Abstract

Software quality is directly correlated with the number of defects in software systems. As the complexity of software increases, manual inspection of software becomes prohibitively expensive. Thus, defect prediction is of paramount importance to project managers in allocating the limited resources effectively as well as providing many advantages such as the accurate estimation of project costs and schedules. This thesis addresses the issues of statistical fault prediction modeling, software resource allocation, and optimal software release and maintenance policy. A software defect prediction model using operating characteristic curves is presented. The main idea behind this predictor is to use geometric insight in helping construct an efficient prediction method to reliably predict the cumulative number of defects during the software development process. Motivated by the widely used concept of queue models in communication systems and information processing systems, a resource allocation model which answers managerial questions related to project status and scheduling is then introduced. Using the proposed allocation model, managers will be more certain about making resource allocation decisions as well as measuring the system reliability and the quality of service provided to customers in terms of the expected response time. Finally, a novel stochastic model is proposed to describe the cost behavior of the operation, and estimate the optimal time by minimizing a cost function via artificial neural networks. Further, a detailed analysis of software release time and maintenance decision is also presented. The performance of the proposed approaches is validated on real data from actual SAP projects, and the experimental results demonstrate a compelling motivation for improved software quality