The Evidence Map for Industrial AI Value
The action layer behind the core verdict: how to fund AI that changes a measurable physical loop, and screen out the generic-copilot POC before it stalls.
Use this as a pre-spend diagnostic for any industrial or infrastructure AI proposal. The question is not whether the model is impressive. The question is whether the initiative can name the exact alarm, dispatch, maintenance order, or control-room decision that changes after go-live, who owns the result when the model is wrong, and the physical metric it moves. Value lives in a bounded operating loop with trusted data and an accountable owner. AI can change that loop; it cannot substitute for it.
Gates before you fund a pilot
Require the sponsor to choose one metric: outage duration, unplanned downtime, OEE, quality escapes, energy intensity, truck rolls, leak loss, safety incidents, permit cycle time, inventory turns, or work-order accuracy. "Improve productivity" is not enough.
Name the exact screen, alarm, maintenance order, dispatch queue, crew briefing, shift handoff, or control-room decision that changes. If the model output does not enter a live operating loop, it is a demo.
Pick 20 real historical cases and trace whether the sensor tags, asset hierarchy, maintenance notes, inspection records, failure labels, weather and load context, and operator decisions are findable and trusted. Do this before vendor demos.
The owner keeps the golden cases, signs the eval standard, decides when the model is wrong, and owns the handoff to frontline training. IT can govern the platform; the business must own the delegated work.
In regulated or safety-critical contexts, begin with inspection, planning, retrieval, anomaly triage, or human-approved work orders. Move toward closed-loop control only after logging, human oversight, rollback, cybersecurity, and assurance gates are proven.
Add the cost of data cleanup, OT access, cyber review, operator training, model monitoring, legal and compliance review, change management, and maintenance. If the business case only counts software and model cost, it is underwritten to disappoint.
Owner, briefing, proof
Owner
One accountable operating owner, not a committee, for each initiative. The owner keeps the golden cases, signs the eval standard, decides when the model is wrong, and owns the handoff to frontline training.
Briefing
A pre-spend decision brief per proposal: the named physical metric, the day-after workflow, the data-lineage proof, and the assurance path, so a paid pilot is funded only when the operating loop is real.
Proof
The chain from a named physical constraint to trusted data to a live operating decision to a finance- or reliability-controlled result, with baseline and comparison, not a demo.
Start with the pre-spend diagnostic on one proposed loop: metric, data, owner, assurance. If the gap is material, widen to a readiness look that designs the operating change across the candidate loops, and build the machinery and the receipts only when the enterprise wants them run.
Claim ledger
Some strong figures here are potential, not guaranteed ROI. Grid-application ranges and cooling-energy reductions are real cases that shouldn't be generalized to every system. Media anecdotes aren't proof of returns; keep them illustrative. And "data readiness" doesn't mean a multi-year governance program before any value; it means the specific data behind the one loop you're changing. The widely-shared "95% of pilots fail" figure is not industrial-specific and is not asserted here.
- Rigorous, industrial-specific pilot-failure-rate data from a verified primary source.
- A peer-reviewed or regulator-filed study showing AI improved downtime, OEE, outage duration, leak loss, or maintenance cost after production integration, with baseline and comparison.
- Causal evidence that a published scaling playbook reproduces results outside the companies that already won.
- Audited operator ROI tied to a physical metric that includes integration and change-management costs.
- The MIT NANDA "95%" report officially hosted, independently replicated, or retracted.