New Model And Optimization Algorithms For Multitasking Scheduling Based On Job And Location

Over the past decade,as competitions have increased and multitasking has become more frequent in everyday life,it has become a natural response in all kinds of work and living environments.Unfortunately,this way of accomplishing tasks forces us to constantly restart and refocus.Attention,at this time,the research of multitasking scheduling is particularly important.On the other hand,with the rapid development of economy and the advent of the Internet era,the personalized demand of customers is becoming more and more prevalent,becoming the dominant factor in the service industry.The real-time update and change of orders make the traditional scheduling mode of manufacturing industry no longer applicable.At this time,the theoretical research and practice of multitasking scheduling are popular today.It plays an important role in the production and service of enterprises.Based on the practical problems existing in manufacturing and service industries,it is of great significance to establish a reasonable scheduling model with multitasking and design an optimal algorithm for it from both theoretical and practical perspectives.Firstly,the research background and significance of multitasking scheduling problem are described,and the related topics in this paper are reviewed.Secondly,this paper studies the multitasking scheduling problems with SLK due-window and learning/deterioration effect.Firstly,we propose a new unified scheduling model which relies on the job itself and processing position.This model sets the interruption factor as a variable coefficient for the first time,and overcomes the drawback that each job must interruption others.Then,two aspects are studied based on the new model.Enterprises need to assign a reasonable order delivery date to achieve a balance among the three key success factors:low cost,fast response,and high customer satisfaction.For this reason,the appropriate due-window assignment has always been an important issue in scheduling research.However,in previous studies,parallel machines usually use the same due-window.Now,due to the demand of personalized production,small batch production has become the mainstream.A due-window cannot flexibly meet the requirements.We assume that we assume that each job has its own due window and shares a common due-window size in the same machine to increase fairness.The size of the due-window is determined by the machine itself.Therefore,our first task is to study parallel-machine scheduling with machine-dependent slack(SLK)due-window assignment in the multitasking environment,and give a polynomial time optimization algorithm to minimize the total cost.The second task is to consider the learning and deterioration effects,which are the internal factors of the machining process.We mainly study the situation that the job processing time depends on the position change.We integrate the learning effect and the aging effect into the same model,and the model overcomes the disadvantage of unbounded time.No research has been done in the multitasking scheduling field.In this part,the minimization of the total maximum completion time and the minimization of the completion time objectives are analyzed,and polynomial time optimization algorithms are given.Finally,we analyze all the algorithms designed in this paper,and use LINGO tools to implement the algorithm.For the design ideas and programs are similar,only one model is selected for example analysis.The validity of algorithms is verified from the examples.