AI-Driven Asset Onboarding at Evergy, RCM-as-a-Service Adoption, and Three Other 2026 Maximo Case Studies Worth Your Time

Most of the AI-in-Maximo conversation is still about demos. The case studies that matter in mid-2026 are the ones where the AI is running in production, with measurable results, in industries where the asset base is the business. This article walks through the most useful ones, what they actually did, and what the practitioner-level lesson is.

Evergy + 1898 & Co.: AI-driven asset onboarding from P&IDs

The 2026 PowerGen session catalog includes a case study from Evergy and 1898 & Co. that is the cleanest real-world example I've seen of AI being used to solve a problem Maximo practitioners have complained about for years: asset data quality at onboarding.

The problem

A utility like Evergy is building or refurbishing a substation. The engineering team produces a thick stack of design documents: P&IDs (piping and instrumentation diagrams), one-line diagrams, equipment lists, DCS I/O files, nameplate photos. The asset data in these documents needs to land in Maximo as structured records — asset, location, classification, spec, vendor, model, rotating equipment hierarchy, and so on.

In the old model, this was a multi-week manual data entry project. Engineers or contracted data entry staff would read the documents, type the records into Maximo, validate, and iterate. The error rate was high. The cycle time was long. The institutional knowledge lived in the documents and in the heads of the engineers who wrote them.

What Evergy built

The 1898 & Co. team combined four AI capabilities on top of Maximo:

1. Optical character recognition (OCR) to digitize the engineering documents at the line and shape level.
2. Computer vision to identify equipment symbols, tags, and connection topology in P&IDs and one-line diagrams.
3. Generative AI to interpret the equipment lists, the DCS I/O files, and the human-readable notes, and to produce structured records.
4. Semantic parsing to map the extracted data into the Maximo asset and location hierarchy, including classification, specification, and rotating equipment structure.

The result is a pipeline that takes a stack of design documents and produces Maximo-ready asset records with high accuracy. The data fidelity is significantly better than manual entry (the error rate is in the low single digits, vs. mid-single-digits or higher for manual). The cycle time is days, not weeks. And the institutional knowledge is captured in the AI pipeline, not lost when the engineers rotate off the project.

Why this matters beyond Evergy

This is the pattern that every asset-intensive industry needs. The bottleneck on Maximo data quality is almost never the platform — it is the data capture at the source. The Evergy case study shows that you can build a data-capture pipeline that:

• Reads the documents the engineers already produce.
• Produces Maximo records that match your classification and hierarchy.
• Frees your engineers to do engineering, not data entry.

The 1898 & Co. team is presenting this session at industry conferences through 2026 (PowerGen being one). If you are looking for a pattern to follow, this is the one.

Entergy: long-running Maximo program, evolving AI/ML integration

Entergy's Maximo program is one of the longest-running in the energy sector — a multi-year, multi-phase rollout covering fleet, gas, transmission, and distribution, with integrations across procurement, AMI (automated meter infrastructure), and contractor compliance.

Two 2026-era patterns worth noting:

• Contractor compliance via API integration. Entergy established an API between Avetta (contractor compliance) and their data lake, and the data flows through into Maximo. The result is that contractor compliance status is visible in the work execution workflow — a technician opening a work order can see whether the contractor on the job is currently compliant. This is a small but high-leverage integration.
• Distribution, transmission, gas, fleet — phased by domain. Entergy's rollout was not a big-bang. It was a phased approach by business domain, with each phase delivered in 6–12 months. This pattern is one that consistently works for large utilities: phase by domain, prove value, expand.

The lesson from Entergy: an asset-intensive utility program is a multi-year commitment, and the integrations you build with the rest of the enterprise (procurement, contractor management, AMI, GIS) are as important as the core Maximo configuration. The platform is the foundation; the integrations are the value.

RCM-as-a-Service: the partner-delivered reliability engineering pattern

The 2026 trend worth flagging is the rise of RCM-as-a-Service offerings from Maximo partners. The pattern: a partner (typically an EAM specialist firm) delivers the FMEA analysis, the maintenance strategy development, the Reliability Strategies library population, and the ongoing governance of the program — on a multi-year managed-service contract.

