Requirements engineering for software product lines: A systematic literature review

摘要

Context: Software product line engineering (SPLE) is a growing area showing promising results in research and practice. In order to foster its further development and acceptance in industry, it is necessary to assess the quality of the research so that proper evidence for adoption and validity are ensured. This holds in particular for requirements engineering (RE) within SPLE, where a growing number of approaches have been proposed.rnObjective: This paper focuses on RE within SPLE and has the following goals: assess research quality, synthesize evidence to suggest important implications for practice, and identify research trends, open problems, and areas for improvement.rnMethod: A systematic literature review was conducted with three research questions and assessed 49 studies, dated from 1990 to 2009.rnResults: The evidence for adoption of the methods is not mature, given the primary focus on toy examples. The proposed approaches still have serious limitations in terms of rigor, credibility, and validity of their findings. Additionally, most approaches still lack tool support addressing the heterogeneity and mostly textual nature of requirements formats as well as address only the proactive SPLE adoption strategy. Conclusions: Further empirical studies should be performed with sufficient rigor to enhance the body of evidence in RE within SPLE. In this context, there is a clear need for conducting studies comparing alternative methods. In order to address scalability and popularization of the approaches, future research should be invested in tool support and in addressing combined SPLE adoption strategies.