Digital Twin-enabled Simulation-optimization Framework for Autonomous Mobile Robot Logistics in Smart Manufacturing

Abstract

Digital twin technology is increasingly used to support the analysis and optimisation of smart manufacturing systems, particularly where flexible intra-plant logistics are required. This study develops a digital twin-enabled simulation-optimisation framework for improving material transportation between assembly and testing stations using Autonomous Mobile Robots (AMRs). A Linear Mathematical Model was formulated to allocate transportation tasks among three AMR types while considering trip demand, payload capacity, travel time, and system constraints. The optimisation problem was solved using the Simplex method, and the resulting logistics configuration was evaluated in a virtual twin environment developed with the DELMIA 3DEXPERIENCE platform. For 30 required trips per shift, the model allocated 12 trips to AMR-1, 10 trips to AMR-2, and 8 trips to AMR-3. The analytical transportation time decreased from 450 minutes under the existing Automatic Electrified Monorail System to 296 minutes under the optimised AMR configuration, corresponding to a 34.2% reduction in cycle time. The model also indicated improvements in average travel time, throughput, path length, and system efficiency under the stated assumptions. The virtual twin was used to visualise routing, congestion, and logistics behaviour within the modelled factory environment. Overall, the framework demonstrates how mathematical optimisation and digital twin simulation can be combined to support comparative evaluation and decision-making for intra-plant logistics before physical implementation.