Collaborative Model-Driven Software Engineering: A Classification Framework and a Research Map

TLDR

Collaborative Model‑Driven Software Engineering (MDSE) involves multiple stakeholders managing shared models, and recent research has produced a diverse body of knowledge on the topic. This study aims to identify, classify, and understand existing collaborative MDSE approaches. We conducted a systematic mapping study that screened over 3,000 papers, selected 106, clustered them into 48 primary studies, and applied a classification framework to extract key information. The analysis shows a growing interest in collaborative MDSE, highlights under‑explored aspects such as multi‑view modeling and synchronous/asynchronous collaboration, reveals that most approaches are UML‑based and language‑specific, and provides a foundation for future research and practice.

Abstract

Context: Collaborative Model-Driven Software Engineering (MDSE) consists of methods and techniques where multiple stakeholders manage, collaborate, and are aware of each others' work on shared models. Objective: Collaborative MDSE is attracting research efforts from different areas, resulting in a variegated scientific body of knowledge. This study aims at identifying, classifying, and understanding existing collaborative MDSE approaches. Method: We designed and conducted a systematic mapping study. Starting from over 3,000 potentially relevant studies, we applied a rigorous selection procedure resulting in 106 selected papers, further clustered into 48 primary studies along a time span of 19 years. We rigorously defined and applied a classification framework and extracted key information from each selected study for subsequent analysis. Results: Our analysis revealed the following main fidings: (i) there is a growing scientific interest on collaborative MDSE in the last years; (ii) multi-view modeling, validation support, reuse, and branching are more rarely covered with respect to other aspects about collaborative MDSE; (iii) different primary studies focus differently on individual dimensions of collaborative MDSE (i.e., model management, collaboration, and communication); (iv) most approaches are language-specific, with a prominence of UML-based approaches; (v) few approaches support the interplay between synchronous and asynchronous collaboration. Conclusion: This study gives a solid foundation for classifying existing and future approaches for collaborative MDSE. Researchers and practitioners can use our results for identifying existing research/technical gaps to attack, better scoping their own contributions, or understanding existing ones.