A review of buildings dynamic life cycle studies by bibliometric methods

Abstract

This paper presents a bibliometric analysis examining dynamic life cycle (LC) studies in the building sector from 2007 to 2024, focusing on works that integrate or apply three key methods: Building Information Modelling (BIM), Machine Learning (ML), and Geographic Information Systems (GIS). By analysing a broad range of publications from the Web of Science database and Scopus database, we investigate publication trends and impacts, collaboration patterns, and highly cited work. A total of 549 core articles were identified within the study scope, with 3,136 additional records included for a sensitivity analysis. Our findings indicate that BIM-LC is the most prevalent, with extensive international collaborations. ML-LC, although a newer area, shows a rapid growth rate and potential in this area. GIS-LC shows steady contributions reflecting ongoing relevance to spatial sustainability. We also propose a conceptual framework illustrating how BIM, ML, and GIS can enhance dynamic LC studies in buildings, highlighting opportunities for practical application, such as standardised dynamic data and workflows, advanced algorithms, pilot experiments and real-world validation. By providing insights into bibliometric analysis and a conceptual framework, this review advances understanding of dynamic LC studies in the building sector.