Motivation
Modern vehicles have up to 180 control units that go through various defined steps of digital value creation during production. These range from checking the installation, flashing the software and individual configuration according to customer requirements through to functional testing and commissioning. The sequence, boundary conditions and dependencies are decisive factors for successful vehicle commissioning. As the scope of digital value creation is constantly growing, a solution is being sought that will make processes more dynamic, individual and intelligent in the future.
The overall aim of this project is to develop an integrated AI methodology for the intelligent automation and optimization of digital value creation processes in automotive production.
At its core, the project pursues three interlocking objectives: Firstly, the automated generation of precedence graphs using artificial intelligence to capture the complex dependencies and sequences in the digital value chains without the need for time-consuming manual modeling. Secondly, the dynamic creation of vehicle-specific OTX checklists using Generative AI, which adapt individually to customer configurations and variant requirements. Thirdly, the implementation of an integrated approach for proactive anomaly detection, which identifies deviations in the digital value chain at an early stage and thus prevents production errors and field failures.
Methodology
The project is based on a three-stage, integrated methodology that systematically uses artificial intelligence to analyze and optimize digital value creation in automotive production. This methodology combines data-driven process modeling, generative AI technologies and intelligent anomaly detection into a coherent solution approach.
Project funding
The KIDWAP project is funded by Dr. Ing. h.c. F. Porsche AG.