結合模糊集合與貝氏分類 應用於無線設備測試良率之研究

A Study on the Combination Fuzzy set and Bayesian classification to Apply the yield of test in Wireless device

范樹根(Fan-shu ken)

指導教授 ： 皮世明

中原大學/商學院/資訊管理學系/碩士(2006年)

https://doi.org/10.6840/cycu200600387

摘要

隨著生產技術快速創新，電腦產品趨向低利化，再加上網際網路及無線通訊之應用需求激增，提高產品生產良率，已成為延續產業競爭力的重要課題。為掌握關鍵製程良率，業者持續朝向智慧型生產領域發展，除資訊硬體工業外，亦涵括電子零組件及無線通訊等產業，逐漸形成完整的分工體系,彼此像混沌般的彼此互相影響。
本研究將探討無線設備廠生產良率改善方法，研究生產之相關的生產改善模式，找出影響生產良率改善之因素的關系，訂定良率改善模式，以利合作單位之實際需求。
在此將利用模糊集合及貝氏網路分析來找出影響無線網路設備製造最終影響生產良率的製程因子及進一步利用貝氏網路來找出製程因子間彼此的關系以作為未來生產良率改善的提升穩定性的方法

關鍵字

良率改善 ； 貝氏網路 ； 模糊集合

並列摘要

Regarding the new production technology, the margin of computer related product is getting lower and lower; also, the needs of Internet and wireless communication applications are increased, to improve the production yield rate becomes an important issue.
To improve the production yield rate, all enterprises keep working and finding an intelligent production process. Besides the hardware information industry, other industries are integrated as a complete functional system; and all industries are effected each other in this system.
This project focuses on the way to improve the production yield rate in the field of wireless product, such as improving the way to produce product to find out the key for the high yield rate and establishing the method to become a working mode for the implementation.
Here, this project will use fuzzy set and Bayesian network to figure out the production factors that effect on the wireless production. Use the relationships among these discovered factors as the keys to improve the production yield rate.

並列關鍵字

Bayesian network ； fuzzy set ； improve the yield rate

參考文獻

[1] Arroyo-Figueroa, G., Sugar, L.E., and Villavicencio, A., “Probabilistictemporal reasoning and its application to fossil power plant operation.”Expert System with Applications 15(1998) 317-324.

[2] Berry, M. J. A., and G. Linoff (1997), Data Mining Techniques, JohnWiley& Sons.

[3] Barrientos M.A., Vargas J.E., “A framework for analysis of dynamicprocesses based on Bayesian networks and case-based reasoning.”Expert System with Applications 15(1998) 287-294.

[4] Bilmes, Jeff(1998). A gentle tutorial of the EM algorithm and itsapplication to parameter estimation for Gaussian mixture and hidden

Markov model. International Computer Science Institute.