Premium
Transformation towards agile software product line engineering in large companies: A literature review
Author(s) -
Klünder Jil AnnChristin,
Hohl Philipp,
Prenner Nils,
Schneider Kurt
Publication year - 2019
Publication title -
journal of software: evolution and process
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.371
H-Index - 29
eISSN - 2047-7481
pISSN - 2047-7473
DOI - 10.1002/smr.2168
Subject(s) - agile software development , software product line , agile unified process , computer science , agile usability engineering , domain engineering , software engineering , lean software development , software development , domain (mathematical analysis) , software , model transformation , requirement , product (mathematics) , systems engineering , process management , software development process , software construction , engineering , artificial intelligence , operating system , mathematical analysis , geometry , mathematics , consistency (knowledge bases)
Almost all companies struggle with software systems that are getting increasingly complex. Therefore, in particular large companies often use software product lines, which provide an efficient way to manage software reuse as well as the high complexity. However, software product lines seem to be too slow to react to changes. Agile development approaches promise to meet this demand. At the same time, integrating an agile approach is not always easy. Consequently, there is a need for a transformation model that supports the agile transformation without harming the benefits of software product lines. In this paper, we present the results of a literature review regarding agile transformation in large software companies. Our results summarize the insights from 85 papers and present a list of 20 tasks and tasks that are recommended by several researchers to be considered during the transformation. On the basis of these results, we create an agile transformation model—the so‐called “agile hamburger”—that contains these tasks and preserves software product lines. Since the model is rather coarse grained and generic, we also support the adaption to different teams by presenting an assessment model tailored to the demands of the automotive domain.