Improving Parameter Traceability in Model-Based Product Development Processes through Integration of Python-Based Domain Models

Abstract

In today's product development processes (PDP), the complexity that companies want to master is growing as markets expand and interdisciplinarity increases. In previous and current research, approaches and methods are being developed to achieve a validated and verified product with the corresponding estimates of cost development as early as possible in the PDP. One of today's research approaches is focusing on modelling parameter and process dependencies that can be used to identify affected stakeholders, models, and process steps in the event of a potential change in a digital PDP so that far-reaching changes can be made efficiently. By the mentioned approach, the traceability of affected parameters into some domain models is possible, such as Microsoft Excel, an analysis into script-based domain models such as Python-based models are not established so far. The already established approach contains a system model in which a digital PDP for fiber-reinforced plastics (FRP) based on load path optimized local reinforced semi-finished products, more specifically tailored textiles (TT), is present. TT offer promising potential to reduce resulting waste compared to conventional FRP. In addition to the process steps, technical parameters from other expert models already exist in the system model and the stakeholders involved are created as roles. The present work will extend this approach with a proprietary approach by which Python-based simulation models can be implemented and whose model-specific parameters can also be efficiently analyzed and traced along the digital PDP. The method presented in this paper is intended to adapt a Python script-based production model for tailored fiber placement (TFP) to the existing system model. By extending the methodology, it is possible to identify parameters right down to the Python models that are relevant for the validation of requirements in the digital PDP. Furthermore, these parameters can also be affected by possible changes and can be adapted efficiently.