Research On Scheduling Problem Considering Machine Availability And Maintenance Resource Constraints

Faced with challenges from the economic globalization,the manufacturing industry suffers more difficulties than opportunities.Therefore,the manufacturing enterprises must exert all of their energies in optimizing the operation and the management of production systems and scheduling the limited resources reasonably to maximize their production efficiency.However,most of the existing researches on production scheduling are divorced from the real production scenarios.They ignore the impact of machine availability on production scheduling and the impact of resource constraints in the scheduling process.This paper concentrates on the scheduling problems with machine availability and maintenance resource constraints,and introduces the relationship between availability and reliability.In addition,an efficient heuristic algorithm is designed to solve the problem.Considering the generality of single machine scheduling problem,the research of this paper is firstly carried out in the single machine environment.Also,the single machine scheduling model with machine availability constraints is established,and the objective is to minimum the total tardiness.The machine availability is defined by the reliability which can be restored through preventive maintenance(PM).A variety of PM with different improvement factors are proposed,and they are flexibly selected according to the machine status.A genetic algorithm based on the Emmons Rules is designed to solve the model,and the results turn out that the proposed model can cope with the impact of machine availability on production scheduling effectively.Through sensitivity analysis,the effects of improvement factor,reliability threshold,and due date tightness on task sequencing are studied.Then,the single machine problem is extended to a more complex environment: parallel machine problem with maintenance resource constraints.The introduction of maintenance resource constraints will increase the complexity of the model,thus we adopt the idea of packing to establish the model.At this point,the variables to be determined are not merely the machine selection,and the task sequencing,but they also consists of the resource allocation.Therefore,the algorithm needs to be redesigned according to the characteristics above.The experimental results show that the algorithm designed in this paper can perform well both in solving speed and precision.Meanwhile,the maintenance strategy can shorten the total tardiness significantly.Through sensitivity analysis,it can be seen that the influence of improvement factor,reliability threshold and the maintenance resource quantity on the total tardiness and machine availability.Moreover,it can also provide theoretical support for the real parallel machine workshop.