Why this is rising now:

• Reliability engineering talent is scarce. Most asset-intensive organizations have one or two people who can credibly lead an RCM analysis. They are stretched.
• The tools are finally there. With the AI-assisted FMEA generation (sfmea template in MAS 9.1 AI Service) and the unified Manage-Health-Predict loop, the productivity of a small reliability engineering team has gone up. But you still need the team.
• The risk of an internal-only RCM program is that it stalls. An RCM program that has to compete with day-to-day work orders for attention is a stalled RCM program.

The practical takeaway: if you are starting a reliability program in 2026 and you don't have 3+ dedicated reliability engineers, look seriously at a partner-delivered model for the first 12–18 months. The cost is real but the alternative — a stalled internal program — is more expensive.

Manufacturing: condition-based maintenance and the OEE link

The manufacturing case studies in 2026 cluster around the link between Maximo Health and Overall Equipment Effectiveness (OEE). The pattern:

1. Health scores the asset's condition.
2. Predict produces a probability of failure in the next N days.
3. The work order generated by the loop is scheduled to minimize OEE impact — typically scheduled for a planned downtime window.
4. The result is measured as a delta in OEE before and after the program.

The IBM-published case study data on manufacturers that have implemented this loop shows mid-single-digit OEE improvements in year one, growing to high-single-digit or low-double-digit improvements by year three. The OEE improvement is the executive-visible KPI. The Maximo APM stack is the engine.

Transportation and aviation: high-criticality asset patterns

Transportation (rail, transit, ports) and aviation MRO are the verticals where the cost of an asset failure is most asymmetric. A failed bearing on a locomotive is recoverable; a failed engine on an aircraft is not. The case studies from these verticals in 2026 show three patterns:

• Condition-based maintenance is the default, not the aspirational state. Inspections and sensor data drive the work.
• The FMEA library is the institutional asset. When a new asset class is introduced, the FMEA is built up-front, not retrofitted.
• Mobile execution is non-negotiable. Technicians work in the field, on the asset, in any weather. The mobile experience has to be as good as the desk experience.

The lesson for other industries: the high-criticality verticals have already done the organizational and process work. Their Maximo configurations reflect that. When a less-criticality industry adopts Maximo, the gap is rarely the platform — it is the institutional discipline that the high-criticality verticals built up over years.

Public sector and facilities: a different optimization

Public sector (federal, state, municipal) and facilities (real estate, hospitals, universities) have a different optimization function. The asset base is broad, the criticality is moderate, and the cost of failure is rarely catastrophic. The 2026 case studies in this space show two patterns:

• Maximo Real Estate and Facilities (RE&F) is the right surface for portfolio-level asset and lease management. The 9.1.10 release (April 30, 2026) added platform UI alignment fixes and security enhancements that are the kind of thing a public-sector IT team needs to check off before a security review.
• The work is the integration. Facilities and RE&F deployments are typically integrated with HR (for space allocation per FTE), with finance (for lease accounting under ASC 842 / IFRS 16), and with project management (for capital projects). The Maximo configuration is the easy part; the integrations are the work.

Healthcare: a quiet but real adoption

Healthcare asset management — the medical equipment fleet — is a smaller segment by Maximo license count but a high-value one. The 2026 case studies show medical centers using Health + Predict to manage their imaging fleet (MRI, CT, X-ray), where unplanned downtime directly affects patient throughput.

The pattern: condition monitoring on critical subsystems (chiller, gradient coil, X-ray tube), Health scoring on the asset, Predict for time-to-failure, and a tightly managed PM program. The result is fewer patient-impacting downtime events and a longer useful life for very expensive equipment.

A practitioner-level read

If you are trying to pick which case study to learn from, the 2026 short list is:

1. Evergy + 1898 & Co. for AI-driven asset onboarding. The pattern of "use AI to read engineering documents and produce Maximo records" is generalizable across industries.
2. Entergy for the multi-year, multi-domain utility rollout. The phasing pattern and the integration pattern are the most useful lessons.
3. A manufacturing OEE-linked CBM program for the reliability-to-OEE story. The KPI is what sells the program to executives.
4. A transportation or aviation deployment for the high-criticality discipline. The FMEA-first, condition-driven, mobile-execution pattern.
5. A partner-delivered RCM engagement if you don't have the internal reliability engineering capacity. The RCM-as-a-Service pattern is real and worth evaluating.

The case studies are converging on a picture: Maximo is no longer just the system of record for work orders and assets. It is the system of record for asset lifecycle decisions, and the AI is making those decisions faster and more reliably than humans can on their own. The organizations that are pulling ahead in 2026 are the ones that have built the operational discipline to use those decisions well.