基于PSO的建筑工程资源均衡优化研究  

Research on Resource Balance Optimization of Construction Engineering Based on PSO

作　　者：唐剑 曹晶莹 王鹏杰 Tang Jian;Cao Jingying;Wang Pengjie(School of Architectural Engineering,Jiangxi Science&Technology Normal University,Nanchang 330013,Jiangxi,P.R.China)

机构地区：[1]江西科技师范大学土木工程学院,江西南昌330013

出　　处：《江西科技师范大学学报》2023年第6期114-118,共5页Journal of Jiangxi Science & Technology Normal University

基　　金：教育部产学合作协同育人项目(202002256012);江西科技师范大学研究生教学改革项目(KSDYJG-2020-05)。

摘　　要：资源管理是建筑工程项目管理的三大目标之一,资源均衡是一种有效的资源管理方法。针对工期固定,资源均衡问题,先利用层次分析法对各资源赋予权重,以最小方差为目标函数作为评价指标,运用微粒群算法进行优化,最后通过实际案例验算,完善工程网络计划优化模型。为解决微粒群算法难以兼顾收敛性和速度的问题,利用个体自我认知、信息共享,充分发挥上一代微粒的速度与位置优势,引入惯性学习因子,对微粒群算法进行优化,提升算法的局部寻优能力和全局寻优能力。Resource management is one of the three objectives of construction project management,and resource balancing is an effective resource management method.For the fixed duration and resource balancing problem,we first use hierarchical analysis to assign weights to each resource,use the minimum variance as the objective function as the evaluation index,use the particle swarm algorithm to optimize,and finally improve the engineering network plan optimization model through actual case verification.In order to solve the problem that the particle swarm algorithm is difficult to balance convergence and speed,individual self-knowledge and information sharing are used to give full play to the speed and location advantages of the previous generation of particles,and inertial learning factors are introduced to optimize the particle swarm algorithm and improve the local and global optimization seeking ability of the algorithm.

关 键 词：多资源均衡 微粒群算法 方差 惯性学习因子 

分 类 号：C931[经济管理—管理学]