Human-machine collaboration in Industry 5.0: An energy-efficient collaborative approach for scheduling intelligent job shops considering automated guided vehicle transportation and worker flexibility

Automated guided vehicles (AGVs) and multi-skilled workers are widely employed in smart manufacturing factories within the context of Industry 5.0. Achieving effective collaboration among machines, AGVs, and multi-skilled workers is crucial for enhancing the production efficiency of intelligent workshops. However, existing studies on production scheduling have not adequately addressed this critical factor. This study investigates an energy-aware collaborative scheduling problem in an intelligent job shop environment, where a limited number of AGVs and workers with diverse skill sets are involved. The primary objective is to address the human–machine collaboration challenge among machines, AGVs, and multi-skilled workers. To achieve this, four distinct objectives are simultaneously optimized: the makespan, total worker cost, total idle time, and total energy consumed by machines and AGVs. An adaptive many-objective evolutionary algorithm is developed for tackling this complex NP-hard problem. In the algorithm, a five-layer collaborative encoding approach is utilized to represent the operations of jobs, machine processing speeds, worker assignments, AGV assignments, and AGV speeds. Additionally, fuzzy correlation entropy theory is integrated into the algorithmic framework and extended to serve as a dynamic fitness evaluation mechanism for assessing the quality of candidate solutions. A new reward function based on this fitness evaluation mechanism is designed to carry out adaptive meta-Lamarckian learning. This learning strategy enables the algorithm to adaptively select an appropriate local search method based on the performance improvement rate of the solutions. Extensive experiments confirm the superiority of our proposed algorithm by comparing it with three well-established many-objective algorithms. This study provides valuable insights and references for production enterprise managers to make effective scheduling decisions in intelligent manufacturing systems that require human–machine collaboration.