Abstract: Disassembly sequence planning, as the first step to achieve product recycling, plays a decisive role in the quality and efficiency of recycling waste products. This paper addresses the three core issues of modelling methods, solution algorithms and uncertainty handling methods in demolition sequence planning. Firstly, the current mainstream modelling approaches for disassembly sequences and the disassembly scenarios tackled by various modelling approaches are collated and analyzed. Secondly, from the perspective of decision solving, the characteristics of three different solving methods, namely mathematical planning methods, intelligent optimization algorithms and reinforcement learning algorithms, as well as the challenges faced when making optimal decisions, are discussed. And the solving methods of single-objective optimization problems and multi-objective optimization problems in the field of disassembly are systematically analyzed. Thirdly, the strategies for dealing with uncertain disassembly problems in the disassembly process are summarized. The advantages and disadvantages of the different strategies are analyzed. Finally, the current state of research on disassembly sequence planning is summarized, and future research directions are discussed.

Key words: disassembly sequence planning, recycling, intelligent optimization, reinforcement learning, uncertainty optimization

摘要： 拆卸序列规划作为实现产品回收再利用的首要步骤，对废弃产品的回收质量和效率起着决定性作用。针对拆卸序列规划中的建模方法、求解算法和不确定性处理方法三个核心问题，整理和分析了当前拆卸序列的主流建模方法以及各种建模方法应对的拆卸场景；从决策求解的角度出发，探讨了数学规划方法、智能优化算法和强化学习算法这三种不同求解方法的特点以及进行最优化决策时所面临的挑战，系统分析了当前拆卸领域中单目标优化问题和多目标优化问题的求解方式。总结了在处理拆卸过程中针对不确定性拆卸问题时的处理策略，并分析了不同策略的优劣；总结了拆卸序列规划的研究现状，并讨论了未来研究方向。

关键词: 拆卸序列规划, 回收再利用, 智能优化, 强化学习, 不确定性优化