Abstract Model Transformations (MT) are a central element of Model-Driven Engineering (MDE) methods. As MT adoption increases in both industry and academia, there is a growing need for systematic software engineering practices, particularly in Requirements Engineering (RE) for MT development. This paper investigates the state of RE in MT through two complementary empirical studies: semi-structured interviews with industry practitioners and a systematic literature review (SLR) analyzing published transformation cases. Both studies address the same research questions but differ in the populations they cover. The interviews focus on industrial settings, while the SLR reviews published work, the majority of which comes from academic sources. Our findings reveal that the RE processes used in MT development tend to be largely informal and lack structured methodologies. While some RE techniques such as prototyping and scenario-based generalization are used, they are typically applied in an ad-hoc manner based on personal experience rather than through a well-defined RE framework. Our studies highlight challenges in stakeholder engagement in MT RE, particularly limited access to stakeholders, which restricts the effective application of RE techniques. Furthermore, our analysis identifies a predominant focus on MT implementation, with limited MT specification and systematic RE activities, which often leads to requirements being implicitly defined rather than explicitly documented. Despite these shared findings, the interview study and SLR differ in their perspectives: the interview study reflects real-world industrial constraints on requirements engineering, while the SLR reflects more research-driven RE practices. These findings underscore the gap between research and practice in model transformations, and highlight the need for lightweight, structured RE frameworks tailored to MT development. Future work should focus on bridging this gap by integrating agile RE techniques with structured methodologies to support flexibility, traceability and stakeholder collaboration in MT projects.
Y. Tehrani et al. (Tue,) studied this question.