Supporting Deviations on Software Processes: A Literature Overview

Software Process (SP) models are the results of the efforts deployed by the software Engineering community to guarantee an advanced level of the SP quality. However, experience has shown that SP agents often deviate from these models to cope with new environments’ challenges. Unfortunately, the appearance of such situations, if not controlled, often lead to the process failure. Since the 90s, several research works have been conducted to handle this problem. Through this paper, we aim at gathering these approaches around a single classification that puts in advance their strengths and their weaknesses. To achieve this goal, we propose two classification frameworks that highlight how existing approaches deal with deviations from two different axes: detection and correction. As a result of this classification, a covering graph is drawn for each framework, which gives an insight about what has been left by the existing approaches and worth to be considered, further. Finally, we introduce briefly the general outlines of a new contribution that we are currently working on to face the shortcomings of the existing approaches.