CIRP CMS 2022: Effective Simulation Software for Production Technology


Chrismarie Enslin from the Cybernetics Lab (IMA & IfU) focuses her research on Machine Learning and Reinforcement Learning in the application domain of digital manufacturing technology and had the opportunity to share her approaches with colleagues from all over the world at this year's CIRP CMS 2022. The conference was the 55th edition of the CIRP Conference on Manufacturing Systems and was held in hybrid format from 29.06 - 01.07.2022 in Lugano, in Switzerland. This year it was hosted by the Innovative Technologies Department of the University of Applied Sciences of Southern Switzerland (SUPSI). Enslin gave a talk on the topic of “ProdSim: An Open-source Python Package for Generating High-resolution Synthetic Manufacturing Data on Product, Machine and Shop-Floor Levels”.

  Lugano © SUPSI

What impressions did you personally take away from the conference?

I attended the CIRP CMS 2022 conference online, which was very convenient and allowed me to still share my paper, even though I couldn’t be at the conference in person. The full effect of the conference was probably better experienced in Lugano, in person. I presented in the Smart Devices, Sensors and Networks session on the last day of the conference. My topic fit into the session fairly well and the questions were very fitting to the topic. Some of the questions were about the usage of the software package I presented, which felt like a confirmation that my message came across clearly.


How can your presentation topic be briefly broken down and what is its relevance for the scientific community?

The topic that was presented is ProdSim, a simulation software tool that can be used for production line simulation. It is an easy to use Python package that is available in an open source format at This software can generate production data on different levels of the production line, which is the product level, the machine level and the shop-floor level. The relevance for the scientific community is that data exactly fitting to a research problem can be generated, where data of this nature is often not found in the real world. Data for the different levels of a production line is often captured separately and can cause a disjointed group of datasets which is hard to unify.

How is your research topic embedded in the IoP complex?

The research topic fits into Workstream B3.II. Its goal is to use data generated in this way to investigate predictive quality on production lines. With full control over the input feature and the influential features in production, a causal model can be determined for the production line, with a ground truth to learn it from. This is a valuable asset in the understanding of production lines and where to most influential points are to increase the quality of produced products.