization debt that makes every change slower and more expensive than it should be. The cloud share data reinforces the direction of travel. Mordor Intelligence estimates the broader utility billing software market at roughly $6.23 billion in 2025, growing to approximately $9.82 billion by 2031 at a 7.88% CAGR, with cloud deployments accounting for approximately 58% of installations. That estimate covers a broader scope than CIS alone, so it should not be treated as a CIS-specific number. For CIS specifically, SNS Insider estimates the market at approximately $1.26 billion in 2023 growing to roughly $3.54 billion by 2032 at 12.23% CAGR, while MarketsandMarkets estimates approximately $1.79 billion in 2024 growing to $3.26 billion by 2030 at 12.8%. The estimates differ because the firms use different scope definitions and methodologies; the consistent conclusion across all three is that sustained double-digit growth is being driven by modernization investment, not greenfield deployment. Gartner reinforced the market direction in its Market Guide for Utility Customer Information Systems, published July 14, 2025 (document G00827595, also covered at openintl.com). The guide named 19 representative vendors and framed modern CIS as extending “beyond billing” with composable architecture and real-time event-driven capabilities. That framing matters for migration planning because it defines what a target-state platform should be able to do, not just what the legacy system failed to do. ## What Real Moves Look Like Two programs announced in 2024 and 2025 illustrate the scale and nature of what utilities are actually doing. CPS Energy in San Antonio, serving approximately 900,000 accounts, committed roughly $304 million to a staged Oracle Cloud migration that replaces a system approximately 25 years old and built on SAP. The program is staged, meaning CPS is not attempting a single cutover across all functions simultaneously. The investment scale signals how seriously large utilities are treating the cost of staying on aging platforms relative to the cost of migration. On the vendor side, Kraken Technologies, the AI-native CIS platform that grew out of Octopus Energy, was spun out as an independent company in December 2025 at approximately $8.65 billion valuation and manages more than 70 million accounts globally. Kraken’s emergence at that scale provides a concrete reference point for what a cloud-native, AI-first CIS looks like in production. It does not make the migration decision for any individual utility, but it removes the argument that cloud-native at scale is unproven. ## The Migration Decision The core decision is not whether to modernize but which path to take. Utilities with substantial SAP ERP investment outside of IS-U, including finance, HR, procurement, and asset management running on SAP, face a different calculus than utilities whose SAP footprint is limited to the CIS. For the former group, moving to S/4HANA Utilities preserves integration investments and keeps the system landscape coherent. For the latter, a competitive evaluation makes more sense. The Oracle vs SAP comparative analysis covers the vendor-specific tradeoffs in detail. The SAP Utilities overview explains the S/4HANA Utilities forward path in the context of the IS-U to S/4HANA migration. And for utilities approaching the CIS evaluation from first principles, the comprehensive guide to choosing utilities software provides the evaluation framework. ## How to Plan the Program The utilities that run successful CIS migrations share a few consistent practices, regardless of which platform they select. First, they freeze integration scope early. A CIS touches billing, collections, customer engagement, meter data management, outage management, field service, and general ledger. Each integration point is a failure mode at cutover. Discovering a legacy interface to a collections agency six weeks before go-live is a program risk that a thorough integration inventory, completed before the design phase, would have surfaced. Second, they run parallel billing. Running the legacy and target systems in parallel for at least two full billing cycles before cutover catches rate calculation differences before they appear on customer bills. The cost of parallel billing is real but predictable; the cost of billing errors at scale after a hard cutover is neither. Third, they treat data quality as a workstream, not a checklist item. Legacy CIS databases commonly contain decades of account history with inconsistent address formats, duplicate service agreements, and rate codes that no longer exist in the active tariff schedule. A data cleanse and transformation effort that is scoped and staffed as a distinct workstream consistently outperforms one that is assumed to happen as part of configuration. Fourth, they staff for change management explicitly. CIS replacements touch every customer-facing team in the organization. Customer service, billing operations, field service, and finance all have workflows that change. Training delivered in the final weeks before cutover does not produce operational proficiency; training delivered over months, against a system that is being configured in parallel, does. The implementation best practices guide covers