整合分群技術與演化式模糊系統於訂單交期預測之研究

A Hybrid Model by Clustering and Evolving Fuzzy Rules for Due Date Forecasting

陳秀榛(Shiou-Jen Chen)

指導教授 ： 張百棧

元智大學/工程學院/工業工程與管理學系/碩士(2005年)

摘要

由於科技的發達，消費者對於產品資訊已逐漸透明化，使得業者對於顧客的滿意度也變的特別的重視。對製造業而言，產品的品質和交期逐漸受到顧客的重視。交期如果不準確，將會降低對顧客的信用，生產者本身則會產生很大的損失，嚴重者可能會有停機待料的問題，造成存貨成本以及其它機會成本的增加。因此，交期指定問題也漸漸成為業者重視的一大課題。近年來，全球半導體的產能快速的以亞太地區為主要來源，估計未來會有更大的市場商機出現在半導體產業中，讓產業陷入了激烈的市場競爭。因此，如何預測正確的訂單交期，將成為業者能夠有效掌握市場佔有率的工具之一。有鑑於此，本研究的目的是在建立一個有別於傳統預測分析模式的新預測方法，讓業者能夠有效的在訂單到達時，預測出產品所需的工作流動時間，使得決策者能進一步推估出交期的確切時間點，完成準時交貨，提高公司在產業界的競爭力。研究中，採用電腦模擬晶圓廠的生產過程所產生的數據，進行對BPN、WM、GA-WM以及本研究所提出的KGA-WM四種不同預測模式的分析與比較。最後，實驗結果顯示，是以本研究所提出的KGA-WM預測模式在集群個數c=3、c=4以及c=5下的預測能力為最佳。

關鍵字

集群分析 ； 模糊規則 ； 基因演算法 ； 預測

並列摘要

This research develops a hybrid model by integrating K-mean Cluster, Genetic Algorithms (GA) and Fuzzy Rule Base (FRB) to forecast the Due Date of Order. The research focuses on Wafer fabrication industry. This hybrid model encompasses two novel concepts: 1. Classify FRB into different clusters, thus the prediction accuracy of FRB is further improved. 2. Evolve FRB by optimizing the number of fuzzy terms of the input and output variables, thus the prediction accuracy of FRB is further improved. Fristly, use the simulated data into the K-mean for classification. Then, corresponding FRB is selected and applied for Job Flow Time forecasting. Genetic process is further applied to fine-tune the composition of the rule base . Finally, using the simulated data, the effectiveness of the proposed method is shown by comparing with other approaches. This research can provide some suggestions for solving the problems existing in the current industry.

並列關鍵字

K-mean ； Fuzzy Rule Base ； Genetic Algorithm ； Forecasting

參考文獻

2. Cheng TCE, MC Gupta, “Survey of scheduling gresearch involving due date determination decision,” European Journal of OperationalResearch, 38, pp.156– 166, 1989.

3. Chung, S. H., M. H. Yang, C. M. Cheng, “The design of due date assignment model and the determination of flow time control parameters for the wafer fabrication factories, ”IEEE Transactions on Components, Packaging, and Manufacturing Technology-Part C, 20, 4, pp.278-287, 1997.

4. Chung, S. H., H. W. Huang, “The block-based cycle time estimation algorithm for wafer fabrication, ”International Journal of Industrial Engineering, 6, 4, pp.307-316, 1999.

5. Chang, P. C., J. C. Hsieh, T. W. Liao, “A case-based reasoning approach for due-date assignment in a wafer fabrication factory, ”Proceedings of the 4th International Conference on Case-Based Reasoning, Vancouver, Canada, July 30-August 2, pp.648-659, 2001.

6. Chang, P. C., J. C. Hiseh, “A neural networks approach for due-date assignment in a wafer fabrication factory, ”International Journal of Industrial Engineering, July 2002.