Recent research in requirements engineering has identified two strong drivers of requirements evolution: environmental turbulence mediated by system stakeholders and the requirements process that accompanies the design process. Current methods in requirements engineering fail in managing evolving requirements. Most of the methods do not support the different views of the stakeholders involved in the requirements process. Moreover, they do not provide any means for predicting the impact in terms of risk and cost of requirements ' changes. The paper shows an empirical investigation of an avionics safety-critical industrial case study pointing out aspects of requirements evolution. In the paper we propose a framework for predictive changes risk/cost analyses relating type of changes into requirements, software life cycle, impact of changes and structure of requirements. The results suggest further works dealing with requirements evolution.
|Title of host publication||Forsight and Precaution, Proceedings of ESREL 2000, SARS and SRA - European Annual Conference, 2000.|
|Publication status||Published - 2000|