Prediction of defect distribution based on project characteristics for proactive project management

As software has been pervasive and various software projects have been executed since the 1970's, software project management has played a significant role in software industry. There are three major factors in project management; schedule, effort and quality. Especially, to represent quality of products, there are various possible quality characteristics of software, but in practice, frequently, quality management revolves around defects, and delivered defect density has become the current de facto industry standard. The researches related to software quality have been focused on modeling residual defects in software in order to estimate software reliability. However, only the predicted number of defects cannot be sufficient information to provide basis for planning quality assurance activities and assessing them during execution. That is, in order to let projects managers be able to identify the project related information in early phase, we need to predict other possible information for assuring software quality such as defect density by phases, defect types and so on. In this paper, we propose a new approach for predicting distribution of in-process defects, their types based on project characteristics in early phase. For this approach, the model for prediction is established using the curve fitting method and the regression analysis. The maximum likelihood estimation is used in fitting the Weibull probability density function to the actual defect data, and the regression analysis is used to identify the relationship between the project characteristics and the Weibull parameters. The research model is validated by using cross-validation technique.