Study On The Multi-objective Optimization Model Of Project And Its Algorithm

With more and more attention paid to the efficiency of project implementation,as well as the development of multi-objective optimization theory, multi-objectiveoptimization problem of project management becomes an active research topic. Sincepeople have put much more attention on protecting the environment and the personalsafety, the traditional studies which focus on the targets of time, cost and qualityunable to meet people’s needs. In modern engineering project management, time, cost,quality, environment and security constitute the main contents of the projectmanagement. How to achieve a balanced optimization of duration, cost, quality,environment and safety in the engineering project, has become an important part inthe research of project management.In this paper, we established four interactional models by detailed analysis of theinteraction between the various objectives: cost-duration, quality-duration,environment-duration and the security-duration. Then the multi-objectiveoptimization model can be formed by combining four single-target models and devoteto solve the new model.Firstly，the non-dominated sorting genetic algorithm wasanalyzed and introduced into the particle swarm optimization, which leads to animproved particle swarm algorithm-based on non-dominated sorting particle swarmoptimization;secondly,self-adaptive inertia weight and velocity mutation methods areproposed to improve searching efficiency and overcome the problem that particle fallinto the local convergence.Furthermore, by the two improved algorithm we obtainedresults of the new model, which demonstrates the reasonability,efficiency, andpracticality of the new model.This article has some guiding signification in theapplication of the management and control the following five aims: time, cost, quality,environment and security in engineering project.