New Models And Algorithms For Machine Scheduling With The Group Technology

In the present, the production and processing model of multi-product is becoming their first option with the fierce environment of the global market and the unsurely needs of the customer. However, in this kind of production model, manufacturing enterprises are often facing with a lot of problems, such as, multi-products, low production and low repetition rate, unregularly replacement of processing technology, etc. Group technology overcomes the difficulties in processing and manufacturing, which is a special kind of production organization and management technology for the production of multi-products. In our country, the application of group technology is widespread, especially in the cigarette making, liquid metal forging, furniture manufacturing and other industries.It is based on the wide application of group technology in the manufacturing field. In this thesis we mainly study two aspects of the research. First, the due window in the past scheduling problem is the only one. However, due to the large of customer’s processing orders, and the single due window is often not flexible to meet the requirements of customer and delivery efficiency. So we need to develop multiple due windows accordingly. This thesis creatively combines multiple due windows with group technology. Our goal is to find the optimal group and job sequence, the optimal due window to minimize total cost, the total cost is including of the advance/delay cost, the start time cost of each due window, and the cost of the size of each due window.The job’s processing time is stable in the first problem we studied. However, in the past few decades, many scholars have made a lot of research on the learning effect of machine processing in the manufacturing field. However, as we know so far, few people have studied the effects of learning and deterioration in the same model. In addition, in the past related research, many scholars study the deterioration effect model of the machine with the increase of the number of times. The processing time of the production process is more and more tend to infinity. But this is obviously not in line with the actual processing environment. Therefore, the research work of this thesis is to propose two position related general processing model:general log-linear model; general exponential model, and combine it with the group technology. This two general process models include the effect of the learning effect and the effect of the study. In addition, in the general position-based deterioration effect model, the actual processing time of the process of production is more consistent with the actual processing environment. On this basis, we mainly study the single machine environment with two classical scheduling objectives, namely minimizing the makespan and the total completion time. In addition, for the two main work of this paper, we propose the corresponding optimal polynomial algorithms. We also give the examples to guide to practice.