Software metrics for policy-driven software development life cycle automation

Abstract

Automation of SDLC requires continuous verification of compliance of the software product under construction to a set of expectations about its quality. We define a policy as an expectation about some aspects of software quality that is expressed as a collection of non-functional requirements (NFRs), compliance to which can be potentially measured. The results of such measurements can be used to verify whether the product meets the expectation set about it. In this paper, we discuss existing NFR taxonomies and propose mapping of software metrics to twelve NFRs. We then propose a model for reliability prediction using publicly available quality metrics for several open source projects.