An empirical study of selecting software development life cycle models

Abstract

This article discussed the problem that is commonly faced by project managers in selecting an appropriate software development model. A good number of models and the ongoing debate between the advocates of agile and traditional models make the selection of a right one uneasy and complicated. The purpose of this study is to propose a new alternative view in selecting development models from the perspective of degree of flexibility in different project situations. The results of a survey conducted to 74 project managers and software developers indicate that there are three naturally occurring development model groups, including heavyweight (Waterfall and V models), middleweight (Incremental and Spiral models), and lightweight (Extreme Programming and SCRUM). A canonical function is identified to help project managers select an appropriate software development model for each potential or planned project.