Study On Multi-resource Leveling Optimization Of Engineering Project Based On Genetic Algorithm

Resources are important factors which affect the schedule, quality and cost of the engineering project. Resource leveling in the engineering project is benefit a lot to reduce the management cost, improve the project quality and guarantees the project to be finished on time, which can provide protection to the control of three major objectives of the engineering project management. Thereby resource leveling has a very important significance to improve the comprehensive benefits of the project.Based on the research of the single-resource leveling optimization basic theories and its drawbacks, this paper evaluates importance of the resources and analyzes the correlations among the resources. A multi-resource leveling optimization simplified model is built and genetic algorithm is used to solve this model. The results of the paper are as the following:(1) The paper contrapose the multi-resource leveling optimization evaluates importance of the resources and analyzes the correlations among the resources. An index system is used to assess the importance of the resource. The importance index system consists of the cost importance index, the activity importance index and the total float importance index. The AHP is applied to identify the weight of each index. Then a comprehensive importance index is summarized for different resource. The gray correlation clustering analyses method is used to cluster various resources and simplified principle is designed, which of availability is validated by example results.(2) This paper designs genetic algorithm to solve multi-resource leveling optimization problem, including chromosome structure and coding design, the fitness function design, genetic operation design, restriction condition handling and algorithmic process design.(3) A multi-resource leveling optimization simplified model is built and genetic algorithm is used to solve this model. The results indicate that this method can simplify the multi-resource leveling model, at the same time considering multiple resources. Moreover, this method can obtain more than one optimal solution, which can provide more reference to guide engineering practice.(4) Matlab genetic algorithm toolbox function and program codes are written to solve the example model. The optimal solution and the corresponding schedule schemethe indicate that the performance of the computations is very good and can provide a new and feasible approach to multi-resource leveling optimization problem in engineering project.