Towards a framework for risk identification and mitigation in Agile software re-engineering: A case study

Abstract

Legacy software is becoming increasingly ubiquitous, and many companies nowadays are facing the challenges associated with this phenomenon. In certain circumstances, re-engineering is the only logical way to deal with legacy software. Such projects, by their very nature, are subject to a wide variety of risk. The first research question of this study was concerned with the identification and investigation of different risks associated with one particular re-engineering project. The potential impact of those risks over a given development phase of the project, along with their actual impact and ultimate mitigation status, have been analysed. The second research question discussed the Scrum practices used on the project and evaluated their usefulness in supporting the management of risk. An interpretive Case Study approach has been followed, with the method of analysis being inductive and reflexive Thematic Analysis. The risks observed were themed around people, process, and technology. While technical and procedural risks are discussed in the literature, the presence of risk in social situations relating to re-engineering appears to have been overlooked. Although these risks do not necessarily have a higher impact, they were found to outnumber those encountered in other aspects of the project by a significant factor. Furthermore, the social risks were often either underestimated or not even recognised. It has also been found that Scrum is an appropriate approach to re-engineering projects. Since many of the re-engineering tasks in the case study were unknown at the beginning, the flexibility brought by Scrum was an important factor in the timely and successful mitigation of emerging risk. My contribution to the field is an initial set of risk categories including their impact on a re-engineering project, which are able to form the basis of further research into different types of re-engineering projects in order to produce a more generalised framework. It is anticipated that the results presented here will help future project teams to prioritise aeras of re-engineering and put adequate risk mitigation into place.