Energy

Service Diagnosis Agent

A colleague had the same fault two years ago. The trouble is, nobody knows what to search for.

Finds a similar past case for the fault pattern, with its documentation, and suggests a diagnostic path.

Open agent
Public administration

Four processes that take up the most time today

Developed in Germany · Hosted in the EU

Developed in Germany · Hosted in the EU or on-premises

Delivered together with

STACKIT Sovereign hosting in Germany, single-tenant from the first user.
Rödl Legal, tax and compliance expertise for legally sound deployment.
4P Consulting Advisory and implementation for the public sector, starting with the AI readiness assessment.

Four hours of diagnosis on a machine whose fault pattern has long been on record in the company's own service history.

The technician is standing at the machine; the fault pattern is an error code plus a particular behaviour on start-up. The service history holds hundreds of reports on the series, the manufacturer's manual runs to eighty pages, and nobody likes searching through either on a tablet in the field. So the measurements start again from scratch, even though the route to the cause is documented.

1fault reportAn inverter reports error code F41 and the machine shuts down with a fault.
380past cases for the seriesThat is how many service reports on this series are in the history.
3hits for the codeThree of them name the same code, and two also record the measurements.
4 hourson-site diagnosisWithout these reports, the cause is narrowed down again, one measuring point at a time.

45% faster fault diagnosis means the technician starts at the measuring point where two past cases ended, rather than from zero.

The 45% gain in diagnosis speed is this agent's headline figure. The 380 past cases, the three hits and the four hours on site sketch a typical service case and were not measured.

Three steps through the process.

For the reported fault pattern, the agent pulls out the similar past cases with the relevant manual section and suggests a diagnostic path; the technician establishes the actual cause.

Compare the fault pattern with past cases

The agent takes the error code, series and description from the fault report and uses them to search the service reports for the series. It shows the reports that record the same code on the same series and marks which of them also record measurements.

Fault report SM-9364, 380 reports searched, 3 hits for code F41

Match the documentation to the assembly

For each hit, the agent adds the relevant section of the manufacturer's manual and the circuit diagram for the assembly. It names the chapter and page, not the whole document.

Manual for U3 series inverters, chapter 7.4, pages 52 to 55

Suggest a diagnostic path for the machine

From the hits that record measurements, the agent builds a sequence of measuring points, starting with the one that led to the cause in two cases. The technician works through them and enters each reading.

Diagnostic path for SM-9364, 5 measuring points, first point DC link voltage

What the agent does not do.

It does not establish the cause.

It matches the fault pattern to past cases and suggests the measuring points. Service engineering establishes what is faulty on this machine, at the unit itself.

It does not clear the machine for operation.

After the assembly is replaced, it records the case with measurements and references. The responsible technician signs off the return to service.

A past case that records only an assembly replacement is of little help to it.

If the report lacks the measurement and the behaviour before the fault, the case still appears among the hits for code F41 but leads to no measuring point. How useful the hits are depends on how reports are written in the field.

Connected systems

  • Field service management
  • maintenance system
  • document repository

Operations

  • Deployment modelEU cloud or private cloud
  • Data processingGDPR-compliant, no model training on your data
  • Responsible teamsService, field service

The questions that come up before the first meeting.

How quickly is the Service Diagnosis Agent ready to use?

A pilot typically runs on your own data within 6 to 8 weeks. The agent connects to your existing systems rather than running as a separate, parallel environment.

Where is the data processed?

EU cloud or private cloud. Processing is GDPR-compliant and models are never trained on your data. Details in the Trust Center.

Where does the figure “45% faster fault diagnosis” come from?

It is based on experience from comparable projects and process benchmarks, not a commitment. We work out the realistic effect for your organisation during the assessment before the pilot.

Related agents.

Intake review, case 2026/04711

Skip to content

Page footer