多项目环境下time-cost置换问题建模与求解  

Model and resolution approach for problem of multi-project time-cost trade-off

作　　者：王勇胜[1,2] 梁昌勇[1] 鞠彦忠[2] 

机构地区：[1]合肥工业大学管理学院,合肥230009 [2]东北电力大学建筑工程学院,吉林132012

出　　处：《计算机工程与应用》2010年第20期237-240,共4页Computer Engineering and Applications

基　　金：国家自然科学基金重点项目No.70631003~~

摘　　要：针对多项目环境下的time-cost置换问题,以活动资源为决策变量,建立了以各项目总延迟最小与总成本最小的双目标time-cost置换模型。在总结前人对于多目标优化求解方法的基础上,针对多项目的特点,提出了基于NSGA-II非劣排序的局部直接搜索改进遗传算法,并考虑资源为连续与离散两种情况进行求解。与NSGA-II相比,算法可以较快地收敛到最优解,并且具有较好的离散变量搜索能力,与ZDT系列测试函数的比较也体现了算法的优越性。For the time-cost trade-off problem with multi-project condition,this paper develops a bi-objective model that the decision variables are the allocated resource quantities,and the objective functions are the projects delay（to be minimized） and the project costs（min）.After summarizing other researchers’multi-objective evolution algorithms and considering the characteristics of multi-project,this paper puts forward a local direct search technology based on Nondominated Sorting Genetic Algorithm II（NSGA-II）.It adopts discrete-continuous decision variables to settle the problem.Compared with NSGA-II,the approved algorithm can find the optimal solutions rapidly,moreover,for discrete variables,the algorithm can also run better.Finally,it compares the improved algorithm to ZDT test problems,the algorithm is superior similarly.

关 键 词：多项目 双目标 time-cost置换 

分 类 号：F2[经济管理—国民经济] TP391[自动化与计算机技术—计算机应用技术]