科技的發展使得人類愈來愈依賴軟體系統，軟體品質因此扮演著愈來愈重 要的角色。為確保軟體品質，許多專家學者們針對軟體可靠度進行了探討和研 究，並建立相關的知識基礎以及推導出許多軟體可靠度成長模型，以預測軟體的 失效行為。本研究提出了一個以非齊次卜瓦松過程（Non-Homogeneous Poisson Process, NHPP）為基礎的可靠度成長模型，當中考慮了測試心力函數的概念，結 合於錯誤偵測率函數。模型建構完成後，透過實際失效資料集進行參數估計與模 型評量，讓新模型與現有模型進行比較分析，藉此評估新模型是否可以更有效地 預測軟體失效行為。

The strong development of technology makes humans rely more and more on software systems, software quality therefore playing an increasingly important role. To ensure software quality, experts has been studying and doing research on software reliability, creating relevant knowledge base and derive many software reliability growth models to predict the failure behavior of software systems. In this study, we propose a reliability growth model base on the Non-Homogeneous Poisson Process, which consider the concept of Testing Effort Function into Fault Detection Rate Function. In addition, numerical examples are given to verify the effectiveness of the proposed models, and compare to the existing ones in order to make sure if the new model had more accurate prediction capability.