於品質管理領域,驗收抽樣計畫(Acceptance Sampling Plan)為一被廣泛應用之統計管制工具,能提供供應商和顧客一決策法則判別貨批接受或拒絕,其目的為確保產品品質及誤判之風險達到雙方既定的要求。而本研究決定出一新型驗收抽樣計畫,以目前廣泛應用於供應商與顧客量化準則之製程能力指標Cpk為發展基礎,且有別於過往文獻只單純應用傳統單次驗收抽樣計畫(Single Sampling Plan)(其操作流程為:從貨批中隨機抽樣n 個樣本,利用這次所抽取的樣本資訊,即判斷整批貨批接受拒絕與否),本研究應用兩階段驗收抽樣計畫(Two-Stage Sampling Plan),克服傳統單次抽樣計畫僅有一次判斷機會且所需抽樣樣本數較高之缺點。此外,為驗證及比較本研究所提出新型驗收抽樣計畫與過往驗收抽樣計畫(單次驗收抽樣計畫與重複群集驗收抽樣計畫(Repetitive Group Sampling Plan))之績效,本研究進行相關分析及探討,結果顯示不論是在效率上或是經濟考量上本研究所提出新型驗收抽樣計畫皆優於過往之驗收抽樣計畫。最後,本研究求解及彙整多種「品質水準」與「風險要求」組合之計畫參數值並提供案例以利實務上應用及參考。
In the field of quality control, acceptance sampling plan has been one of the widely used statistical quality tools which provides the producer and the customer a general rule for lot sentencing by satisfying their specified quality and risk requirements. In this paper, we propose a new-type acceptance sampling plan based on the popular used capability index Cpk which is specialized in measuring the required process capability for the producer and the customer. Different from the existing acceptance sampling plans which were developed by using single sampling strategy which is a lot-sentencing procedure in which one sample of n units is selected at random from the lot, and the disposition of the lot is determined based on the information obtained in this sample, we develop the new-type sampling plan by using Two-Stage sampling strategy which requires less sampling and provides one more chance to the submitted lot compared with the traditional single sampling plan. Otherwise, we analysis and discuss the performance of the proposed variables plans compared with the existing sampling plans including single sampling plan and repetitive group sampling plan. The results indicate that variables two-stage sampling plan can be served as an ideal sampling method for quality assurance whether in economically or efficiently. Finally, we not only tabulate the plan parameters for various required quality levels and allowable risks and provided an example for practical applications.