Abstract: To recycle and dispose the waste in production process from multiple factories efficiently, aiming at the system contained multi-factories and a recycling center, the decision makers in system were devided into two decision types. In type 1, the recycling center was the only decision maker, and two recycling models：independent recycling and joint recycling were developed. In type 2, the decision makers included all the factories and the recycling centers, and three recycling models：disperse recycling, integrated recycling and joint recycling were developed. To find the optimal disposal model, the system total cost minimization was used as target to construct formulas. In addition, a heuristic search algorithm was proposed to solve the joint recycling model. By changing the factory numbers and associated costs parameters to observe cost change of system. The experiment results showed that the joint recycling model obtained much lower total relative costs for all the systems, and when the degree of waste amount difference between factories gradually become larger, the effect of joint recycling become better.

Key words: multi-factories, waste, disposal, joint mechanism, heuristic search algorithms

摘要： 为了有效率地回收与处置多个制造厂在生产过程中产生的废物，针对多个制造厂与单一回收中心构成的系统，将系统中的决策者分成两种决策类型。在类型1中决策者为回收中心，提出2种回收处置模型：独立模型（模型1-1）、协同模型（模型1-2）；在类型2中决策者为各制造厂与回收中心，提出3种回收处置模型：分散模型（模型2-1）、整合模型（模型2-2）、协同模型（模型2-3）。以系统总相关成本最小化为目标构建公式，以找出最优回收处置模型，并提出一套启发式搜索算法用于协同模型中的求解。通过改变制造厂数量与相关成本参数，来观察系统成本的变化。实验结果分析表明，协同模型可使系统总成本最小，且当各制造厂之间废物量的差异程度逐渐变大时，其他回收模型较协同模型的成本差距也变大，因此协同的效果越好。

关键词: 多制造厂, 衍生废物, 回收处置, 协同机制, 启发式搜索算法