Waste electrical and electronic equipment (WEEE) not only occupies resources but also pollutes the environment. Disassembly and recycling of WEEE is an important part of sustainable manufacturing. However, the number of parts in WEEE is large and the relationship between parts is complex, which increases the difficulty of disassembly task planning. Therefore, this paper proposes a precedence graph based on associated parts and adopts a partial disassembly to reduce the number of disassembly tasks. To evaluate the green performance of the disassembly line, a partial disassembly line balancing model considering efficiency, profit, energy consumption, and hazard is proposed. Then, a multi-objective genetic simulated annealing algorithm is developed. To ensure the effectiveness of the algorithm, novel encoding, decoding, genetic operation, and simulated annealing operation based on problem constraints are designed. The proposed model and algorithm are applied to a waste refrigerator disassembly line in China, and the results show that the proposed algorithm is superior to the comparison algorithms. Compared with the original disassembly scheme, the new schemes have higher efficiency and profit, as well as lower energy consumption and hazard.

废弃电子电气设备（WEEE）不仅占用资源，而且污染环境。WEEE的拆卸和回收是可持续制造的重要组成部分。但是，WEEE中的零件数量很大，零件之间的关系也很复杂，这增加了拆卸任务计划的难度。因此，本文提出了一种基于相关零件的优先级图，并采用部分拆卸来减少拆卸任务的数量。为了评估拆卸线的绿色性能，提出了考虑效率，利润，能耗和危害的部分拆卸线平衡模型。然后，开发了一种多目标遗传模拟退火算法。为确保算法的有效性，采用新颖的编码，解码，遗传运算，设计了基于问题约束的模拟退火算法。将该模型和算法应用于我国的废旧冰箱拆装生产线，结果表明该算法优于比较算法。与原来的拆卸方案相比，新方案效率更高，利润更高，能耗和危害更低。