容許區間樣本數的決定 －藥物製程檢測樣本數的應用

Abstract

容許區間是一種被廣泛利用的統計工具，一直以來更是藥物製程的檢驗時經常使用的工具。藥物製程檢驗時所需抽取的樣本數是各大製藥公司長久以來所關注的議題，其背後代表的不僅僅是成本的考量，也是藥物品質的保證。因此本篇論文利用文獻上所提供的一些準則，著眼於研究容許區間樣本數選取的方法，並提供一套演算法來計算不同資料所需的容許區間樣本數，並在最後利用各種模型假設下進行統計模擬以驗證研究結果。此論文所提出的這一套樣本數選取方法，可適用於不同分配，相信對各大藥廠在關於藥物製程檢驗時所抽取的樣本數決策時有不少的幫助。

Tolerance is a widely used statistical tools, and has been a useful tool for pharmaceutical testing. The determination of sample size of a pharmaceutical testing has been one of important issues for pharmaceutical companies, because it does not only represent cost considerations, but also is a quality assurance.Therefore, in this study, we develop an algorithm for determining the sample size of a tolerance interval based on historical data under some criteria provided in literatures focusing on the study of determination of sample size for tolerance interval. The required sample sizes for different cases are provided in this study. A simulation study is conducted to verify the result derived from the algorithm. The method proposed in this study can help pharmaceutical companies in choosing the sample size for drugs testing.