Investigate an energy asset before planning work

When plant equipment runs outside its normal range, an agent gathers the evidence from the historian, the asset management system and engineering tools. An engineer decides what work to plan.

A gas turbine bearing runs hot, or a feed pump vibrates more than usual. An alarm or an asset performance alert flags it. Before anyone plans work, someone must find out why. Today a reliability engineer owns that investigation, with the control room and the maintenance planners. The evidence sits in five or more systems, and the work takes days. In this design an agent gathers the evidence into one case and drafts a work notification. An engineer decides, in the asset management system. The agent never touches the plant. The same work applies to a gas compressor on a platform, or a mill bearing at a mine.

Four steps, left to right. Observe: an anomaly in readings or thermal images. Compare: maintenance records and operating conditions such as load, inlet temperature and flow. Investigate: request the missing evidence, such as recent images, inspection reports and operating logs. Prepare work: an engineer reviews the proposal and approves the work order. A banner says there is no automatic plant control. — **Investigate before planning work**The useful output is a checked finding or a clear evidence gap. An engineer then decides.Open image at full size ↗

The systems involved

The investigation touches seven of the fourteen layers. The full estate is on the energy and resources page.

Layer	Systems in this use case
Data platform (R02)	Process historian (AVEVA PI), which stores every sensor reading over time. Lab system for oil samples. Lakehouse and reports
Integration fabric (R03)	The buffer network between plant and business networks, with a historian copy. A read broker for agents. The EAM programming interface
Systems of record (R04)	Enterprise asset management, or EAM (SAP PM, IBM Maximo): notifications, work orders and history. ERP for parts. Permit to work. Document store with equipment manuals
Line of business and OT (R05)	Distributed control system (DCS) and its alarms. Asset performance management, or APM (AVEVA, GE Vernova, Aspen Mtell). Condition monitoring (Bently Nevada). For a generator, the bidding link to the market operator
Intelligence and learning (R06)	Anomaly models, performance models, vibration analysis and a digital twin
Agent platform (R07)	New: the agent, its case store, and a search index over manuals and past work orders
Operating model and supervision (R13)	Control room operator, reliability engineer, maintenance planner, superintendents, trading desk or production planner

Layer: Data platform (R02)
Systems in this use case: Process historian (AVEVA PI), which stores every sensor reading over time. Lab system for oil samples. Lakehouse and reports

Layer: Integration fabric (R03)
Systems in this use case: The buffer network between plant and business networks, with a historian copy. A read broker for agents. The EAM programming interface

Layer: Systems of record (R04)
Systems in this use case: Enterprise asset management, or EAM (SAP PM, IBM Maximo): notifications, work orders and history. ERP for parts. Permit to work. Document store with equipment manuals

Layer: Line of business and OT (R05)
Systems in this use case: Distributed control system (DCS) and its alarms. Asset performance management, or APM (AVEVA, GE Vernova, Aspen Mtell). Condition monitoring (Bently Nevada). For a generator, the bidding link to the market operator

Layer: Intelligence and learning (R06)
Systems in this use case: Anomaly models, performance models, vibration analysis and a digital twin

Layer: Agent platform (R07)
Systems in this use case: New: the agent, its case store, and a search index over manuals and past work orders

Layer: Operating model and supervision (R13)
Systems in this use case: Control room operator, reliability engineer, maintenance planner, superintendents, trading desk or production planner

How it runs today

The work starts with an APM alert or a DCS alarm. It can also start with an email from the equipment maker's diagnostics centre, or an operator who notices a trend. It then runs in eleven steps.

Step	Who	System	How the data moves
1. Alert	APM software, or the control room operator	APM reading the historian. Or a DCS alarm	APM raises an alert when a reading leaves its normal band. It shows on a screen and goes out by email
2. First look	Control room operator, shift supervisor	DCS trends, historian screen (PI Vision)	The operator looks at the trend and checks load and air temperature. A line goes in the shift log
3. Raise a notification	Operator, or diagnostics analyst	EAM (SAP PM notification, Maximo service request)	Typed by hand: free text, equipment number, priority
4. Engineering review	Reliability or plant engineer	Historian, APM, condition monitoring, EAM history, equipment manuals	Exports trends to Excel. Reads past work orders on screen. Emails the equipment maker
5. Ask for more evidence	Engineer, then a condition monitoring technician or an oil lab	Vibration route, oil sample, thermal camera	Asked by email or notification. Results come back as PDF reports, often days later
6. Decide	Engineer, maintenance and operations superintendents	Morning meeting, Teams	A spoken decision: keep watching, change how the unit runs, fix at the next outage, or fix now
7. Plan the work	Maintenance planner	EAM work order. ERP parts reservation	The notification becomes a work order with tasks, parts and labour
8. Tell the market or production planning	Trading desk, or production planner	Bidding system to the market operator (AEMO). Production forecast	A generator rebids its availability, with a reason. An oil and gas operator books the lost volume
9. Control of work	Area authority, isolating authority	Permit to work system, or paper permits	A permit is issued. The equipment is isolated and locked
10. Do and close	Technician, supervisor	Mobile work app (Maximo Mobile, SAP Asset Manager)	Hours, parts and findings recorded. Failure codes are often left blank
11. Report	Reliability engineer, plant manager	Reports over copies of EAM and historian data	A monthly reliability report. US generators also file event reports to NERC

