不确定环境下医疗废物回收网络鲁棒优化研究  

Robust optimization of medical waste recycling network in uncertain environment

作　　者：李昌兵[1] 梁琴 伍凌 LI Changbing;LIANG Qin;WU Ling(School of Economics and Management,Chongqing University of Posts and Telecommunications,Chongqing 400065,China)

机构地区：[1]重庆邮电大学经济管理学院,重庆400065

出　　处：《河北科技大学学报》2024年第5期549-561,共13页Journal of Hebei University of Science and Technology

基　　金：国家自然科学基金(62272077);重庆市自然科学基金(cstc2021jcyj-msxm2047);重庆市教委科学技术项目(KJQN202100604)。

摘　　要：针对医疗废物回收网络的不确定性,选取医疗废物量和运输成本作为关键变量,构建有多个不确定参数的多目标非线性整数规划模型,并引入鲁棒优化来处理不确定因素。结合多目标粒子群优化算法(MOPSO)和遗传算法(GA)求解该模型,外层GA负责选址决策,内层MOPSO针对选址结果进行配送路径优化,并以国内某城市为实证对象进行仿真验证。结果表明:相较于传统遗传算法,所提算法总成本降低了10.37%,总风险减少了1.86%,工作量偏差缩减了50.18%;敏感性分析表明,医疗废物量的不确定性对目标函数的影响更为显著。所提模型使得决策者可根据风险偏好调整不确定参数,以获得最佳的医疗废物回收网络优化方案,为进一步的医疗废物回收研究提供了参考。Aiming at the uncertainty of medical waste recycling network,with the quantity and transportation cost of medical waste as the key variables,a multi-objective nonlinear integer programming model with multiple uncertain parameters was constructed,and robust optimization was introduced to deal with the uncertain factors.Multi-objective particle swarm optimization(MOPSO)and genetic algorithm(GA)were combined to solve the model.The outer GA was responsible for location decision,and the inner MOPSO was responsible for distribution path optimization based on location selection results.A domestic city was selected as the empirical object for the simulation.The results show that compared with the traditional genetic algorithm,the proposed algorithm reduces the total cost by 10.37%,the total risk by 1.86%and the workload deviation by 50.18%;Sensitivity analysis proves that the uncertainty of medical waste volume has more significant influence on the objective function.The proposed mode can help the decision makers adjust the uncertain parameters according to the risk appetite to obtain the best medical waste recycling network optimization scheme,which provides some reference for further study of medical waste recycling.

关 键 词：城市运输经济学 医疗废物 回收网络 鲁棒优化 多目标粒子群优化算法 

分 类 号：U121[交通运输工程]