Impact of the Cluster Topology of Autonomously Controlled Material Flow Networks on the Performance of a Logistic System

Abstract

The planning and control of production processes are increasingly centralized, requiring advance scheduling. Changes like rush jobs often lead to costly rescheduling. Autonomous production systems provide a solution by allowing logistic objects to make independent decisions based on current conditions. This thesis examines the role of cluster topology—identified through clustering algorithms—in enhancing autonomous control. By investigating decentralized job sequencing within these clusters, simulations reveal that considering cluster topology improves logistical outcomes. Additionally, size and number of clusters prove to be important parameters for further research into autonomous production systems.