Empirical study of embedded software quality and productivity

On a day-for-day basis, software managers struggle with the quality of the software process applied to their software engineering projects. Oftentimes the measurement and analysis needed to determine software process and product quality and to derive and set in motion strategic and tactical plans for the software organization are only partially conclusive. This paper provides a way for software organizations to study their software project data and characterize software productivity and product quality for individual projects and aggregates of software projects. Thirty-nine avionics software development projects are studied and results discussed. Armed with the results of these analyses, software managers can be more resolute in identifying specific candidate software engineering activities targeting key improvement areas -- areas most likely to improve software productivity, reliability, and quality directly affecting the company software bottom line.