Skip to content
Service & spare partsFor every industry

Spot serial failures in the field early

demi finds patterns across units, series and customers and gives engineering solid cases instead of one-off reports.

Today

The same failure is repaired in the field again and again, and nobody connects the cases. The reports are scattered across tickets, e-mails and return slips. Engineering hears about it late, and then only as an anecdote.

An example

At Talwerk Antriebstechnik the agent shows 37 shaft seal failures on TG 80 gearboxes from one serial range. All the seals come from the same supplier batch. The agent drafts a change request with the 37 cases as evidence; the head of engineering reviews and approves it.
Fictional company, typical workflow.

Your data, wherever it lives

Field failures are scattered today across service tickets and reports, technicians' e-mails, return slips, often only as scans, and complaints in the QM system; BOMs sit in the PLM, serial numbers and batches in the ERP. demi reads, understands and links them per component, serial number and batch, with your rights. The AI ties every new case to the earlier ones at once and shows a pattern with evidence while it is still small, and the head of engineering decides on the change.

Where the data lives today

  • Tickets
  • E-mails
  • Scans & paper
  • QM system
  • PLM
  • ERP
Data LayerOne view of everything, with your rights

How demi solves it

Three layers in this order: connect the data first, then the rights, then the AI.

  1. 1 · ConnectData Layer

    Data sources in this case: the CMMS, QMS, PLM and ERP. demi replaces nothing and links service reports, tickets, returned parts and complaints to the BOMs of the affected units. Every failure is tied to the component, the serial number and the supplier batch.

  2. 2 · PermitPermission Layer

    Engineering sees the patterns in aggregated form, and only service sees customer names and contacts. The head of engineering approves the change request before anything is written back, and this is recorded.

  3. 3 · ActIntelligence Layer

    The agent groups failures by component, serial range and supplier and counts the cases exactly from the records, each one linked. From that it drafts the feedback to engineering with all the evidence.

How it runs

  1. 1

    Link service, complaint and return data to the BOMs

  2. 2

    Group failures by component, series and batch

  3. 3

    Show striking patterns with case counts and evidence

  4. 4

    Draft a change request for engineering and get it approved

What changes

  • Root causes fixed in the design

  • Fewer repeat repairs in the field

  • Engineering works with facts

What does it bring you?

Check it with your own numbers

Every starting value is an assumption you can change. The calculation sits right below, step by step.

Assumption, editable

Your numbers

Result

+86 hfreed per year

Assumptions and calculation

demi spots patterns across machines and series and bundles the cases; you hand them to engineering.

6 h → 4.2 h per case

How does it run today?

What is already digital saves less: the factor lowers the share demi brings.

How we calculate

4 field reviews × 6 h × 30% × 1.0 (manual) × 12 months = 86 h

Value per year, before demi costs

Freed per year+86 h

Assumption: labor cost of $48.62 per hour in US manufacturing in June 2026 (BLS), plus workplace, IT and overhead.

= Value per year, before demi costs

+$5,590

gained per year

An estimate, not a promise. The starting values are assumptions for an industrial company with 150 to 500 people.

Combine with more cases and get the package

Questions about this use case

How does demi detect a pattern?

The agent assigns every failure to the component, the serial number and, through the BOM, the supplier batch. When cases cluster clearly more in one group than in the rest, it shows that as a pattern. The counts come from your records, and every case is linked so you can check it.

Does engineering see our customer names?

Not unless you want it to. Engineering sees component, series, operating time and finding. Customer names and contacts stay with service, as your org chart defines.

Can demi create a change request?

Yes, as a draft with all the evidence. An authorized person reviews and approves it; only then may the request be written back. The log records who approved it.

Our service reports are free text. Is that enough?

Yes. The agent reads free text and assigns finding and component, even when technicians use different terms for the same thing. It flags uncertain matches so someone in service can confirm them.

See this case with your data

In the demo we show you the workflow on your own systems.

Your configurator

No use cases yet

Your configurator is still empty

Add a use case from any card or case page and see what it brings per year, with the calculation. Free, no sign-up, no commitment: try it out and change any number.

Browse use cases