Step: 1. Alert
Who: APM software, or the control room operator
System: APM reading the historian. Or a DCS alarm
How the data moves: APM raises an alert when a reading leaves its normal band. It shows on a screen and goes out by email

Step: 2. First look
Who: Control room operator, shift supervisor
System: DCS trends, historian screen (PI Vision)
How the data moves: The operator looks at the trend and checks load and air temperature. A line goes in the shift log

Step: 3. Raise a notification
Who: Operator, or diagnostics analyst
System: EAM (SAP PM notification, Maximo service request)
How the data moves: Typed by hand: free text, equipment number, priority

Step: 4. Engineering review
Who: Reliability or plant engineer
System: Historian, APM, condition monitoring, EAM history, equipment manuals
How the data moves: Exports trends to Excel. Reads past work orders on screen. Emails the equipment maker

Step: 5. Ask for more evidence
Who: Engineer, then a condition monitoring technician or an oil lab
System: Vibration route, oil sample, thermal camera
How the data moves: Asked by email or notification. Results come back as PDF reports, often days later

Step: 6. Decide
Who: Engineer, maintenance and operations superintendents
System: Morning meeting, Teams
How the data moves: A spoken decision: keep watching, change how the unit runs, fix at the next outage, or fix now

Step: 7. Plan the work
Who: Maintenance planner
System: EAM work order. ERP parts reservation
How the data moves: The notification becomes a work order with tasks, parts and labour

Step: 8. Tell the market or production planning
Who: Trading desk, or production planner
System: Bidding system to the market operator (AEMO). Production forecast
How the data moves: A generator rebids its availability, with a reason. An oil and gas operator books the lost volume

Step: 9. Control of work
Who: Area authority, isolating authority
System: Permit to work system, or paper permits
How the data moves: A permit is issued. The equipment is isolated and locked

Step: 10. Do and close
Who: Technician, supervisor
System: Mobile work app (Maximo Mobile, SAP Asset Manager)
How the data moves: Hours, parts and findings recorded. Failure codes are often left blank

Step: 11. Report
Who: Reliability engineer, plant manager
System: Reports over copies of EAM and historian data
How the data moves: A monthly reliability report. US generators also file event reports to NERC

A swimlane with six lanes: plant systems, control room, engineers, maintenance, trading or planning, and area authority. Step 1: the APM or a DCS alarm raises an alert. Steps 2 and 3: the control room takes a first look at the trend and types a notification into EAM. Step 4: an engineer reviews the evidence in Excel and email. Step 5: a technician or oil lab sends results back days later. Step 6: the engineers decide at the morning meeting. This is an approval. Step 7: the planner builds the work order. Step 8: the trading desk rebids or the planner updates the forecast. Step 9: the area authority issues the permit and isolation. This is an approval. Step 10: the technician does the work and closes it, often with thin findings. Step 11: the engineer reports monthly. From alert to decision takes one to ten working days, typical and unsourced. — **Asset investigation today: who does what**Today the evidence moves between five or more systems by export and email.Open image at full size ↗

Handoffs

The case passes from the control room to engineering, then to maintenance planning. For a generator it then goes to the trading desk. For an oil and gas operator it goes to production planning. Finance codes the cost as operating or capital spend.

Where reporting comes in

APM and historian analytics start the work. Monthly dashboards report availability and the forced outage rate. Board packs carry availability and major failures. Generators also report availability to the market and the regulator.

Cycle time and volumes

From alert to an engineering decision: one to ten working days (typical, unsourced).
From notification to scheduled work, when not urgent: one to four weeks (typical, unsourced).
A fleet diagnostics centre may review tens to hundreds of alerts a week (typical, unsourced).

Pain points

The evidence sits in five or more systems. Engineers move it by Excel export and email.
Many alerts are false or already known. So people stop reading them.
A failing sensor can look like a failing machine.
Work orders close with thin findings and no failure code. The next investigation starts from scratch.
Know-how sits with a few experienced engineers.
The equipment maker's diagnostics data and the site's data rarely join.

