A passed acceptance test is a snapshot of a sample: it holds for the section that was checked and for the moment it was checked. Anyone turning that into a number has to say where it came from. So every case carries the same grid — and where a figure is missing, it says so.
Industry · project type · scope of the model · baseline before the project · result afterwards · assumptions marked as open. Without a baseline no improvement is stated — only what holds without comparison: scope, effort, duration. An improvement without a starting value is not a measurement but an impression.
No baseline, no number. No measurement note, no effect claim.We name customers only with written consent. Six of the eight cases therefore carry their name. Two stand without one — at the customer's request, complete in substance.
A logo without consent is not evidence, it is a claim about a third party.Where a baseline is documented, the full grid stands and the effect may be named. Where none is documented, only figures that hold without a point of comparison are given.
Building materials - cement-bonded particleboard · Retrofit of an automatic board cutting line
Existing line with Beckhoff control (TwinCAT 3), operator interface and ERP link. Faulty sensor behavior was not detectable in operation.
Core software commissioned in five days. Actuators and sensors taken over from EPLAN, the process recorded on video and modeled. Monitoring exposed hidden sensor faults.
A trade article reports 25% higher throughput for a line with identical figures, without naming the company. We do not state that number until the attribution is confirmed.
Welding automation · New robotic welding cell with virtual commissioning
Usual commissioning time for a machine of this kind: three person-months.
Two weeks to the finished process model, one week of virtual commissioning. One engineer modeled the machine without classic PLC experience.
The three person-months are igm's experience figure, not a measured comparison on the same machine.
Automotive · Retrofit of an end-of-line test station for quality control
Outdated proprietary control mastered only by individual specialists. A necessary change could not be made because the people responsible were no longer available.
A few weeks from analysis to commissioning. Migration to an open, standardized platform. Higher quality, lower complexity and shorter cycle times. A second test station followed.
The shorter cycle times are not quantified in the source.
Beverage industry · New juice blending and filling plant
Manual process without continuous recording of recipe, parameters and cycle.
Process model in two weeks, tested and commissioned virtually. The real plant was built afterwards to the tested specification. Recipe, parameters and cycle are traceable per batch.
Traceability covers recipe, parameters and cycle, not the entire plant.
Automotive supply, tool making · New machine controlling two intelligent press tools
High-strength sheet in difficult grades caused elevated scrap rates.
Less scrap, higher forming accuracy and better quality. 250 ms process accuracy at high speed, without special code.
The scrap reduction is not quantified. Electrical design and installation came from Duatec Automation GmbH, so the effect cannot be attributed to the method alone.
Plant engineering, raw material handling and bulk solids · Modernization of an existing plant
No documented baseline.
Under 80 hours of engineering effort for the complete modernization. The process was visible and verifiable before commissioning.
No further figures are available. No improvement is claimed without a documented baseline.
Automotive · Proof of concept, full automation of a camshaft assembly line
No documented baseline.
Under two weeks from the first workshop to live commissioning. The behavior was known before the start of production.
A proof of concept evidences feasibility, not series operation.
Test equipment · New machine
No documented baseline.
Commissioned in two days. Six years of operation without retraining and without a service call.
The six years are an operating period, not a comparison with another solution.
These values do not come from a single case but from several completed projects. They are project-specific and therefore carry a condition.
Ranges from completed customer projects. The actual value is project-specific and is measured on your line during the assessment.
That is not caution, it is the statement itself: structure transfers, effect is measured.
What was measured on one line holds for that line. On yours it is measured anew — in the assessment, with a stated baseline.
All eight applications stand here — including the three without a documented baseline. They stand without an improvement claim instead of disappearing.
The industry of a case says where it happened — not that it would yield the same elsewhere.
It exists before the code. What these cases evidence is therefore the path, not a result explained in hindsight.
Verification asks: was it built right? Validation asks: was the right thing built? The cases above evidence the first question. The second was answered in each project by the operator. WhiteBox does not mean you can see the code. WhiteBox means conformity becomes decidable instead of a matter of judgment — and that is what a case here can evidence.
Selmo replaces neither commissioning nor experience — it reduces the space of possible causes. And what is checked is always the defined behavioral space: what is not modeled is not covered.
The question we like most
15–30 minutes. We bring the answer, not the presentation.