The most important points at a glance
AI-based modeling makes production systems usable as a verifiable model from process descriptions and operating data. What matters is not automation per se, but a understandable representation with which scenarios can be analyzed and decisions prepared. Professionals must always review the AI's suggestions to ensure rules, exceptions and real limitations are accurately reflected.
From our practice, the value lies not in automating as much as possible, but in creating a model that can be trusted in operation. For this purpose the symas GmbH combines software development, model building, simulation and AI into a comprehensible solution.
What does AI-based modeling of production systems mean?
A production model structurally maps relevant machines, workstations, orders, material flows, resources, rules and dependencies. AI helps prepare, organize and convert these elements from existing process descriptions and operating data into model building blocks. The goal is a robust digital representation, not an opaque automation.
What questions must the model answer?
Before building, it must be clear which question the model should answer. Whether it concerns capacity, bottlenecks, sequences, staffing, inventories or an investment, exactly this question determines how detailed the model must be. Everything that is not needed for the answer can be deliberately omitted.
- Check capacity and utilization
- Analyze bottlenecks and sequences
- Better estimate staffing needs
- Evaluate inventories
- Prepare investments with scenarios
How does AI support the construction of the model?
AI can take on or prepare several steps: structure information from documents and process descriptions, detect patterns in operating data, propose model building blocks or generate variants for experiments. The actual technical assessment remains with the professionals, because only they can determine whether exceptions, priority rules and real constraints are correctly considered.
Thomas Kehl collaborates with research and application partners in the field of AI-based modeling of production systems. A publicly documented Innosuisse workshop connected symas, an application partner and OST – Ostschweizer Fachhochschule institutes. The focus was on the Swiss industry’s competitiveness through practical modeling.
How is a robust result tested?
- Document data sources and assumptions
- Compare the model with known operating situations
- Have professionals assess behavior for plausibility
- Carry out sensitivity analyses to understand the response to changed inputs
| Method | Task | Project benefit |
|---|---|---|
| AI | Identifies, structures and derives | Speeds up model development |
| Simulation | Executes defined scenarios reproducibly | Makes impacts comparable |
| Professional review | Evaluates rules, exceptions and plausibility | Ensures traceability |
What matters in practice
A model is only as good as its assumptions and data sources. If you do not document them and align them with real operating situations, you risk results that appear precise but are technically weak.
Why are AI and simulation stronger together?
AI and simulation fulfill different tasks. AI helps with recognition, structuring and derivation, while simulation runs defined scenarios reproducibly and makes their effects comparable. Together they accelerate model development without sacrificing traceability.
For symas, this combination is central because it links understandable production models with robust decision bases. This leads to solutions that work in operation and remain maintainable and explicable.
Frequently asked questions
When is AI particularly worthwhile in modeling a production system?
Does AI replace professionals in modeling?
How can one tell if a model is truly reliable?
How does simulation differ from AI in this context?
Anyone who wants to use AI-based modeling sensibly does not need a black-box system, but a clear goal, clean data and technical validation. This is where the difference between a quick idea and a robust model for production systems lies.
The symas GmbH supports companies with AI-based modeling, simulation and software development from a single source. When process knowledge should become a viable model, this structured approach is often the fastest path to better decisions.
Möchten Sie ein belastbares Modell für Ihre Produktion?
symas GmbH, Oberegg