How agents run it

The agent runs the evidence gathering. People keep every decision. The agent reads plant data only from the historian copy, through a read broker in the buffer network. It never reads from a controller. It writes only to its own case store and to a draft notification in EAM.

Step	Agent or person	Reads from, and how	Writes to	Who approves, and where
1. Open a case	Agent	APM alerts and alarm history, as events through the read broker	Its own case store	No one. Opening a case changes nothing in the plant
2. Check the reading	Agent	Historian programming interface (PI Web API, Seeq): quality flags, flat lines, drift, nearby sensors	Case store	No one
3. Check how the unit ran	Agent	Historian: load, air temperature, fuel, recorded setpoints. Shift log. Dispatch targets	Case store	No one
4. Check recent work	Agent	EAM tool, read only: recent work orders, open notifications. Maximo offers this through an MCP server, a standard way to give agents tools	Case store	No one
5. Run engineering tools	The agent calls them. The tools calculate	Anomaly model, performance model, vibration analysis, digital twin	Case store, with each output and tool version	No one
6. Check each finding	Fixed rules	Live data and similar past cases	Drops a claim, or marks it unverified	No one. Only checked findings go on
7. Ask for evidence	Agent drafts	Gaps in the case	A draft request to the condition monitoring queue	The engineer. Or an agreed rule for internal requests
8. Propose	Agent	The case	A finding pack, and a draft EAM notification marked "agent draft"	The engineer, in the EAM worklist
9. Decide and plan	Engineer, superintendents, planner	The finding pack	EAM work order	People, as today
10. Market or production change	Trading desk, or production planner	The work plan	Rebid with a reason. Outage plan. Production forecast	People. The participant owns every bid
11. Control of work	Area authority	Permit system	Permit and isolation	People only. The agent has no path to it
12. Learn	Engineer	Findings on the closed work order	An "accepted" or "rejected" label on the case	The engineer reviews each label

Step: 1. Open a case
Agent or person: Agent
Reads from, and how: APM alerts and alarm history, as events through the read broker
Writes to: Its own case store
Who approves, and where: No one. Opening a case changes nothing in the plant

Step: 2. Check the reading
Agent or person: Agent
Reads from, and how: Historian programming interface (PI Web API, Seeq): quality flags, flat lines, drift, nearby sensors
Writes to: Case store
Who approves, and where: No one

Step: 3. Check how the unit ran
Agent or person: Agent
Reads from, and how: Historian: load, air temperature, fuel, recorded setpoints. Shift log. Dispatch targets
Writes to: Case store
Who approves, and where: No one

Step: 4. Check recent work
Agent or person: Agent
Reads from, and how: EAM tool, read only: recent work orders, open notifications. Maximo offers this through an MCP server, a standard way to give agents tools
Writes to: Case store
Who approves, and where: No one

Step: 5. Run engineering tools
Agent or person: The agent calls them. The tools calculate
Reads from, and how: Anomaly model, performance model, vibration analysis, digital twin
Writes to: Case store, with each output and tool version
Who approves, and where: No one

Step: 6. Check each finding
Agent or person: Fixed rules
Reads from, and how: Live data and similar past cases
Writes to: Drops a claim, or marks it unverified
Who approves, and where: No one. Only checked findings go on

Step: 7. Ask for evidence
Agent or person: Agent drafts
Reads from, and how: Gaps in the case
Writes to: A draft request to the condition monitoring queue
Who approves, and where: The engineer. Or an agreed rule for internal requests

Step: 8. Propose
Agent or person: Agent
Reads from, and how: The case
Writes to: A finding pack, and a draft EAM notification marked "agent draft"
Who approves, and where: The engineer, in the EAM worklist

Step: 9. Decide and plan
Agent or person: Engineer, superintendents, planner
Reads from, and how: The finding pack
Writes to: EAM work order
Who approves, and where: People, as today

Step: 10. Market or production change
Agent or person: Trading desk, or production planner
Reads from, and how: The work plan
Writes to: Rebid with a reason. Outage plan. Production forecast
Who approves, and where: People. The participant owns every bid

Step: 11. Control of work
Agent or person: Area authority
Reads from, and how: Permit system
Writes to: Permit and isolation
Who approves, and where: People only. The agent has no path to it

Step: 12. Learn
Agent or person: Engineer
Reads from, and how: Findings on the closed work order
Writes to: An "accepted" or "rejected" label on the case
Who approves, and where: The engineer reviews each label

The finding pack holds the finding, the other explanations considered, the sources, the gaps and a suggested work scope. Step 6 is the validation loop described on the line of business and OT layer (R05) page. Suggestion tools already ship, so this check is the part worth designing.

