活动重叠模式与资源约束下的项目调度优化

发布时间：2018-03-07 01:04

本文选题：项目调度　切入点：遗传算法　出处：《管理科学学报》2017年09期 　论文类型：期刊论文

【摘要】：采取活动重叠模式通常是加速研发的有效手段,带有活动重叠的资源受限项目调度问题是经典资源受限项目调度问题的扩展.首先,深入分析了活动重叠对于项目调度的影响,对活动重叠及其不确定进行详细描述与建模,提出了活动重叠导致下游活动返工时间的二项分布概率模型;其次,构建了以最小化研发项目期望工期为目标的优化调度模型,设计了基于串行进度生成机制的遗传算法对大规模问题进行优化求解;最后,基于PSPLIB J60问题库中480个算例分析了该算法的计算结果,并考察了网络参数、资源参数和重叠参数变化时,采用活动重叠模式对缩短项目工期的影响.研究结果表明:活动对资源的需求强度越小或资源稀缺程度越低,可重叠活动对数量就会增加,项目工期缩短得越明显;网络复杂度的变化对缩短项目工期的影响不大;项目中重叠活动对越多,重叠导致的下游活动返工的概率越小,项目工期缩短的越明显.
[Abstract]:Activity overlap mode is usually an effective means to accelerate R & D. The resource-constrained project scheduling problem with overlapping activities is an extension of the classical resource-constrained project scheduling problem. Firstly, the influence of activity overlap on project scheduling is deeply analyzed. The activity overlap and its uncertainty are described and modeled in detail, and the binomial distribution probability model of activity overlap leading to the rework time of downstream activities is proposed. Secondly, the optimal scheduling model with the goal of minimizing the expected duration of the R & D project is constructed. The genetic algorithm based on serial schedule generation mechanism is designed to solve large-scale problems. Finally, 480 examples in the PSPLIB J60 problem library are used to analyze the results of the algorithm, and the network parameters are investigated. When the resource parameter and overlapping parameter change, the effect of the activity overlap mode on shortening the project duration. The research results show that the smaller the activity demand for resources or the lower the resource scarcity, the more overlapping activities will increase. The shorter the project duration is, the less the network complexity changes to shorten the project duration; the more overlapping activities in the project, the smaller the probability of rework the downstream activities caused by overlap, the shorter the project duration is.
【作者单位】：天津理工大学管理学院;北京航空航天大学经济管理学院;
【基金】：国家自然科学基金资助项目(71571005;71271019) 教育部人文社会科学基金资助项目(16YJC630159)
【分类号】：F273;TP18

【相似文献】