these program disciplines in depth. For a view of how billing and ERP functions connect in the broader utility technology stack, utility billing ERP provides the systems context. And for the wider market picture, modern software solutions transforming the utilities industry traces how the platform landscape has shifted over the past several years. The modernization wave is not a trend to watch. For utilities still running IS-U on ECC, it is a program to start planning now. The 2027 mainstream maintenance end date and the 2030 extended maintenance close are fixed points. The utilities that begin scoping, budgeting, and vendor evaluation in 2026 will have more options and more bargaining room than those that wait until the deadline is closer. ## Frequently asked questions When does SAP IS-U maintenance actually end? SAP IS-U on the ECC platform reaches the end of mainstream maintenance in 2027, with extended maintenance available through 2030. After that, the supported forward path is S/4HANA Utilities. Utilities that want to remain on a supported SAP system need to plan a migration to S/4HANA or evaluate alternative platforms before the extended maintenance window closes. How long does a CIS migration typically take? Large, complex programs at investor-owned utilities commonly run three to five years from vendor selection through final cutover. The CPS Energy Oracle Cloud migration, one of the more visible recent programs, was announced in 2024 as a staged program for roughly 900,000 accounts. Mid-market utilities with simpler rate structures and fewer legacy integrations often complete migrations in 18 to 30 months, but that range requires a realistic integration inventory before it can be trusted. Should a utility migrate to S/4HANA Utilities or evaluate alternatives? That depends on the existing SAP estate, the complexity of the billing environment, and the utility's tolerance for a multi-year program versus a platform replacement. Utilities already running deep SAP ERP integrations often find S/4HANA Utilities the lower-risk path. Utilities with fewer SAP dependencies, or those whose current CIS is not SAP at all, have more reason to run a competitive evaluation that includes cloud-native alternatives. What is the biggest risk in a CIS migration program? Integration scope underestimation is the most common cause of cost and schedule overruns. A CIS touches billing, collections, customer engagement, meter data management, outage management, field service, and general ledger. Each integration point is a potential failure mode at cutover. The utilities that run the smoothest programs typically freeze integration scope 12 to 18 months before go-live and run parallel billing for at least two full billing cycles. --- ## Utility Customer Experience: A CIS-Centric Guide URL: https://www.utilitieslabs.com/utility-customer-experience-cis-centric-guide/ # Utility Customer Experience: A CIS-Centric Guide AvanSaber Research/July 25, 2026/8 min read Utility customer experience programs usually start in the wrong place: a new front end, a journey map, a satisfaction dashboard. Meanwhile the thing every customer actually touches twelve times a year, the bill, is produced by a customer information system nobody in the CX meeting wants to talk about. This guide takes the opposite route. It starts from the structural facts that make utility CX unlike any other industry, then works through the CIS capabilities that decide most of the experience, and ends with a sober look at where AI helps and where it reliably hurts. ## Key Takeaways - Utility CX hit a measured low in 2025: the J.D. Power residential electric study recorded satisfaction of 499 on a 1,000-point scale, the lowest across its residential utility studies, with rising costs the main driver. - Customers cannot churn, so dissatisfaction flows to regulators instead. The CX budget case is cost-to-serve and rate-case credibility, not retention. - The CIS is the CX backbone: billing accuracy, start/stop/transfer, payment arrangements, and every fact a portal or bot can show a customer. - Affordability is now a CX workload: household energy arrears reached $23.0 billion by mid-2025 per NEADA, which makes payment-arrangement design a frontline experience, not a back-office one. - Proactive beats reactive: high-bill alerts and outage notifications measurably cut inbound calls. - AI belongs behind the agent and in front of routine tasks. It does not belong in front of a billing dispute. “A utility’s CX budget is usually spent on the front door while the bill is treated as plumbing. The bill is the product. Fix the data behind it before you buy another engagement layer.” Nikhil Jathar, founder of AvanSaber ## Why Utility CX Is Structurally Different A retailer with unhappy customers loses revenue. A utility with unhappy customers keeps every one of them and inherits their frustration through other channels: the contact center, the mayor’s office, and the public utility commission. Because the customer cannot leave, the feedback loop routes through the regulator, who scrutinizes service quality and customer complaints when the utility asks for a rate increase. That is the structural difference, and it means utility CX is partly a regulatory asset. The current