基于遗传算法的项目管理优化仿真研究

The optimization simulation model based on the project's three-dimensional target preferences

传统的项目进度一维优化扩展至有偏好的二维目标(进度、成本)优化,同时将成本优化目标分解为项目成本大小以及资源均衡度从而构成三维目标优化,将无资源约束的环境扩展至资源约束下的复杂环境,将局部搜索优化领域扩展至全局范围内的优化.在内容上,先对项目的单目标优化管理理论进行详尽研究并指出其现实的局限性,同时提出了智能启化式方法-遗传算法在资源约束下项目管理优化方面的优势.在此基础上本文构建了基于三维目标偏好的项目管理优化仿真模型,解决了项目管理优化理论中最为重要的两大问题:资源约束下的项目进度优化以及资源约束下的三维目标(项目进度、项目成本以及资源均衡度)的优化问题.为了验证此模型对以上问题的有效性,本文应用Matlab仿真技术进行仿真模拟并与传统方法做比较,从结果可以看出遗传算法能够更好的解决此类问题.

The paper extends the traditional one-dimensional project schedule optimization to a preference of 2D target （schedule, cost） optimization, and then to a three-dimensional target optimization. First this article shows the theory of dynamic optimizing project schedule in detail and points out the limitations of existing methods. At the same time, it expounds the intelligent and mineralization type method, the basic principle and application of genetic algorithm, and how to use Matlab genetic algorithm of intelligent optimization function. This paper constructs the multidimensional model of optimizing management of project schedule. At last, the model solves the project schedule management problem respectively： no resource constraints project schedule optimization, the resource constraints project schedule optimization and multi-objective optimization （resource, cost, efficiency）. Meanwhile, in order to verify the validity of the genetic algorithm to solve above problems, this paper applies universality case for validation. The results show that the genetic algorithm can solve such problem better compared with other traditional methods.