Collective Intelligence Application in a Kitting Picking Zone of the Automotive Industry

Abstract

The durability of an automobile factory depends on its flexibility and its evolution capacity to meet market expectations. These expectations tend increasingly to the vehicles’ customization. Therefore, automobile factories may be able to manufacture several vehicle models on the same assembly line. It makes automobile manufacturers face big logistic challenges in their production sites. They must be capable of simplifying, synchronizing and proposing intelligent and flexible logistic flow. Thus, digital tools for decision support are needed. This paper aims to propose an architecture to model the logistic process of supplying materials to the assembly line as a multia-gent system. Thus, multiagent learning and collective intelligence techniques can be applied to guarantee a good performance of the process. The case study focuses on a kitting picking zone from a Renault production site which manufactures six different vehicle models, each one with its variants.