Use of Classification Techniques to Design Laser Cutting Processes

In the field of manufacturing engineering, process designers conduct numerical simulation experiments to observe the impact of varying input parameters on certain outputs of a production process. The disadvantage of these simulations is that they are very time consuming and their results do not help to fully understand the underlying process. For instance, a common problem in planning processes is the choice of an appropriate machine parameter set that results in desirable process outputs. One way to overcome this problem is to use data mining techniques that extract previously unknown but valuable knowledge from simulation results. Our research examines the use of such techniques within the field of Virtual Production Intelligence (VPI). This paper proposes a novel approach for applying machine learning models, namely classification and regression trees, to design a laser cutting process. The evaluation shows that the models accurately identify regions in the multidimensional parameter space that increase the quality of the process (i.e. high cut quality). We implemented the models in the web-based VPI-platform, where the user is able to gain valuable insights into the laser cutting process with the aim of optimizing it.