臺灣博碩士論文加值系統

在半導體測試廠中，測試機台之成本佔半導體測試廠建構資本的70％以上，而提升測試機台的使用率，將可使機台成本攤提至更多的晶圓上，因此可降低晶圓的單位測試成本，進一步提升測試廠的競爭力。
本研究目的係分析電性測試人員以及測試機台在不同的搭配比例下，對於機台使用狀況的影響。當電性測試人員負責看管較多的測試機台時，因為負責的作業員正忙碌於其他測試機台的作業，而造成其他亦需要服務的測試機台處於閒置的頻率提升。此類機台閒置，乃是由於機台間共用的作業員在同一時間僅能服務單一測試機台而產生，稱為機台干涉時間。機台干涉時間隨著電性測試人員負責較多的測試機台而增加，為一種產能的損失。因此如何求得最佳之電性測試人員與測試機台的組合方式，以使系統達到最高之績效，即成為一個重要的議題。
本研究利用反應曲面法以及基因遺傳演算法搭配系統模擬方法，提出一個啟發式演算法以搜尋測試機台與測試人員的適當組合方式，作為人機指派之決策依據。利用模擬軟體建構出測試廠區的作業流程後，可求得在各種不同的機台與作業員組合比例下的系統績效，並利用反應曲面法找出相對較優的測試機台與電性測試人員的指派方式。此外，藉由人機指派問題之遺傳演算法，亦可搜尋出相對較優的測試機台語電性測試人員的指派方式，但須耗費較多之時間，可在反應曲面法失效時作為替代之演算法。本研究結果發現，本研究演算法求得之指派方式與具有經驗之工程師之指派方式具有相同之優良系統績效，可知本研究之演算法具有尋優之有效性，亦可作為新進工程師決策之模式依據。

Equipment costs account for more than 70% of the capital investment in a semiconductor test facility. To improve the utilization of the test machines shares the cost of test machines to more wafers, thus decreases the unit test cost of the wafer and improves the competitive strength of a semiconductor test facility.
The purpose of this thesis is to develop a algorithm to find a well-performed assignment of test machines and operators for different test product mix. Assigning multiple machines to an individual operator may cause the machine interference problem. That is, when the operator is tending a machine, another machine that needs to be tended by the operator at the same time has to stay idle and wait until the operator is available again. The idle time that caused by assigning multiple test machines to an individual operator is termed as machine interference time and is a kind of capacity loss. Machine interference time raises while decision maker assigning the test machines to operators poorly. Thus, it is an important issue to find an assignment of test machines and operators that leads to relatively better system performance. In this thesis, we apply response surface methodology and genetic algorithms together with simulation method to develop the heuristic algorithms for searching the well-performed assignment of test machine and operators. The assignments found by the heuristic algorithms that we proposed have the same system performance with the assignment decided by the experienced engineer, thus to show the validity of the heuristic algorithms and can be the decision support model for new coming engineer.

Contents
Chapter 1 Introduction 1
1.1 Background & Significance 1
1.2 Motivation 2
1.3 Research Aims 3
1.4 Overview of This Thesis 4
Chapter 2 Literature Review 5
2.1 Machine Interference Problem 5
2.2 Existing Methods 6
2.3 Conception of Simulation 10
2.4 Response Surface Method with Mixture Experiment 12
2.5 Fundamentals of Genetic Algorithms 16
Chapter 3 Research Framework 19
3.1 Problem Structuring 20
3.2 Simulation Model 26
3.3 Assignment Algorithm 33
Chapter 4 Empirical Study 46
4.1 System Description 46
4.2 Simulation Model 47
4.3 Search the Well-Performed Assignment 51
4.4 Results and Discussion 56
Chapter 5 Conclusion and Future Research 60
Reference 63
Appendix 66

1.Ashcroft, M. A. (1950), “The productivity of several machines under the care of one operator”, Journal of the Royal Statistical Society (B), Vol. 12, No. 1, pp. 145-151.
2.Bahnasawi, Ahmed A., Magdi S. Mahmoud and Shawki Z. Eid (1996), “Sensitivity analysis of machine interference in manufacturing systems”, Computers Ind. Engng, Vol. 30, No. 4, pp. 753-764.
3.Chen, Pangwei (2000), “A Queueing Network model for Machine Interference Problem with External Activities”, Institute of Manufacturing Engineering National Cheng Kung University.
4.Chien, C., J. Deng (2001), “Optimization of wafer exposure patterns using a two-dimensional cutting algorithm”, International Transaction in Operational Research, Vol. 8, No. 5, pp. 535-545.
5.Cornell, J. A. (1990), Experiment with Mixtures: Designs, Models, and the Analysis of Mixture Data, 2nd Edition, John Wiley & Sons, New York.
6.Gaafar, L. K. and Tariq A. Aldowaisan (1994), “Fitting a mixture-based response surface using computer simulation”, Proceedings of the 1994 Winter Simulation Conference, pp. 1424-1427.
7.Gen, M., and R. Cheng (1997), Genetic Algorithms and Engineering Design, John Wiley & Sons, New York.
8.Gupta,U.C., T.S.S. Srinivasa Rao (1996), “Computing steady state probabilities inλ(n)/G/1/K queue”, Performance Evaluation, Vol. 24, pp. 265-275.
9.Holland, J. H. (1975), Adaptation in Natural and Artificial Systems, Michigan University Press.
10.Jackman, John and Eric Johnson (1993), “The role of queueing network models in performance evaluation of manufacturing system”, Journal of the Operation Research Society, Vol. 44, No. 8, pp. 797-807.
11.Lachenmaier, Lynn S. and Jeffery K. Cochran (1996), “Use of resource blocks in SLAM II for the evaluation of operator-machine assignments”, Proceedings of the 1996 Winter Simulation Conference, pp. 1135-1140.
12.Law, Averill M., W. David Kelton (1991), Simulation Modeling & Analysis, 2nd Edition, New York.
13.Lee, R. S., M. J. Shaw (1997), “A genetic algorithm-based approach to flexible flow-line scheduling with variable lot sizes”, IEEE Transactions on System, Man, Cybernetics-Part B: Cybernetics, Vol. 27, No. 1, pp. 36-54.
14.Little, J. D. C. (1961), “A proof for the queueing formula L=λW”, Oper. Res. ,Vol. 9, pp. 383-387.
15.Myers, R. H. and D. C. Montgomery (1995), Response Surface Methodology: Process and Product Optimization Using Designed Experiments, John Wiley & Sons, New York.
16.Oh, Keytack H. (1996), “The computerized operator-machine system (OMS) for the least cost combination of operators and machines”, Computers Ind. Engng. ,Vol. 31, No. 1/2, pp. 159-162.
17.Raviv, Amnon (1998), “Optimal Staffing in Semiconductor Manufacturing: A Queuing Theory Approach”, Solid State Technology, Vol. 41, Issue 4, pp. 77.
18.Rossetti, Manuel D. and Gordon Clark (1998), “Evaluating a queueing approximation for the machine interference problem with two types of stoppages via simulation optimization”, Computers Ind. Engng., Vol. 34, No. 3, pp. 655-668.
19.Stafford, Edward F. (1988), “An optimal solution technique for the operator-machine assignment problem”, Production and Inventory Management Journal, Vol. 29, No. 9, pp. 1354-1359.