Resource Constrained Multi-project Scheduling Optimization Based On Transfer Time-Transfer Cost

Resource constrained project scheduling problem(RCPSP)minimizes the duration of the project(make span)by scheduling a series of activities under the constraints of limited resource availability and task priority relationships.Traditional resource constrained project scheduling optimization problem more rarely consider the transfer time and transfer cost of resources within the project and between of the project,and project execution requires specific resources(people,equipment,material)appear at the appointed place,so the resources in one place after the task of the project is transferred to the next place,usually requires additional time and cost.Because of the uncertainty in the project scheduling process,it is difficult to ensure the project can be implemented smoothly according to the established scheduling scheme.In this case,it is particularly important to take into account the uncertain factors of resource transfer time and transfer cost in the process of multi-project scheduling.The purpose of this paper is to propose a resource constrained multi-project scheduling problem considering resource transfer time and transfer cost,and to build the corresponding model and generate the corresponding optimal scheduling.First,the thesis reviewed that researchers constructed more different model and the objective function respectively according to different resource-constrained project scheduling problem.However,the existing research either consider the transfer time,or only considering the transfer cost,did not consider the resources transfer time and transfer costs at the same time,and objective function of total duration don’t include transfer time.Then,a multi-project scheduling model is established to minimize project duration(including transfer time)and transfer cost.Furthermore,an improved genetic algorithm is proposed to solve the resource constrained multiproject scheduling problem considering the transfer time and transfer cost.The improved genetic algorithm adopts the natural number coding method based on priority,and introduces the multiobjective sequencing method of non-dominant Pareto solution for optimization,which ensures the priority relationship constraint of tasks and reduces the complexity of understanding.On this basis,the objective function is selected by calculating the crowding distance of chromosomes.Finally,the design of the genetic algorithm is applied to the practical example,the initial population generation and iteration to 50 and 100 of the objective function values,the solution of pareto optimal solutions,and it is concluded that one group of the optimal solution of scheduling gantt chart,the level of resource utilization of two types of scheduling under comparison,concluded that the average utilization of the three kinds of resources,and will eventually solve the result with the traditional heuristic algorithm(the shortest project tasks(SASP),minimum completion time(MINLFT)at the latest)comparative analysis,to verify the effectiveness of the algorithm;Combined with the comparison of the use of three renewable resources and the resource transfer cost,it reflects the concept of resource utilization maximization in resource-constrained multi-project scheduling problem,and illustrates that the multi-project scheduling problem with transfer time-transfer cost is more efficient in the use of resources.