一种项目群级联优化调度方法

A Program Cascade Optimization Scheduling Method

多项目管理不仅需要对执行层的多项目进行优化调度,还需要在策略层通过项目群的形式对项目之间的关系进行协调。本文在资源约束多项目调度问题的基础上,面向项目群策略管理的实际需求,提出一种新的项目群级联优化调度方法。对项目群计划调度的过程进行分析,给出项目群级联调度的业务过程模型。以目标级联分析方法(ATC)为框架,以多项目工期最短为目标,考虑资源约束、项目内部的紧前关系约束和多项目之间的紧前关系约束,构建项目群级联调度的问题模型。结合ATC方法提出项目群级联优化调度问题的求解算法。该算法采用自上而下、反复迭代的方式运行。处于策略层的项目群向处于执行层的项目分配资源并下达目标,然后对每个单项目在执行层先分别优化,同时把反馈结果传递到策略层对各执行层优化结果进行协调。若不满足项目群总体目标,则对项目群再进行协调,重新分配资源并下达目标且反复迭代,最终达到整体最优。将本文所提出的方法与遍历算法进行充分的比较,数值实验结果表明,本文所提出的方法具有较高的计算效率,能够满足项目群级联优化调度要求,是一种有效的项目群调度方法。

Multiple project scheduling at the operation level is an important part of multi-project management.Similarly,the program planning at the tactical level is also essential since it coordinates the relationships of projects.Based on the resource constrained multi-project scheduling problem,a new cascade scheduling method is proposed for programs to enhance linkages between the multi-project scheduling at operation level and program planning at the tactical level.The process of program planning and scheduling is analyzed,and a busi⁃ness process model is given for program cascade scheduling.The analytical target cascading(ATC)as a framework to construct the problem model of the program cascade scheduling problem,which minimizes the program duration under resource constraints and precedence relationship constraints within projects and between multiple projects.The algorithm for solving the program cascade optimization scheduling problem is proposed in combination with the ATC method.The algorithm runs from top to bottom and iteratively.The program at the tactical level allocates objectives and resources to each project at the operation level,and then each single project is scheduled independently.At the same time the feedback results are passed to the tactical level,where the program evaluates the scheduling results of the operation level.If the overall objectives of the program are not achieved,the program will coordinate the objectives and resources for all the projects again,and then real⁃locate goals and resources to each project.The process is repeated until the overall objective is achieved.A practical experiment is conducted comparing the proposed method with the traversal algorithm,and the results show that the method proposed in this paper has good computational efficiency and can meet the practical requirements of the program scheduling.