This study seeks to bridge the gap between artificial intelligence (AI) innovation and facility management (FM) practice by identifying dominant research clusters, examining thematic evolution and revealing opportunities for sustainable and intelligent facility operations. Specifically, it addresses how generative models including large language models, generative adversarial networks and explainable AI are reshaping FM workflows and decision-making processes.
FM research is witnessing increasing attention toward generative artificial intelligence (GenAI), yet current studies remain fragmented and lack a systematic understanding of how these technologies are being integrated into FM research and practice. This study aims to systematically map and analyze the thematic evolution of GenAI applications in FM using structural topic modeling (STM) combined with qualitative synthesis. STM was used to identify latent themes and track their temporal evolution, supported by descriptive trend analysis. Manual triangulation and qualitative synthesis were conducted to validate topic labels and interpret emerging research trajectories, technological applications and integration pathways across FM contexts.
Five thematic clusters were identified: (1) GenAI for visual and spatial applications in built environments, (2) industrial GenAI for sustainability optimization, (3) predictive maintenance with explainable and generative AI, (4) smart buildings for energy and comfort optimization and (5) natural language processing for FM automation.
To the best of the authors’ knowledge, this paper offers one of the first STM-based systematic reviews of GenAI in FM. The findings provide a replicable, data-driven framework and a practical roadmap for integrating GenAI-driven automation into FM processes, contributing to the theoretical and operational advancement of AI-enabled facility ecosystems.
