Experimental Evaluation of AGV Dispatching Methods in an Agent-Based Simulation Environment and a Digital Twin

A critical part of Automated Material Handling Systems (AMHS) is the task allocation and dispatching strategy employed. In order to better understand investigate this component, we here present an extensive experimental evaluation three different approaches with randomly generated, as well custom designed, environment configurations. While previous studies typically focused on use cases based highly constrained navigation capabilities (e.g., overhead hoist transport systems), our built around mobile, free-ranging vehicles, i.e., Autonomous Mobile Robots (AMR) that are gaining popularity in a broad range applications. Consequently, experiments conducted using microscopic agent-based simulation, instead more common discrete-event simulation model. Dispatching methods often assumption asynchronous event-based model, vehicles trigger cascade individual decisions, e.g., when reaching intersections. We find does not translate very fleet mobile systems can change direction at any time. With mind, formulations known suited for synchronous decisions. The Stable Marriage Problem (SMP) Linear Sum Assignment (LSAP). matching assignment algorithms compute actual selected evaluated multi-agent environment. To integrate centralised management, digital twin concept proposed implemented. By approach, management independent implementation specific agents, allowing quickly adapt other simulation-based or real application scenarios. For evaluation, two new performance measures related efficiency material handling system proposed, Travel Efficiency Throughput Effort. indicates reassignment mechanisms method help increase overall fleet. did significant differences absolute terms throughput rate. Additionally, difference between SMP- LSAP-based seems negligible. conclude discussion, where consider potential confounding factors relate findings previously reported results found literature.