Abstract
During collaborative configuration of software product lines (SPL), multiple stakeholders contribute together in building a single product specification. Conflicting situations can arise during the configuration process due to contradictions between some/all stakeholders’ configuration choices. Detecting and resolving such situation rise two major challenges: choosing which choices to omit, and taking stakeholders’ preferences into account. Several SPL collaborative configuration approaches are available. However, they either do not present detailed information on the strategies for conflict resolution, or they rely on a systematic process that resolves conflicts by prioritizing configuration decisions made at earlier stage, constraining therefore some of stakeholders’ choices. The lack of flexibility may hinder conflict resolution as choices taken at earlier stages overlay those at later phases. To mitigate these limitations, we propose a new collaborative configuration approach (Colla-Config) that provides a preference-based conflict resolution method within a free-order configuration process; each stakeholder expresses his/her preferences through a set of substitution rules, and freely makes his/her configuration decisions towards the desired product, without being constrained by the configuration decisions made by the other ones. To assess the feasibility and the usability of the proposed approach, we conducted a usability test designed following the ISO/IEC 25062:2006 Common Industry Format for usability tests. Results of the experiments provide preliminary evidence of the approach feasibility and the tool ability to properly support the SPL collaborative configuration.
•Proposing a new approach for product lines collaborative configuration.•Providing algorithms to ensure a preference-based conflict resolution method within a free-order configuration process.•Supporting the approach with a tool, and evaluating its usability.•Following the ISO/IEC 25062:2006 Common Industry Format for usability tests.•Evaluating the scalability of the solution.