摘 要 過去我們常見到各種品質工具在產品開發階段或生產製造階段的運用，但是如果只是單 獨運用各品質工具，則會流於局部與片段的改善。由於一般企業大都資源有限，如何透過良 好的程序與使用簡易的方法來處理產品設計開發上的品質問題，則往往才是企業界最關注的 重點。本研究嘗試以六標準差設計 (DFSS) 的系統觀，透過運用Kano模式、電子零件失效率 分析、環境可靠度試驗等之品質工具，結合DMADV (Define, Measure, Analyze, Design, Verify) 程序作創新流程的整合，來提升ODM 電子產品之設計品質。 本研究方法實際在M 公司運用後，以TL9000 通訊產業品質管理系統國際標準所引用之 退貨指標YRR (One-Year Return Rate)，作為量化衡量產品設計品質與可靠度的依據，並將最 近一年實際數據作比較分析，發現採用DFSS 之ODM 產品族群在YRR 指標的表現上，要明 顯比未採用DFSS 之OEM 產品族群好，而且其相對值可高達平均約25 倍的差距，若轉換成標 準差，則約有1.12 σ 的差距。在DFSS 的應用領域上，如果能充份抓住客戶與市場的需求，並 有效分配公司可運用之資源，發展出適合自己公司產品開發的方法，可大幅提高企業競爭力。

Abstract Organizations that have adopted the principles of the Six Sigma methodology have become aware that once they have achieved the five sigma quality levels, the only way to progress further is to redesign their products or processes. This has led to the development of Design for Six Sigma (DFSS). Recently, DFSS has gained much popularity due to the widespread application of Six Sigma principles, particularly in new product development. This research employs the systematic view of the DFSS process and combines it with quality tools to improve the design quality of Original Design Manufacture (ODM) electronic products. A wireless communication company applying the DFSS methodology was analyzed, and an index of one-Year Return Rate (YRR) from TL 9000 Quality Management System Requirements of the worldwide telecommunication industry was used to evaluate the measurable effect of product design quality. The average YRR result of an ODM product group employing the DFSS process was observed to be better than an Original Equipment Manufacture (OEM) product group without the DFSS process. Furthermore, its relative value can be improved up to 25 times or 1.12 sigma. By applying the DFSS methodology, the demands of customers can be further satisfied, and the specific needs of the customer are fulfilled.