考虑资源技能和时间窗特性的任务可拆分项目调度
发布时间：2018-12-18 19:16
【摘要】：资源受限项目调度问题是项目管理中的典型问题，问题中一般假设任务不允许被拆分，而在实际项目中，资源工作时间的不连续、紧急任务的加入、市场因素等通常需要任务被拆分成多次执行。同时企业发展要求资源拥有多种技能。将这些因素引入问题中具有十分重要的理论和现实意义。 本文首先分析了研究多技能资源任务可拆分项目调度的现实和理论意义，，并对RCPSP、PRCPSP、多技能资源的RCPSP问题的研究现状进行总结。对问题进行清晰的界定和描述，构建了数学模型，提出了求解此问题的分支定界算法。此算法求解过程中以可行调度活动集进行分支，而利用节点的任务完成率来优化分支和优选过程；而从资源约束条件来看，资源拥有单技能时，通过对技能总量的比较而判断任务组合的资源约束条件是否满足，而资源拥有多技能时，通过对资源—技能矩阵的处理、对资源指派方案的编码和解码等过程计算资源对技能的提供量是否能满足任务组合对技能的需求，从而判断任务组合是否满足资源约束条件。最后，通过Patterson实例集的测试，比较混合遗传算法、粒子群算法与分支定界算法的目标函数和运行时间，可以看出中小型项目调度时，分支定界算法对解决多技能资源任务可拆分项目调度问题的有效性和可行性。 本文的研究成果，理论上丰富了资源受限项目调度问题的研究内容，而在动态环境下，给定项目和资源配置参数，可以得到较好的调度方案，为项目经理合理安排项目进度提供指导。
[Abstract]:Resource constrained project scheduling problem is a typical problem in project management. It is generally assumed that tasks are not allowed to be split, while in actual projects, the discontinuity of resource working hours and the addition of urgent tasks are involved. Market factors, etc., usually require tasks to be split into multiple execution. At the same time, enterprise development requires resources to have a variety of skills. It is of great theoretical and practical significance to introduce these factors into the problem. This paper first analyzes the practical and theoretical significance of the research on multi-skill resource task divisible project scheduling, and summarizes the research status of RCPSP,PRCPSP, multi-skill resource RCPSP problem. The problem is clearly defined and described, a mathematical model is constructed, and a branch and bound algorithm is proposed to solve the problem. In this algorithm, the feasible scheduling activity set is used to branch, and the task completion rate of the node is used to optimize the branch and optimal selection process. From the point of view of resource constraints, when resources have a single skill, the resource constraints of task combination are judged by comparing the total skill amount, and when resources have multiple skills, they are processed by resource-skill matrix. The coding and decoding of resource assignment scheme can calculate whether the resource supply to skills can meet the requirement of task combination and determine whether task combination meets resource constraints. Finally, through the test of Patterson instance set, comparing the objective function and running time of hybrid genetic algorithm, particle swarm optimization and branch and bound algorithm, we can see the small and medium-sized project scheduling. The effectiveness and feasibility of branch and bound algorithm for solving the problem of split project scheduling of multi-skill resource tasks. The research results of this paper theoretically enrich the research content of resource-constrained project scheduling problem, but in dynamic environment, given project and resource allocation parameters, a better scheduling scheme can be obtained. Provide guidance for project manager to arrange project schedule.
【学位授予单位】：华中科技大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F224;F062.4
本文编号：2386368
[Abstract]:Resource constrained project scheduling problem is a typical problem in project management. It is generally assumed that tasks are not allowed to be split, while in actual projects, the discontinuity of resource working hours and the addition of urgent tasks are involved. Market factors, etc., usually require tasks to be split into multiple execution. At the same time, enterprise development requires resources to have a variety of skills. It is of great theoretical and practical significance to introduce these factors into the problem. This paper first analyzes the practical and theoretical significance of the research on multi-skill resource task divisible project scheduling, and summarizes the research status of RCPSP,PRCPSP, multi-skill resource RCPSP problem. The problem is clearly defined and described, a mathematical model is constructed, and a branch and bound algorithm is proposed to solve the problem. In this algorithm, the feasible scheduling activity set is used to branch, and the task completion rate of the node is used to optimize the branch and optimal selection process. From the point of view of resource constraints, when resources have a single skill, the resource constraints of task combination are judged by comparing the total skill amount, and when resources have multiple skills, they are processed by resource-skill matrix. The coding and decoding of resource assignment scheme can calculate whether the resource supply to skills can meet the requirement of task combination and determine whether task combination meets resource constraints. Finally, through the test of Patterson instance set, comparing the objective function and running time of hybrid genetic algorithm, particle swarm optimization and branch and bound algorithm, we can see the small and medium-sized project scheduling. The effectiveness and feasibility of branch and bound algorithm for solving the problem of split project scheduling of multi-skill resource tasks. The research results of this paper theoretically enrich the research content of resource-constrained project scheduling problem, but in dynamic environment, given project and resource allocation parameters, a better scheduling scheme can be obtained. Provide guidance for project manager to arrange project schedule.
【学位授予单位】：华中科技大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F224;F062.4
【参考文献】
相关期刊论文 前1条
1 雒兴刚;汪定伟;唐加福;;任务可拆分项目调度问题[J];东北大学学报(自然科学版);2006年09期
本文编号：2386368
本文链接：https://www.wllwen.com/jingjilunwen/jingjililun/2386368.html
