Research On Multi-objective Optimization Model And Algorithm Of Fuzzy Resource-strained Project Scheduling

With the development of science and technology,as well as the progress of the society, the scale and complexity of project are constantly increasing, meanwhile increasingly highlight the importance of project management. Under the condition of the existing resource constraints, how to take into account the multiple target projects, realize the reasonable allocation and utilization of resources, so as to realize more effective project scheduling, has become a universal concern of all the project managers. At the same time, the resource-constrained project scheduling problem(RCPSP) is very important problems of project management category, so how to solve the problem of RCPSP has very high research significance in project management. Traditional project scheduling problem assumes that each task completion time and resources are needed for certain. But in the actual project operation process, due to the project by the uncertainty of the environment and resource constraints, led to a number of discrepancy between the project completion time and the estimated time beforehand. Hence we will type the fuzzy theory into the resource-constrained project scheduling problem, thus make up a new type of project scheduling-- the fuzzy resource-constrained project scheduling problem.We adopt the method of triangular fuzzy number to represent respectively fuzzy completion time and every job of resource demand in this article, establish single objective fuzzy scheduling optimization model for project under the condition of uncertain resource constraints. And take the minimization the completion fuzzy time for the project as a target function. We use the genetic algorithm to solve first, And aiming at the shortcomings of the single genetic algorithm, to improve the original genetic algorithm and the genetic operation. We change the original binary encoding to gray coding way, amend the original way of the roulette wheel selection method for chaos traversal search, and improved the fitness function. In the Improvements of the algorithm, on the basis of the original genetic algorithm, the join operation of simulated annealing, the original genetic algorithm into the hybrid genetic simulated annealing algorithm. At the last, we use an example to solve single objective fuzzy scheduling optimization problems to make a comparison between the two kinds of methods.Then on the basis of solving the project single objective fuzzy scheduling optimization problem in this paper, we increase project fuzzy resources cost of the objective function, change the original single objective fuzzy project scheduling problem into a multi-objective fuzzy project scheduling problem, and establish project multi-objective fuzzy scheduling optimization model.In solving multi-objective fuzzy scheduling problem, we adopt a double linked list based on models and activities, using a fast not dominant order based on the genetic algorithm(NSGA II),in order to ensure better achieve the reasonable use of resources. Finally through a concrete exampleobtaining good results, the design is verified by the effectiveness and robustness of the algorithm.