Reconfigurable manufacturing system scheduling: a deep reinforcement learning approach

Reconfigurable Manufacturing Systems (RMS) bring new possibilities toward meeting demand fluctuations while, at the same time, challenges scheduling efficiency. This paper presents a novel approach that, for problem of RMS on multiple products, finds dynamic control policy via group deep reinforcement learning agents. These teamed agents, embedded with shared value decomposition network, aim minimising make-span constant updating order by guiding automated guided vehicles to move modules machine, raw materials, and finished products inside system.