半導體產品種類多、變異性大，造成以代工為主的封裝業面臨少量多樣的快速反應生產模式，因此完工時間必須有效縮短，客戶滿意度也必須進一步提升，以符合市場要求。以半導體封裝廠瓶頸工作站銲線區為例，其加工時間長、機台數量多，機器經常會由於某些原因而必須停止，無法持續加工，有些因素是可預期的，如定期預防保養等；有些因素是不可預期的，如突發性的故障、機台作業或傳送不順等。若是預防保養排程決策不良，將導致每日產出不平均，在製品（Work-In-Process，WIP）水準起伏很大，系統資源浪費，系統流程阻塞，進而影響其他相關工作站。因此提供良好的預防保養排程讓每日銲線站之產出平均，WIP保持最低水準，增進供貨穩定性，以符合市場需求提昇公司競爭力。
本研究提出運用基因遺傳演算法在預防保養排程計畫上，期望在最有效率的情況下求得預防保養最佳化的排程。並以一IC封裝廠之瓶頸製程銲線站來模擬比較並以田口式（Taguchi）實驗設計法驗證此方法，以作為實際工廠訂定決策的參考依據。
在實證測試方面，整理了目前的資料，分析其中的要點與限制，並將實際作業上的習慣融入演算法中，本研究最終得到以下幾點結論：
(1) 研究結果證明基因演算法等啟發式方法可用來取代
系統化的人工編排方式，且能得到不錯的效果。
(2) 研究結果證明應用基因演算法所求得之總待機時間
最小，不只使工廠之產能提升，並可將工廠所損失
之總懲罰成本最小化，進而提升工廠之競爭優勢。

The variety in type and variability of semiconductor products have led the ODM-based assembly industry to face a quickly-reactive production mode with low quantity and high variety. Hence the completion time must effectively be reduced and customer satisfaction must further be improved to meet the market demand. The choke point of a semiconductor assembly house, the wire bonding station, is taken as an example. Its process time is long; a large number of machines is required, and discontinuous process due to frequent stops of machines constantly happens. Some of the factors causing the machine stops are expectable, such as regular maintenance, but some are contingent, like a sudden breakdown, unsmooth work or conveyance on machines, etc. When a bad decision of regular maintenance scheduling is made, it may cause uneven daily yield, fluctuant standard of Work-In-Proces (WIP), resource waste of system and block of system flow, and further influence other relative work stations. Therefore, a proper scheduling of maintenance which will make the daily yield of wire bonding station even, keep WIP in the lowest standard and improve the stability of product supply shall be provided to fulfill market demand and enhance the competitiveness of a company.
This research introduces the application of the genetic algorithm into regular maintenance scheduling plan in order to find the optimal scheduling in the most efficient condition. Moreover, the wire bonding station of an IC assembly house is taken for simulation and Taguchi experimental design method is then used to verify the applicability. This is to provide the house a reference for actual decision-making.
In the empirical tests, the current data are organized, key points and restrictions thereof are analyzed, and the usual practice during operation is incorporated into the algorithm. The study hence reaches the final conclusions as follows:
(1) The study results demonstrate that heuristic
methods like genetic algorithm can be used
to replace the systematic manual arrangement
and good effects can be obtained.
(2) The study results prove that total idling
time derived from genetic algorithm is the
smallest. It not only will increase the
production capacity but also minimize the
total penalty cost an assembly house will
lose, and therefore advance the domination
of the house to compete in the field.

中文部份
王清平，「退化製造系統下多種動作的動態維護策略之研究」，國立清華大學工業工程與工程管理學系，碩士論文（2003）。

朱玉芬，「應用基因演算法在專業IC 設計業的供應鏈生產排程之研究」，私立輔仁大學資訊管理研究所，碩士論文（2000）。

沈怡瑄，「以適應性批貨/設備配對策略與遺傳演算法於晶圓測試廠最佳化之派工與排程」，國立台灣大學資訊工程學研究所，碩士論文（2003）。

李嘉柱，李佳穎，「半導體後段廠之現場生產流程與作業管制條件分析辦法探討」，機械工業雜誌，12 月，109-115.（1999）。

林慈傑，「以遺傳演算法求解類運輸問題模式化的多廠訂單分配問題」，國立台灣大學工業工程研究所，碩士論文（2002）。

施嘉昶，「利用遺傳演算法及隱價求解生產規劃含裝設問題」，國立清華大學工業工程研究所，碩士論文（1998）。

陳怡文，「多種保養動作下動態預防保養策略之研究」，國立清華大學工業工程與工程管理學系，碩士論文（2001）。

張添香，「廣義預防維修置換策略」，國立台灣科技大學工業管理系，博士論文（2000）。

英文部分
Beasley, J.E. and Chu, P.C. (1996). A genetic algorithms for the set covering problem. European Journal of Operetional Research, 94, 392-404.