A swimlane with seven lanes: agent, engineering tools, fixed rules, engineer, trading or planning, area authority, and the case store. Step 1: the agent opens a case from an APM alert, read through the broker. Steps 2 to 4: the agent checks the reading, how the unit ran, and recent work in EAM. Step 5: engineering tools do the calculations. Step 6: fixed rules check each finding against live data and past cases. Step 7: the agent drafts a request for more evidence. The engineer approves the request. Step 8: the agent proposes a finding pack and a draft EAM notification. Step 9: the engineer approves the draft in EAM, which is an approval, and plans the work. Step 10: the trading desk or planner changes the bid or forecast. Step 11: the area authority issues the permit. This is an approval, and the agent has no path to it. Step 12: the engineer labels the finding accepted or rejected. The case store keeps every source, tool output, check and approval. — **Asset investigation with an agent: who does what**The agent gathers and checks. Engineers decide in EAM. Permits and bids stay with people.Open image at full size ↗

What stays with a person

How the unit runs, and any setpoint change. That is the control path.
The safety instrumented system and protection settings. Safety actuation is a deterministic zone.
Permits and isolation. Control of work is a legal safety record.
Outage commitments and rebids. The market participant owns every bid (Australian Energy Regulator, 5 August 2026).
Work priority and spend. These are delegated authorities.

Where the evidence is kept

The case store keeps each source reference: the tag, the time range and the query. It keeps each tool output with its version, the check result and who approved. The EAM notification carries the case ID. The operator holds the logs, not the vendor, and the agent cannot change them. This is the observability and audit trail capability at work.

What changes in the systems

A new EAM notification type or status: "agent draft".
A case ID field on notifications and work orders.
A review queue in the engineer's EAM worklist.
An event subscription to APM alerts.
A read-only agent identity on the historian and in the buffer network.
An EAM tool that can only read history and create a draft notification. It sits behind the tool gateway.
A search index over equipment manuals and past work orders, held in knowledge and memory.
Model hosting that respects where plant data may go. Plant data often may not leave the site or the country.

What does not change

The DCS, the SIS, alarm settings and protection settings stay as they are. So do the APM models, unless reviewed. The permit process, the maintenance strategy, the market bidding rules and finance approvals do not change.

Who does this today

Vendors now ship the part that covers one system. No one was found doing the join across historian, EAM, lab and permits.

TotalEnergies and Chevron with Honeywell. Honeywell's Experion Operations Assistant launched on 19 March 2026. Chevron and TotalEnergies piloted it. It warned operators 5 to 10 minutes before alarms, and operators acted [vendor]. At TotalEnergies' Port Arthur refinery it forecast five events, about 12 minutes ahead (Hydrocarbon Processing, November 2025) [vendor]. Status: launched. It advises.
Aker BP with Cognite. Aker BP uses Cognite Atlas AI agents for root cause analysis. Cognite says they cut engineering time by more than 70 percent (World Oil, 16 September 2025) [vendor]. The article does not say who measured it. Status: in use, by the release. It advises.
Shell with C3 AI. C3 AI reliability agents find root causes, draft work orders and check parts, "with human approval or override". The source is one article (Portal Tela, 5 June 2026). Status: unverified. The article is machine-written and cites no source.

Asset system vendors have announced agents too. IBM announced Maximo Condition Insight on 5 December 2025. It says automatic work order creation is coming "soon" [vendor]. SAP planned its Maintenance Planner Agent for late 2025. This research could not confirm it shipped [vendor]. Seeq (3 March 2026) and Emerson (11 May 2026) announced investigation agents and advisors [vendor].

The same pattern runs in network planning. The agent chooses the studies, a solver calculates, and an engineer decides. AgenticGrid and Duke Energy's work with AWS are described on the energy and resources page.

Where this sits

The four questions place this agent as follows. Its role is assist: it does work for an engineer. Its permissions are read only on the historian copy, the APM events and the EAM history. It can write only to its own case store and create a draft EAM notification. It has no path to the control system, the permit system or the bidding system. It runs at delegated tasks (A2) on the autonomy contract, both as designed and as permitted. It does a set task, and an engineer approves every output. Its impact is small. A draft can be rejected, and nothing physical moves.

On the risk profile it is Tier 2 (enhanced), because it writes drafts into a production EAM. Many operators class historian data as control-system data. If yours does, it is Tier 3 (strict). It touches no deterministic zone. Safety actuation stays out of its reach by design.

