Quantitative framework for managing software life cycle

Abstract

The paper presents a rigorous and practical way of managing the quantitative yields of new (improved) software engineering methods in the software life cycle. A meaningful set of metrics and models is used to measure the value of applying tailored methods. The value-based framework is proposed as a system of integrated models (such as cost models, productivity models, quality-related models, benefit models, and value-related models) that combine project data and expert opinion. The framework is proactive, as it allows to estimate the value-based metrics of a software engineering method in order to monitor it, improve software quality and convince developers and managers that the method and related investment are worthwhile. To demonstrate the applicability of the proposed framework, the results of a case study are presented and used as an initial validation of the framework. In this paper we use the model system concept to address the problem of quantifying the yields of new (improved) software engineering methods and suggest a value-based framework (VF). VF entails the application of interrelated metrics and models from (3-7) for estimating the relevant costs, the benefits translated into software quality management and improvement, as well as the resulting value-related metrics evaluated during the software system life cycle. The framework is applicable at a high level of abstraction and offers useful results from a practical point of view, as our case study indicates.