Abstract
Quality of software product is mainly dependent on the influential human factors for software development, and introduction of software faults may be caused in the early development process such as design and coding phases. For improvement of software reliability and productivity, it is necessary to assess and predict software reliability more quickly, and to perform quality improvement actions such as quantitative clarifying and controlling the influence by the human factors to affect software reliability in the early development process. For the software design process, we have already proposed a new conceptual model and the "Taguchi-Method" based approach for analyzing the relationship among human factors and design review quality. On the other hand, based on the method of multiple regression, we have proposed a several software reliability prediction models for predicting the quality of a final product in the early development phase. In this paper, we discuss a new software reliability model to predict the number of faults remaining in the software product at the end of production phase based on the design reviewing characteristics. The result of an experiment shows that the proposed model is meaningful to improve the productivity of software development.
