Resource Constrained Project Scheduling Optimizaion Based On Improved Genetic Algorithm

The resource-constrained project scheduling problem(RCPSP)is to schedule the activities of a project so that neither the resource constraints nor the precedence relationships are violated,with the objective of minimizing the completion time of the last activity,i.e.,the project makespan.For one thing,RCPSP is widely applied in diverse industries,such as construction engineering,software development and advanced manufacturing,which plays an important role in predicting project duration and arranging project activities.It is also of great significance for society to save resource and reduce waste.For another,RCPSP,as one of the crucial topics in the project scheduling research area,has attracted many scholars to conduct extensive researches on it.Although RCPSP itself is easy to describe,improving the solution methodology is of great difficulty.Moreover,there is a big difference between current project scheduling methods and the reality of scheduling situation.Hence,RCPSP is a significant and challenging problem both in theory and practice.RCPSP with specified availabilities for multiple renewable resources is considered in this paper.First,the research background is introduced,the project scheduling theory,schedule generation scheme and genetic algorithm are combed in detail as the theoretical basis of this paper.At the same time,the related concepts of RCPCP are introduced at length in the literature review,and the existing solution methods are compared.Second,a mixed integer programming(MIP)model is proposed,and a new schedule generation scheme is designed,which is named sequential placing approach.This approach locates the activities within divided zones formed by respective resource availability line one by one on ground of a placing sequence and a certain rule,in order to generate a feasible scheduling plan.On this basis,an improved genetic algorithm which adopts the sequential placing approach as the decoding rule is proposed to solve RCPSP.Finally,the computational experiments are conducted to examine the performance of the proposed algorithm.The theory and algorithm should ultimately serve the reality and provide feasible solutions for project management problems in modern enterprises.Case study on RCPSP is carried out to prove the practical application value of the proposed algorithm.