Application of Regression-Based and Function Point Software Sizing Techniques to Air Force Programs

摘要

This study investigated the ability to develop regression models to predict the number of source lines of code (SLOC) and the degree of correlation between function points and the number of SLOC. Since software size -- or the number of lines of code (LOC) -- is considered to be the primary software cost driver, accurate software size estimates are critical because of the growing importance of software in today's Department of Defense (DoD) weapon systems. For the regression analysis, four sizing data bases, containing various functional (Independent) variables, were used. These variables included complexity, reliability, the program's quality of specification, etc. Theses. (JES)