Deriving pertinent systems requirements : a methodology

Poor understanding of needs results in incompletely captured requirements and causes project failures. Analysts and developers by training, practice preoccupation and lacking suitable methodologies are ill equipped to capture various dimensions of requirements. In early stages needs are vaguely expressed. They have to be extracted and reasoned with stakeholders using relevant vocabulary. Business reality limits time at stakeholders' disposal to participate in requirements development. Methodologies that are more workable are required. Working with early stage goal oriented concepts from Goal Oriented Requirements Engineering (GORE) and fuzzy set theory based value aggregation, this thesis proposes a methodology to develop and select pertinent requirements. We use GORE concepts for identifying requirements and fuzzy aggregation to select alternatives. Use of fuzzy set theory and Ordered Weighted Aggregation (OWA) operators allows quantitative structuring of early stage decision problem consistent with human thinking and reasoning. The methodology is illustrated via two case studies.