| Summary: | Due to the ever-changing needs of stakeholders, changes in environment and technology requirements have a tendency to evolve during system development. It is essential that prior to incorporating change the analyst must determine the impact of change on the system. It is not,desirable to incorporate a change without understanding and determining its impacts on the system. Requirements traceability (traceability) is the ability to "follow the life of a requirement. in both a forwards and backwards direction (i .e .. from. its origins. through its development and specification. to its subsequent deployment and use. and th.rough periods of on-going refinement and iteration in any of these phases)" (Gotel and Finkelstein 1994; Gotel and Finkelstein 1997). When requirements evolve, the analyst determines the impact by following the traces from the changing requirement to the dependent requirements, design, architecture, and source code. A few traceability approaches and tools capture requirements traces by referential links (trace From and traceTo, or source and target. id= "]". "2" ... id= "n ") or hyperlinks. Such traces are not expressive and they do not explicitly capture the rationale for the existence of dependencies between requirements. When a requirement evolves the impact set includes all the requirements that trace from the changing requirement. The analyst will have to analyse each individual requirement to identify the requirements actually impacted. For a large system with many requirements such trace information is not sufficient to rationalize and assess the impact of change. Hence, finer and more expressive traces will provide an effective and improved impact analysis. This thesis presents a requirements dependency taxonomy that comprises of the minimal set of semantically expressive and architecturally significant dependencies. It captures the rationale for the existence of dependencies between artefacts through explicitly defined meta-models that provide guidelines for capturing requirements dependencies. The aim is to enrich the traces with additional information (metadata), which makes the traces less coupled to the structure of the artefacts and more meaningful independently. There is a huge traceability gap between requirements and architecture (Griinbacher et at. 2001; Matthias 2006; Omoronyia et al. 2011). The existing approaches fall short in providing a better understanding of the system. A few deficiencies in the traceability approaches are because of the syntactic nature of the traces and mapping requirements to standalone architecture. The syntactic traces couple tightly to the structure of the artefact. The requirements trace to a standalone architecture that may not capture complete information and may not be adequate for change impact analysis. The requirements dependencies syntactically map to subsequent phase artefacts, which causes a loss of useful traceability information.
|
|---|
