Use Case · Plant Engineering · 2026

How project folders become usable knowledge.

A mid-sized plant engineering firm · 10 designers, 100 projects · POC in one week, implementation step by step.

Keep engineering knowledge, even with staff turnover.

Baseline

Engineering grown over years.

Gas panel systems, pressure vessels, process components – the experience sits in folders and in the heads of individual designers.

Design FTE10engineering capacity
Projects completed~100history ~6 years
Queries / year~150quotes & clarifications
Project data~20 GBCAD, calculations, docs, emails

With every staff change, part of the experience disappears. On a 10-person team that has an outsized effect.

Problem

Three ways today. None scales.

With every new query the designer decides, with no guarantee that similar experience is found again.

Way 1Start from scratch
  • expensive and slow
  • no learning from the past
  • risks rediscovered
Way 2Get lucky by chance
  • a gamble, often unsuccessful
  • 2–5 h of searching per query
  • experience not systematically usable
Way 3An experienced colleague remembers
  • knowledge of individuals
  • not scalable
  • lost with turnover

At ~150 queries × ~3 h of searching, roughly 450 h of engineering time per year go without direct value, about 11 weeks of one FTE.

Data landscape

Where the knowledge sits today.

Four sources, four structures, no common index – the designer has to search each source separately.

Source 01
Network drives~100 folders · ~20 GB

Project folders with CAD, calculations, documentation. Structure differs per project.

Source 02
Emails & attachments~500 threads / project

Clarifications with customers, suppliers, authorities. Decisions often only documented here.

Source 03
ERP systemlimited · records only

Invoices and delivery notes. Supplier link visible, hardly any technical project knowledge.

Source 04
Engineering headsimplicit

Experience of individual designers – lessons learned, risks, supplier ratings. Recorded nowhere.

Each source useful on its own, together unfindable. The agent builds the missing index across all four sources.

Output

What the engineer sees in the end.

A structured dossier per project – directly searchable, with risks, suppliers and KPIs from history.

SEARCHmedium:hydrogen & certification:ATEX & pressure:>30bar
3 hits · showing top 1
PROJECT 2023-047
Gas panel system for a hydrogen plant
✓ Docs complete
CRITERIA
Medium: Hydrogen (gaseous)
Pressure / temp: 35 bar / 120 °C
Material: 1.4571 (austenitic stainless)
Application: Outdoor · temperate climate
Customer: Research institute
Components: Panel system · pressure control · filter
AD 2000PED 2014/68/EU Mod. HATEX Zone 1
KPI · FROM PROJECT HISTORY
8 monthsduration (plan: 6)
+6 weeksdelivery delay 1.4571
480 hengineering effort
LESSONS LEARNED
  • ATEX zone clarification iterative – align with the operator up front
  • Plan stainless 1.4571 with a 4–6 week lead time
  • Pressure regulators: not all standard models are approved for H₂ – check up front
Impact

Search reduced. Knowledge kept. Risk visible early.

Search & research35 % → 20 %of the engineering day. ~15 points of freed capacity per designer.
Knowledgeindependentexperience from 100 projects stays available, even with staff turnover.
Riskvisible earlylessons learned prevent repeat mistakes, risks visible at the quoting stage.
Use case as PDF

Download the full use case.

Including architecture, the full output example and roadmap. You get the PDF right after submitting, plus a short email from me.

Next step

Start with one of your project folders.

We test the value with a real use case from your plant engineering, within one week, with your data. If the POC doesn't convince you, it costs nothing.

Book a 30-minute call →