Journal "Software Engineering"
a journal on theoretical and applied science and technology
ISSN 2220-3397

Issue N9 2017 year

DOI: 10.17587/prin.8.387-395
Product Line Engineering: Standards and State-of-Practice
A. S. Khritankov, anton.khritankov@acm.org, Moscow institute of physics and technology, Dolgoprudny, Moscow Region, 141701, Russian Federation
Corresponding author: Khritankov Anton S., Associate professor, Moscow institute of physics and technology, Dolgoprudny, Moscow Region, 141701, Russian Federation, E-mail: anton.khritankov@acm.org
Received on June 14, 2017
Accepted on June 20, 2017

A recent version of the ISO/IEC 26555 standard comes into force as a national standard in Russia in June 2017. The standard describes technical management processes for systems and software product lines. In this paper, we review the standard and the current state of research and industrial practice in the field. Mass product customization in Industry 4.0, Industrial Internet (IoT), ubiquitous computing are major pillars in the Horizon 2020 programme of the EC. And this means there is need for a lot of customizable software systems. Software product lines may be the key. While a popular topic worldwide, SPLE has missed attention in Russian Federation with only a few papers on the topic. Thus we briefly introduce main concepts of the SPLE such as product line, variability model, strategic reuse, reference architecture, feature and decision models, product configuration. Then we provide an example feature model built with FeatureIDE. Applications of PLE include shared development of a family of products each targeting a specific market segment, development of a large evolving cyber-physical system with several versions delivered over time and a combination of both. International standardization process started with a failed attempt of ISO/IEC 26520 series and continued with more successful ISO/IEC 26550-26564 standards, some of which is still under development. Russian national standardization committee adopted the SPL technical management ISO/IEC 26555:2013 as national standard in 2016, omitting other published standards in the series. Extensions and additions to the standard should consider tracking rationale for the changes being introduced to the SPL configuration and variability model in particular. Another shortcoming is absence of a model for gradual introduction of the standard (e.g. capability or level-based). Process capability assessment may be governed by ISO/ IEC 33001, but this option is not mentioned in the standard. Current research centers in SPLE are Univ. Litz, TU Darmstadt, Univ. Magdeburg, Univ. Passau and other in EU; Univ. Pernambuco, Univ. Namur, Univ. Bahia and other in Brazil; Univ. Waterloo, Univ. Toronto, SEI/CMU in US and Canada, several universities in China and South Korea. EU also runs a series of projects uniting research universities and commercial companies to help spread fresh developments in PLE, such as FAMILIES, VARIES and current REVAMP2 (2016-2019). Major research directions are variability modeling and languages (with and without UML/SysML), verification and traceability, variability model extraction, concurrent evolution of variability models and assets, transition to clone-and-own and PLE, and, notably, model-based software engineering (MBSE) for SPLE. Notable industrial users of SPLE are Siemens, Philips, General Motors, ABB Group with others following, including Thales and SAAB (through Celsius Tech). Although PLE may be the next (old) big thing in software development to support mass customization through strategic reuse, it requires a considerable upfront investment and strong process discipline. Currently developed standards and available tools may help practitioners and tool vendors lower this investment to a level, manageable by small and mediun range companies.

Keywords: Internet of Things (loT), PLE, product lines, ISO/IEC 26550, strategic reuse
pp. 387–395
For citation:
Khritankov A. S. Product Line Engineering: Standards and State-of-Practice, Programmnaya Ingeneria, 2017, vol. 8, no. 9, pp. 387—395.