numbers show how much pressure that loop is under. The J.D. Power 2025 U.S. Electric Utility Residential Customer Satisfaction Study, based on 127,103 interviews across the 152 largest electric utility brands, put overall satisfaction at 499 on a 1,000-point scale, the lowest score recorded across J.D. Power’s residential utility studies. The same study found average monthly residential electric costs up 34% since 2020, reaching $189 for 2025. Utilities cannot CX their way out of a price increase, but the study’s own commentary points at what they can control: reliable service, easy access to information, and frequent updates on outages and restoration. Affordability stress compounds it. The National Energy Assistance Directors Association reported that household energy arrears rose about 31%, from roughly $17.5 billion to $23.0 billion, by June 30, 2025, with about 21.5 million households behind on their energy bills. One in six residential customers being in arrears is not a collections statistic. It is a customer experience population, and how a utility treats it is increasingly what commissions judge. ## The CIS Is the CX Backbone Strip utility CX to the interactions that actually happen and almost all of them resolve to the customer information system. What SAP, Oracle, and Cayenta platforms actually cover is mapped in our overview of what a utility CIS includes; here is why it decides the experience. Billing accuracy is the foundation. An accurate, explainable bill is the single largest CX deliverable a utility produces, and every estimated read, misapplied rate, or unexplained true-up manufactures a contact. Exception handling inside the CIS (catching a billing anomaly before the bill goes out, rather than after the customer calls) is invisible CX work, and it is where modern AI earns its keep, as covered in AI in utility billing. Start, stop, and transfer is the first impression. Move-in and move-out is often the first and last interaction a customer has, and it is pure CIS workflow: identity, service point, deposit rules, final bill. A move-in that completes online in minutes sets the relationship’s tone; one that requires a phone call and a follow-up sets a different one. Payment arrangements are frontline CX now. With arrears at the levels NEADA documents, the flexibility of the CIS’s payment-arrangement engine (installment plans, deferred balances, assistance-program integration, safe dunning rules) determines whether a struggling household experiences the utility as workable or hostile. The collections side of this, and where agentic AI is starting to help, is the subject of our piece on AI in utility collections and dunning. The wider modernization story, why utilities are re-platforming these functions at all, is covered in customer experience as a CIS-led digital transformation. ## Self-Service Portals: The Second Front Door A portal or app is where most routine interactions should land: bill view, payment, usage history, outage reporting, and service moves. The catch is that a portal is only as good as its CIS integration. A portal that shows a three-day-old balance, or cannot set up a payment arrangement because that logic lives only on the agent’s screen, teaches customers to call instead, and every taught call is a permanent cost. J.D. Power’s commentary on the 2025 study makes the same point from the data: utilities that deliver streamlined billing, digital self-service, and timely outage alerts see satisfaction rise even in a cost-pressured environment. The practical test for any portal project is coverage of the top ten call reasons. If a customer can see why the bill went up, fix a payment problem, and report an outage without calling, the portal is doing its job. The build-versus-integrate details are in our guide to customer portals in utility services. ## Proactive Communication: Outages and Usage The cheapest contact is the one the utility sends first. Two categories dominate. Usage and bill alerts. Oracle’s Opower Proactive Alerts watch interval data for customers trending toward an unusually high bill and message them before it lands; Oracle’s documentation credits high bill alerts with a 9% reduction in high-bill calls plus 0.3% energy savings on top of the 1.5% from Home Energy Reports. The scale is no longer experimental: Oracle reported approximately 45 million North American households in Opower AI-driven programs as of April 2026. A bill shock predicted is a dispute avoided. Outage communication. Customers tolerate outages far better than silence about them. The J.D. Power finding above lists outage updates and planned-restoration times among the top drivers utilities control. The prerequisite is unglamorous: the outage system and the CIS have to agree on who is affected, so the alert reaches the right households with a restoration estimate the utility can stand behind. ## Where AI Helps and Where It Hurts AI has a real place in utility CX, and a well-documented failure mode. The map of what is production-grade across the whole operation is in our pillar on where AI in utilities actually lands; this is the CX-specific cut. Where it helps. Behind the agent: Oracle shipped AI call summarization and tagging for utility contact centers in May 2025, cutting