Improved Proximal Policy Optimization Algorithm for Multi-objective Disassembly Line Balancing Considering Hazardous Tasks

As disassembly lines evolve, keeping workers safe from all potential dangers has become paramount. This work proposes an improved approach, Proximal Policy Optimization for Disassembly Line(PPO-DL), to address Multi-objective Dis-assembly Line Balancing Problem(MDLBP) that integrates task switching time and hazardous tasks. The MDLBP model aims to optimize disassembly net profit, balance workstation utilization, and mitigate penalties from hazardous tasks. The proposed PPO-DL modifies the action space of PPO to accommodate MD LBP characteristics, ensuring efficient learning by restricting actions to valid ranges through action masking. Comparative experiments with PPO, Q-learning, and A2C demonstrate PPO-DL's superiority in finding near-optimal solutions across various scenarios.