Bris, R., Châtelet, E., Yalaoui, F. (2003). New method to minimize the preventive maintenance cost of series–parallel systems. Reliability Engineering and System Safety, 82, 247-255.

Cavory, G., Dupas, R. and Goncalves, G. (2001). A genetic approach to the scheduling of preventive maintenance tasks on a single product manufacturing production line. International Journal of Production Economics, 74, 135-146.

Chambers, L. D. (2001). The Practical Handbook of Genetic Algorithms: Applications, CRC Press, Boca Raton.

Chang, P. C., Chen, S. H. and Lin K. L. (2005). Two-phase sub population genetic algorithm for parallel machine-scheduling problem. Expert Systems with Applications, 29, 705-712.

Charles, A. S., Floru, I. R., Catherine, A.P., Pibouleau, L., Domenech, S. (2003). Optimization of preventive maintenance strategies in a multipurpose batch plant: application to semiconductor manufacturing. Computers and Chemical Engineering, 27, 449-467.

Dohi, T., Matsushima, Kaio, N. and Osaki, S. (1996). Non-parametric repair-limit replacement policy with imperfect repair. European Journal of Operation Research, 96, 260-273.

Goldberg D. E. (1989). Genetic Algorithms in Search, Optimization, and Machine Learning, Reading, Mass.: Addison Wesley.

Gupta, M.C. Gupta, Y.P. and Kumar, A. (1993). Minimizing flow time variance in a single machine system using genetic algorithms. European Journal of Operational Research, 70, 289-303.

Grefenstette, J. J., (1986). Optimization of control parameters for genetic algorithms. IEEE Transactions on systems, 16, 122-128.

Holland, J. H. (1975). A daptation in natural and artificial systems. Ann Arbor: The University of Michigan Press.

Jhang, J. P. and Sheu, S. H. (2000). Optimal age and block replacement policies for a Multi-component system with failure interaction. International Journal of System Science, 31, 593-603.

Jiang, R. and Ji, P. (2002). Age replacement policy: A Multi-Attribute value model. Reliability Engineering and System Safety, 76, 311-318.

Kad, D. A. (1991). Replacement strategies with mixed corrective actions at failure. Computers and Operations Research, 18, 141-149.

Lee, C. Y., Piramuthu, S. and Tsai, Y. -K. (1997). Job shop scheduling with a genetic algorithm and machine learning. International Journal of Production Research, 35, 1171-1191.

Liu, J. and Tang, L. (1999). A modified genetic algorithm for single machine scheduling. Computers and Industrial Engineering, 37, 43-46.

Lotka, A. J. (1939). A contribution to the theory of self-renewing aggregates with special reference to industrial replacement. Annals of Mathematical Statistics, 10, 1-25.

Nakagawa, T. and Kowada, M. (1983). Analysis of a system with minimal repair and its application to replacement policy. European Journal of Operational Research, 12, 176-182.

Pintelon, L. M., and Gelders, L. F. (1992). Maintenance management decision making. European Journal of Operation Research, 58, 301-307.

Schaffer, J. D., Caruana, R. A., Eshelman, L. J. and Das, R.(1989). A study of control parameters affecting online performance of genetic algorithms for function optimization. Proceedings of the Third International Conference on Genetic Algorithms, 51-60.

Sortrakul, N. , Nachtmann, H. L. and Cassady, C.R. (2005). Genetic algorithms for integrated preventive maintenance planning and production scheduling for a single machine. Computers in Industry, 56, 161-168.

Tadahiko, M., Hisao, I. and Hideo T. (1996). Genetic Algorithms for Flowshop Scheduling Problems. Computers and Industrial Engineering, 30, 1061-1071.

Tasi, Y. T., Wang, K. S. and Teng, H. Y. (2001). Optimizing preventive maintenance for mechanical components using genetic algorithms. Reliability Engineering and System Safety, 74, 89-97.

Wang, H. and Pham, H. (1999). Some maintenance models and availability with imperfect maintenance in production systems. Annals of Operations Research, 91, 305-318.

Wang, H. (2002). A survey of maintenance policies of deteriorating systems. European Journal of Operational Research, 139, 469-489.