Automatic Design of Dynamic Collaboration Strategies for Machines and Automated Guided Vehicles via Multiobjective Genetic Programming.

This article addresses the dynamic collaboration problem between processing machines and automated guided vehicles (AGVs) in flexible manufacturing systems (FMS), aiming to minimize both makespan and energy consumption. The authors propose a novel multiobjective genetic programming (GP) approach that evolves collaboration strategies by representing priority functions as GP trees, incorporating ten status statistics related to handling time and energy use. The method dynamically selects job-machine-AGV combinations during production simulation and is evaluated against 28 basic collaboration strategies (BCSs) across multiple training and testing scenarios. Results demonstrate that the GP-evolved strategies outperform BCSs in terms of adaptability, scalability, and solution quality, effectively balancing production efficiency and energy consumption in dynamic, large-scale manufacturing environments.