after-call work so agents spend time on the conversation instead of the notes. In proactive alerts, as above. And in routine self-service: balance, due date, outage status, straightforward payment arrangements, where the rules are deterministic and the bot is just a faster interface. Where it hurts. In front of a billing dispute. A Verint survey of 1,500 consumers found more than two-thirds had experienced a bad chatbot interaction, with failure to understand the issue as the top complaint, alongside self-service flows that never reach a live agent. A Five9 study of 4,000 consumers in October 2024 found 75% prefer a human for customer service, 56% are often frustrated by AI chatbots, and 48% do not trust information from them. A disputed bill is high-stakes, emotionally loaded, and account-specific, which is the exact profile bots handle worst. A customer who is already angry about a $400 bill and then loops through a bot that misunderstands the problem does not file that experience under technology; they file it under the utility’s name, sometimes in a commission complaint. The design rule that follows: automate the status update, the outage map, and the payment plan setup, and hand the dispute to a person, quickly and with the full account history on screen. Escalation speed is a feature, not an admission of failure. How to build the agent-assist side of that is covered in AI in utility customer service. “Never let a chatbot argue about money. The bot’s job is to notice the dispute early and get it to a human with the context attached. Every vendor demo shows the bot answering; ask instead to see it escalating.” Nikhil Jathar, founder of AvanSaber ## The CX Moments, Mapped to the CIS | CX moment | CIS capability behind it | Common failure mode | First bill after move-in | Start/stop/transfer workflow, rate assignment | Wrong rate or deposit surprise on bill one | Bill higher than expected | Interval data analytics, proactive high-bill alerts | Customer discovers it first, calls angry | Payment trouble | Payment-arrangement engine, assistance integration | Rigid plans, dunning that ignores an active arrangement | Outage | OMS-CIS linkage for affected-customer lists | Silence, or alerts sent to the wrong households | Billing dispute | Exception history, adjustment audit trail | A chatbot in front of it | Routine account tasks | Portal integrated to live CIS data | Stale balances that teach customers to call ## What to Measure and What to Ask Measure utility CX where it is actually created: first-bill accuracy for new accounts, exception rate per billing run, percentage of top call reasons fully closable in self-service, alert reach before high bills and during outages, arrangement completion rate for customers in arrears, and time-to-human on escalations. Those six numbers predict the J.D. Power score and the commission’s mood better than any survey the utility runs on itself. And when a vendor pitches a CX layer, ask the CIS questions first: what data does it read, how fresh is it, what actions can it complete rather than merely display, and what happens at the moment of escalation. The platform context for those questions is in our CIS-led digital transformation guide and the vendor AI landscape in the SAP vs Oracle vs Cayenta comparison. For utilities working through these CIS assessment questions, the AvanSaber team provides vendor-neutral advisory on CIS capability gaps, portal integration, and customer experience program design across SAP, Oracle, and Cayenta platforms. ## Frequently asked questions Why is utility customer experience different from retail or banking CX? Utility customers cannot leave, so satisfaction does not show up as churn. It shows up as call volume, complaints to the public utility commission, and opposition in rate cases. The regulator effectively acts as a second customer. That changes the economics: the return on utility CX comes from lower cost-to-serve and smoother regulatory proceedings, not from retention. What system actually drives utility customer experience? The customer information system (CIS). Billing accuracy, start/stop/transfer service, payment arrangements, and the data behind every portal screen and outage alert all come from the CIS. Front-end tools can only present what the CIS knows, so CX programs that skip the CIS data layer tend to disappoint. How bad is utility customer satisfaction right now? The J.D. Power 2025 U.S. Electric Utility Residential Customer Satisfaction Study, released in December 2025, recorded overall satisfaction of 499 on a 1,000-point scale, the lowest across its residential utility studies, driven mainly by rising costs. Average monthly residential electric costs were up 34% since 2020, reaching $189 for 2025. Where does AI help utility customer experience, and where does it hurt? AI helps behind the agent (call summarization, account context, next-step suggestions), in proactive alerts (high-bill and outage notifications), and in routine self-service like balance checks and payment arrangements. It hurts when a chatbot is put in front of a billing dispute: surveys in 2024 found most consumers report bad chatbot experiences and a strong preference for humans on complex issues. Automate the routine, escalate the disputed. ---