A Systematic Review of Traceability in Requirements Engineering of Socio-technical Systems: Industrial Practices and Needs

Abstract

Requirement traceability (RT) is useful for determining the effects of differences between various artefacts in the software development lifecycle. RT assures the whole system's quality and accuracy. It also helps the change management process by providing a logical relationship between software development practice deliverables. This thesis examined important methods in current RT activity and classified existing RT techniques based on their domains and requirements. Furthermore, we examined the tools, industrial practices, approaches, and their contributions and trade-offs and highlighted prospective future study topics with important results. Empirical, qualitative, quantitative, agile-based, hybrid, blockchain, natural language processing (NLP), and machine learning (ML) techniques are examined using 46 research studies. The systematic literature review has identified (i) the practices that are popular in literature for RT of socio-technical systems (STSs) and their implications in the industry and (ii) the new trends and approaches that are adopted by the literature and industry. The results and findings from the literature show the variability of requirements and their impact on the overall system. Also, the systematic review (SR) has tried to cover the research gaps of previous literature reviews that only covers the domain-specific literature and case studies. Overall, the SR has analysed the work done in the past five years in the field of RT in STSs. The SR concludes that various practices need to be explored for further studies, which include blockchain, NLP, and ML.