Modelling for Decision-making in Dynamical Line-less Assembly Systems

Abstract

The shift towards electro mobility has a significant impact on the entire value chain of the automotive industry. The parallel production of Internal Combustion Engine, Battery Electric and Plug-in Hybrid Electric Vehicles con-fronts manufacturers with major challenges. This requires flexible production systems that are able to operate efficiently even under the influence of dynamic conditions. As could be seen in the COVID-19 pandemic, production systems must be able to react flexible and resilient even to these kinds of situations. Within the production system there is a multitude of temporally and spatially determined neuralgic points at which a reaction to dynamic conditions is possible and necessary. In order to make target-oriented decisions at these points, a large amount of data from many different sources is required as input parameters. The decisions can be made manually, semi-automatically or automatically by a human, a machine or a (software) system. Therefore it is essential that influences, input and output variables as well as decisions and the solution space are provided in a defined manner enabling all involved units to process the information. The paper therefore describes the first steps in modeling decisions for dynamic line-less assembly systems. Starting with an introduction to the topic and the multi-variant automotive production, the architecture of this kind of flexible assembly system is briefly described. Afterwards a description of the formal modeling of decisions in assembly systems is presented. Finally, after a discussion of the results, an outlook on the next steps and further research needs is given.