Dismantling sequence planning of end-of-life products considering parts recovery strategies

Efficient disassembly and resource recovery of end-of-life products are increasingly important to circular and green manufacturing. This paper proposes a recovery-strategy-driven disassembly sequence planning method. A disassembly hierarchical information graph and a disassembly priority relationship matrix (DPRM) jointly represent structural constraints and recovery attributes. Component recycling potential is evaluated by integrating the analytic network process (ANP), Criteria Importance Through Intercriteria Correlation (CRITIC), grey relational analysis (GRA), and the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), with GRA replacing Euclidean distance in TOPSIS, and the resulting scores determine recovery strategies. A dual-objective optimization model is developed to minimize disassembly time and maximize net economic benefit. The model is solved by a hybrid ant colony optimization-genetic algorithm with pheromone-guided initialization, constraint-aware crossover and mutation, and Pareto non-dominated sorting. An electric vehicle traction battery pack case study shows that incorporating recovery strategies reduces disassembly time from 1,694.10 s to 1,679.25 s and increases net economic benefit from CNY 1,096.30 to CNY 1,118.69. Compared with conventional genetic and ant colony algorithms, the proposed method shows favorable case-specific convergence behavior and objective-space trade-offs, together with a comparatively steady late-stage hypervolume trajectory.