A Prescriptive Analysis Tool for Improving Manufacturing Processes

Recently there has been an increase in the development and use of Digital Twins manufacturing industry. Different kind models can be developed depending on that is required. There very specific detailed from a particular machine/process oriented to condition monitoring or tool wear degradation, also others full plant planning production. In any case, each these will involve lot parameters (of different model) and, priori, it not easy evaluate effect change one them produce final result how they are correlated among them. This critical optimally parametrize machine plan production plant. The current paper presents allows testing ranking multiple simulations Twin having saved previously standard format (namely FMU) for physics-based other formats (ONNX example) data-driven approaches. prescriptive analysis thus simulation, optimization prescription tool. presented have available set allow configuration scenarios by defining variations both, input signals model. user wants modify generated range values covered step. scenarios, turn, with all possible combinations defined inputs. One scenario per combination. then perform exhaustive scenarios. simulation scenario. After this evaluation function defined, which results evaluated, simulations. So, only obtain best scenario, but stablish some limits guarantee value another one, even check worst visualization options, allowing analysing outputs every detail performing graphs visually present obtained ranking. later used configure optimize results.