Optimizing Software Quality Assurance

Abstract

A major concern for managers of software projects are the triple constraints of cost, schedule and quality due to the difficulties to quantify accurately the trade-offs between them. Project managers working for accredited companies with a high maturity will typically use software cost estimation models like COCOMO II and predict software quality by the estimated number of defects the product is likely to contain at release. However, most of these models are used separately and the interplay between cost/effort estimation, project scheduling and the resultant quality of the software product is not well understood. In this paper, we propose a regression-based model that allows project managers to estimate the trade-off between the quality, cost and development time of a software product, based on previously collected data.