02807nam a2200253   450000500170000000800410001703500150005810000240007324500900009726000520018730000140023949000270025350001440028050200580042452017250048265000280220765000240223570000390225970000410229870000390233971000510237881000590242985600650248820260818134557.0030798         th                  eng    a.b114087770 aSuthad Sinsuesatkul10aImprovement of forecasting and scheduling procedures in the Siam-Fibre Cement Company  aBangkok :bAsian Institute of Technology,c1994  a95 leaves1 aThesis ;vno. IE-94-02  aA thesis submitted in partial fulfillment of the requirement for the degree of  Master of Engineering, School of Engineering and Technology  aThesis (M.Eng.) - Asian Institute of Technology, 1994  aThis study concerns with the production planning problem of a multi-product  single-stage system of the Siam-Fibre Cement company. At first, the current  practices of the company related to the production planning, forecasting,  production planning, and scheduling are investigated. Then, practical and  appropriate procedures to improve the efficiency are developed.  Considering data availability and ease of implementation, two types of time  series forecasting models, i.e., Winter's model and ARIMA model, are suggested.  Data of company sale are divided into two data sets. The first data set is used for  model developing and the second data set is used for model validation. Model  evaluation is done by comparing the results of both models with current practice  in terms of forecasting error measures. Both models give better results than those  of current practice.  For production planning, the current practice model uses linear programming  to optimize total cost. All relevant costs are included in the model and the  assumption of linear costs is also reasonable. Thus, it is not necessary to develop a  new model. For the implementation period, the output of model indicates the  products to be produced on a specific machine. Now, the final task is how to  schedule these products on each machine to minimize setup time.  The problem of machine scheduling is analogous to combinatorial problem  known as the traveling salesman problem. This problem can be solved by mixed  integer programming. However, considering ease of implementation, a heuristic is  suggested for ease of solution, and implementation. Both the results of heuristic  and traveling salesman algorithm yield the same optimum results.10aScheduling (Management)10aProduction planning1 aNagarur, Nagendra N.,eChairperson1 aPastijn, H., eExamination Committee1 aKuehn, W., eExamination Committee2 aThe Siam Cement Co., Ltd., eScholarship Donor2 aAsian Institute of Technology.tThesis ;vno. IE-94-02  3Full-Textuhttp://203.159.5.9/ait-thesis/detail.php?q=B15378