建设物流方案多目标均衡优化模型研究  

Multiobjective tradeoff optimization model of implementation scheme in project logistics

作　　者：佟士祺[1] 王诺[2] 谢春晓[2] 

机构地区：[1]大连理工大学管理学院,辽宁大连116024 [2]大连海事大学交通与物流工程学院,辽宁大连116026

出　　处：《计算机集成制造系统》2008年第12期2312-2316,共5页Computer Integrated Manufacturing Systems

摘　　要：为解决建设物流项目实施方案中的工期、成本、安全等多目标综合优化问题,提出了以决策网络计划法为基础,以多属性效用函数为理论建立的"工期—成本—安全"均衡优化模型。采用自适应遗传算法求解,并利用自适应调节参数提高模型的求解效率。通过某大型化工设备运输与安装方案的优化实例,对模型的合理性和算法的有效性进行了验证。结果表明,采用该模型求解建设物流项目实施方案的多目标优化问题,可以解决工序变量函数关系不明确、不同优化目标相互之间难以兼顾、决策点数目较多的难题。To solve comprehensive multiobjective optimization of implementation scheme about time, cost and security in project logistics, a time-cost-security tradeoff optimization model, which was based on decision network planning technique and multi-attribute utility function theory, was proposed. Self-adaptive genetic algorithm was applied and self adaptive control parameters were provided to improve the resolution efficiency. Finally, a case study of transportation and installation implementation scheme of large chemical equipment was given to verify the validity of the model and feasibility of solution method. Results showed that the presented method could solve problems such as uncertain function relationships among variables in one process, tradeoff optimization of different objects and calcu lating difficulties caused by the excessive amount of decision points.

关 键 词：建设物流 多目标优化 决策网络计划 多属性效用函数 遗传算法 

分 类 号：F224.33[经济管理—国民经济]