Automation in Model-Driven Engineering: A look back, and ahead

TL;DR

The paper revisits automation in Model-Driven Engineering (MDE), tracing how automation has evolved from early model-to-code and model-checking activities to runtime execution and ongoing model management. It surveys the main automation activities, the enabling technologies that support them, and provides a mapping between tasks and technologies, highlighting AI, human factors, and non-functional concerns as central future directions toward 2030. A key contribution is documenting the state of the art across four enabling-technology families—Formal Methods, Search-based Techniques, extensional/intensional knowledge, and AI/Large Language Models—and identifying gaps, such as data scarcity and interoperability, that constrain AI-driven automation in MDE. The work emphasizes the need for open datasets, benchmarks, and evaluation frameworks, and advocates a vision of AI-augmented, human-centered MDE that can handle increasingly complex, socio-technical systems across diverse domains.

Abstract

Model-Driven Engineering (MDE) provides a huge body of knowledge of automation for many different engineering tasks, especially those involving transitioning from design to implementation. With the huge progress made in Artificial Intelligence (AI), questions arise about the future of MDE, such as how existing MDE techniques and technologies can be improved or how other activities that currently lack dedicated support can also be automated. However, at the same time, it has to be revisited where and how models should be used to keep the engineers in the loop for creating, operating, and maintaining complex systems. To trigger dedicated research on these open points, we discuss the history of automation in MDE and present perspectives on how automation in MDE can be further improved and which obstacles have to be overcome in both the medium and long-term.