Data-Driven Adaptive Processes - A Potential Enabler for Flexible and Versatile Automotive Body Shops

摘要

Automated robotic processes are difficult to stabilize, which requires a high ramp up effort and complex variant-specific geometry jigs especially in automotive body shops. This effort increases by higher variance and shorter product life cycles, a current trend in the automotive industry. To avoid this, the approach of this paper aims to use self-learning data analytics models to adapt process controls according to features of individual joining partners. Therefore, as a first step, this paper shows an overall concept for intelligent integrated adjustment planning in body shops as well as a dive-in into adaptive welding processes.