Software Product Line Adoption in Industry: Challenges and Opportunities
Software systems are often developed as variants to target a variety of market segments. In addition, software working with hardware is deployed on various hardware architectures and is required to process various signals across those deployments. Handling such a high degree of variance in product development and testing requires a variability-aware process.
RISE Research Institutes of Sweden and Mälardalen University, in the scope of the ITEA XIVT project project, performed an investigation regarding the current state of product-line engineering adoption in the railway domain for the development and management of variant-intensive software systems. The findings are collected using mixed-method research through current practice analysis and focus groups.
The investigation shows that incremental adoption of software product line engineering (SPLE) even with unsystematic reuse may reduce development and testing time, boost confidence in the product, and reduce safety certification efforts. However, identifying reuse opportunities based on requirements, change propagation and impact analysis, test automation, and SPLE awareness remain open areas for improvement.
The future vision the XIVT project identified is a take-away for the industry: establish a product-wide regression suite, introduce a continuous integration pipeline, and apply automated test generation in combination with reuse recommendation, test reuse, variability-aware testing and verification, change impact analysis, and traceability link recovery. This will improve your way of working in product line engineering, enhance the quality of software, and reduce the lead time by lowering the product development time.
More information:
Related Papers:
[1] Muhammad Abbas, Robbert Jongeling, Claes Lindskog, Eduard Paul Enoiu, Mehrdad Saadatmand, and Daniel Sundmark; “Product line adoption in industry: an experience report from the railway domain”; 24th ACM Conference on Systems and Software Product Line: Volume A (SPLC); 2020; ACM, New York, USA; https://doi.org/10.1145/3382025.3414953
[2] Muhammad Abbas, Mehrdad Saadatmand, Eduard Paul Enoiu, Daniel Sundmark, Claes Lindskog; "Automated Reuse Recommendation of Product Line Assets Based on Natural Language Requirements”; International Conference on Software and Systems Reuse (ICSR); 2020; Lecture Notes in Computer Science, vol 12541. Springer; https://doi.org/10.1007/978-3-030-64694-3_11
