Scheduling Problems With Learning Effects And Due Date Assignment

Scheduling is a kind of important field of combinatorial optimization. Itsbackground is mainly machine manufacture, later it be widely used in areassuch as transportation dispatch, production management and computer systems.Scheduling considering learning efect and due date assignment simultaneously isprovided with very important practical significance and has received considerableattention in many literatures, which had made a lot of results. We consider twokinds of scheduling problems about learning efect and due date assignment inthis paper. The structure of the paper is organized as follows.In the first chapter, some notations, definitions and basic background infor-mation about the subject are introduced, and then the main research results andinnovation in this dissertation have been summarized.In the second chapter, we consider single-machine scheduling with learningefects and job scheduling under a group technology environment can be assignedone of the three kinds of due date FML-CON, FML-SLK and DIF. The goalis to determine an optimal combination of the due date assignment strategyand job schedule so as to minimize objective function that includes earliness,tardiness, due date assignment and flow time costs. An unified polynomial-timeoptimal algorithm is provided for all of the problems with above three due dateassignments. Furthermore, for the case that machine with special learning efect,we also present optimal algorithms.In the third chapter, we consider scheduling under the learning efect andjob due dates can be assigned as follow two due date assignment methods: CONand SLK. The problem is to determine the optimal due dates and the processingsequence simultaneously to minimize costs for due date assignment, earliness and penalties of tardy jobs. We present polynomial-time optimal algorithms to solvethe problems. Additionally, based on the problem that scheduling deterioratingjobs with CON/SLK due date assignment, learning efects be considered simul-taneously and we show that the new problems can also be solved optimally inpolynomial-time.