The continuous traceability of all artefacts of a software development process is demanded by major standards like DO-178B, the V-Model XT and CMMI. Safety-critical systems are not accepted without verification of continuous traceability - from requirement to model down to the code and its test. The captured traceability links, as part of the software system documentation, serve as basis for the following analyses:
According to empirical studies, traceability is a process which so far has not been adopted on a wide front in software engineering, despite its clear advantages. Reasons for this lie in the lack of a generic and formal definition of a traceability model, in the low degree of integration of CASE tools concerning traceability purposes and in the resulting considerable effort for maintaining and analyzing traceability links, in addition to the lack of knowledge regarding the evolution of traceability during software development.
Current research in the area of traceability focuses on the automated capture of traceability links to guarantee high quality link bases. What, however, is still missing is a generic approach to the systematic and quantifiable assessment of the available link bases, such as to guarantee the reliability of impact and derivation analyses and the identification of trace anomalies far beyond the capabilities of gap analyses.
Furthermore no approaches exist allowing to derive correction strategies once trace anomalies are located and preferably in early project phases.