AI-Assisted Data Modelling & Design: A Systematic Literature Review

Purpose: This paper explores how artificial intelligence (AI) supports Data Modelling and Design (DM&D) across its lifecycle and how human-in-the-loop (HITL) mechanisms can enhance model quality in enterprise governance. Methodology/Approach: A PRISMA 2020–guided systematic review (2023–2025) of Scopus, Web of Science, and ACM Digital Library identified 28 eligible studies. Evidence was synthesised along the DAMA P–D–C–O cycle, with a focus on HITL. Findings: AI supports planning, building, reviewing, and managing data models through schema generation, enrichment, validation, and optimisation. Results vary with model accuracy, data quality, and semantic gaps. Effective use relies on HITL workflows such as propose–validate, tutoring, co-editing, and feedback loops. A scorecard combining technical, performance, efficiency, and governance indicators traceable via metadata helps demonstrate impact. Research Limitation/Implication: Findings reflect studies published between 2023 and 2025; results may evolve as AI capabilities progress. Originality/Value of paper: This study presents the first lifecycle synthesis of AI-assisted DM&D. It organises evidence from 28 studies, defines key HITL patterns for quality assurance, and outlines ways to evaluate AI’s role in modelling governance. The findings provide a basis for further research and for developing frameworks that connect AI-driven modelling with enterprise data governance.