It starts at curated learning (L1). People choose the past cases it uses. Accepted and rejected findings can later feed reviewed learning (L2), through the learning flywheel.

Measure the result

Compare the agent with existing alarms, fixed checks and engineer-led investigation. Count accepted findings, missed issues in a reviewed sample, false work requests and review time. Track the time from alert to decision against today's one to ten days. Reduced downtime is a separate result. It needs a fair comparison over a long period.

Start with a trial

Take a year of closed investigations on one unit or one class of pump. Run the agent on that history, with no live drafts. Ask two engineers to grade each finding pack against what was actually found. Then allow live drafts at one site. Register the agent through the agent lifecycle, with a named sponsor, before it goes live.

Evidence and limits

The as-is steps are typical industry practice, not one company's process. No published cycle times or volumes were found, so all of them are marked typical and unsourced.

The vendor results are the vendors' own. The Aker BP figure does not say who measured it. Shell's story exists only in a machine-written article.

No vendor agent found joins the historian, EAM, lab, permits and production accounting. That join is the part this design adds. It has not been tested.

The IEA's Energy and AI report (2025) discusses AI in maintenance. It does not report a result for this workflow.

Next step

Score this agent on the risk profile. Then see how the read broker works on the line of business and OT page.

Explore this topic

Sources were read on 15 October 2026 unless a line says otherwise. [vendor] marks a vendor's claim about its own product or customer.

Named deployments:

Honeywell, commercial launch of Experion Operations Assistant, 19 March 2026 [vendor].
Hydrocarbon Processing, Honeywell and TotalEnergies pilot, November 2025 [vendor].
World Oil, Aker BP expands its AI-first strategy with Cognite Atlas AI, 16 September 2025 [vendor].
Portal Tela, C3 AI agents at Shell, 5 June 2026. Machine-written, no source named. Unverified.

Vendor agents for this process:

Vendor	Agent or feature	Date	Status as of 15 October 2026
IBM	Maximo Condition Insight	5 December 2025	Announced. Automatic work orders "soon" [vendor]
IBM	Maximo Application Suite 9.2, MCP server for outside agents	25 June 2026	Announced [vendor]
SAP	Maintenance Planner Agent	Planned for Q4 2025	Not confirmed. SAP pages refused access [vendor]
Seeq	Agent Q and Build Your Own Agent	3 March 2026	Announced [vendor]
Emerson	AspenTech AVA, four advisors	11 May 2026	Launched. Advisors not named [vendor]
Interloc	Interloc AI agents on Maximo, with approval steps and audit trails	Date not found	Announced [vendor]
Siemens	Senseye maintenance copilot. Pilots reported 25 percent less reactive maintenance time	Date not found	Not confirmed [vendor]

Vendor: IBM
Agent or feature: Maximo Condition Insight
Date: 5 December 2025
Status as of 15 October 2026: Announced. Automatic work orders "soon" [vendor]

Vendor: IBM
Agent or feature: Maximo Application Suite 9.2, MCP server for outside agents
Date: 25 June 2026
Status as of 15 October 2026: Announced [vendor]

Vendor: SAP
Agent or feature: Maintenance Planner Agent
Date: Planned for Q4 2025
Status as of 15 October 2026: Not confirmed. SAP pages refused access [vendor]

Vendor: Seeq
Agent or feature: Agent Q and Build Your Own Agent
Date: 3 March 2026
Status as of 15 October 2026: Announced [vendor]

Vendor: Emerson
Agent or feature: AspenTech AVA, four advisors
Date: 11 May 2026
Status as of 15 October 2026: Launched. Advisors not named [vendor]

Vendor: Interloc
Agent or feature: Interloc AI agents on Maximo, with approval steps and audit trails
Date: Date not found
Status as of 15 October 2026: Announced [vendor]

Vendor: Siemens
Agent or feature: Senseye maintenance copilot. Pilots reported 25 percent less reactive maintenance time
Date: Date not found
Status as of 15 October 2026: Not confirmed [vendor]

Context:

Equinor, AI saved USD 130 million, 7 January 2026. Equinor monitors over 700 rotating machines with 24,000 sensors, mainly with traditional machine learning.
EDF and Mistral partnership, 28 May 2026. Agents to query the technical memory of the nuclear fleet, on sovereign hosting, not for plant control.
K&L Gates on the AER automated bidding bulletin, 10 August 2026. The participant owns every bid.
IEA, Energy and AI, 2025. Reviewed 18 September 2026.
Race for 2030, AgenticGrid at Broken Hill, undated, accessed 14 October 2026, and T&D World, AWS and Duke Energy, 18 September 2026. Both show the pattern where the agent plans, tools calculate and an engineer decides.