整合台灣製造業不同分群下存貨掌控度關鍵影響因素之研究

A research of integrating various clustering models to find the key factors on inventory control for Taiwan's manufacturing

吳家齊(Jia-Chi Wu)

指導教授 ： 張淳智

國立臺中科技大學/資訊與流通學院/流通管理系碩士班/碩士(2014年)

https://doi.org/10.6826/NUTC.2014.00084

摘要

由於台灣製造業體系與供應鏈有所改變，對於製造業加以分群及分析的研究與日俱增，在眾多的分群模式之下，應選擇何種模式方能符合特性不一的廠商成為一個值得研究的課題。因此，若能整合各分群模式並給予相同比較基礎，則將更能針對不同廠商提供不同模式的改善方針。

本研究以張淳智與顏憶茹(民102)之研究作為模式的基礎，在改善方針方面整合張淳智與顏憶茹(民101)、張淳智與顏憶茹(民102)及黃彥翔(民102)三種分群模式，並修正張淳智與顏憶茹(民102)及黃彥翔(民102)兩模式之疏漏並利用倒傳遞網路重建因果模型以找出關鍵影響因素。

本研究將原始資料重新代入各模式並計算誤差，結果顯示本研究所整合之模式之誤差皆低於其他三個模式，表示在特性不一的廠商中，整合模式比單一模式更能有效地找出其關鍵影響因素。

關鍵字

台灣製造業 ； 分群模型 ； 關鍵因素

並列摘要

Because Taiwan's manufacturing system and supply chain have changed, the researchers on Clustering and analysis of manufacturing are steadily increasing. Under the numerous clustering models, which model should be selected to meet the manufacturers with different characteristics is worthy of study. Therefore, if we could integrate the clustering models and given the same basis of comparison, it will be better to provide improved guidelines for various manufacturers.

In this study, the integrated model is based on Chang and Yen (2013). Three clustering models (Chang and Yen, 2012; Chang and Yen, 2013; Huang, 2013) were integrated in order to improved guidelines for firms. Furthermore, this research corrected the deficiencies of Chang and Yen (2013) and Huang (2013), and reconstructed the causal models using back propagation network model to identify the key factors on inventory control.

Finally, we put the raw data into each model and calculate the prediction error, the results showed that the prediction error of proposed integrated model was lower than the other three models. That improved the integrated model is more effective than the single model to identify the key factors of the manufacturers with different characteristics.

並列關鍵字

Taiwan’s manufacturing ； clustering models ； key factors

參考文獻

賴瑩嬛(2012)。應用群集分析探討侵臺颱風之分類特性。國立交通大學土木工程學系碩士班學位論文。

Budayan, C., Dikmen, I., & Birgonul, M. T. (2009). Comparing the performance of traditional cluster analysis, self-organizing maps and fuzzy C-means method for strategic grouping. Expert Systems with Applications, 36(9), 11772-11781.

De Reyck, B., Degraeve, Z., & Vandenborre, R. (2008). Project options valuation with net present value and decision tree analysis. European Journal of Operational Research, 184(1), 341-355.

Fayyad, U., Piatetsky-Shapiro, G., & Smyth, P. (1996). From data mining to knowledge discovery in databases. AI magazine, 17(3), 37.

Han, J., & Kamber, M. (2006). Data Mining, Southeast Asia Edition: Concepts and Techniques. Morgan kaufmann.