Achieving robustness via autonomous control: examples from production logistics

Autonomous control is the ability of a logistic object of information processing, decision rendering by its own and realization of the decision result. As it is necessary to equip logistic objects with necessary technologies and decision methods simulation studies were performed in order to evaluate the effectiveness of different decision strategies. In one example of a manufacturing scenario autonomous control is used as decentralized approach of allocating materials to customer orders. The data base and the decision strategy will be presented. As autonomous control is a decentralized control approach using so far unlocked flexibility potentials it is able to cope with the growing complexity in logistics processes. In order to also theoretically understand the relation of complexity, degree of autonomous control and performance, we employ graph coloring dynamics as a minimal model. This refers to the key question in logistics research aiming for the understanding of the interplay of logistics processes in networks. Firstly it is aimed to demonstrate this non-trivial dependency of degree of autonomous control, logistics system complexity and logistics target achievement using Graph Coloring (GC) dynamics. The focus lies on GC dynamics in order to analyze the interplay between the temporal pattern changing colors on nodes until the graph is solved. A suitable local performance measure in GC dynamics is the number of color changes. Secondly, we show the application of GC dynamics in order to predict network performance in dependence of the network architecture. Based on this fundamental understanding rules of the design of robust networks will be able to derive in the future.

Return