A Prescriptive Analysis Tool for Improving Manufacturing Processes
16th WCEAM Proceedings Lecture Notes in Mechanical Engineering(2023)
Abstract
Recently there has been an increase in the development and use of Digital Twins in the manufacturing industry. Different kind of models can be developed depending on the use that is required. There can be very specific and detailed Digital Twins from a particular machine/process oriented to condition monitoring or tool wear degradation, and also others from a full manufacturing plant oriented to planning and production. In any case, each of these models will involve a lot of parameters (of different kind depending on the model) and, a priori, it is not easy to evaluate the effect a change in one of them will produce in the final result or how they are correlated among them. This is critical to optimally parametrize a machine or to optimally plan the production of a plant. The current paper presents a tool that allows the testing and ranking of multiple simulations from a Digital Twin having been saved previously in a standard format (namely FMU) for physics-based models or other kind of formats (ONNX for example) for data-driven approaches. This prescriptive analysis tool is thus a simulation, optimization and prescription tool. The presented tool will have available a set of models and it will allow the configuration of scenarios by defining variations of both, the input signals and the parameters of the model. The variations of the parameters the user wants to modify are generated by defining the range of values to be covered and the step. The scenarios, in turn, are generated with all the possible combinations of the defined variations of parameters and inputs. One scenario per combination. The tool will then perform exhaustive simulations with all the defined scenarios. One simulation per scenario. After this an evaluation function can be defined, with which the simulations results can be evaluated, that will result in a ranking of all the simulations. So, the tool will not only allow to obtain the best scenario, but also stablish some limits on parameters to guarantee a specific value of another one, or even check the worst scenario. The tool will also have some visualization options, allowing analysing the outputs of every simulation in detail and performing some graphs to visually present the obtained ranking. The prescription obtained by this simulation tool can be later used to configure a machine or to change some production parameters to optimize the final results.
MoreTranslated text
AI Read Science
Must-Reading Tree
Example
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
Generate MRT to find the research sequence of this paper
Chat Paper
Summary is being generated by the instructions you defined