A comprehensive literature review of 5D building information modelling using latent dirichlet allocation analysis

Purpose – This research evaluates the current state of 5D building information modelling (BIM) in cost estimation and management within the architecture, engineering and construction/FM industry. It identifies key areas of interest, synthesizes industry needs and highlights research gaps by analysing a large corpus of academic articles. Design/methodology/approach – A quantitative approach using latent Dirichlet allocation (LDA) analysed articles from Scopus and Web of Science. The methodology involved querying these databases for relevant articles on 5D BIM and cost estimation, applying LDA to identify and analyse emerging themes and synthesizing the findings for a comprehensive overview of the research landscape. Findings – The LDA analysis reveals themes in 5D BIM for cost estimation and management, including integration with cost tools, implementation challenges and practical case studies. It also highlights technological advancements, software innovations and educational needs for effective BIM adoption. Research gaps are noted in standardization, interoperability and the long-term impacts of 5D BIM on project lifecycle costs. Research limitations/implications – This study’s reliance on abstracts may miss key terms in full texts, though full-text analysis could introduce noise. Despite this, the findings offer valuable insights into trends and opportunities in 5D BIM research, guiding future studies and highlighting underexplored areas. Originality/value – This study provides novel insights by using LDA to analyse a large volume of 5D BIM research articles systematically. Unlike traditional reviews, LDA enables the objective identification of latent themes across over a thousand abstracts, reducing bias and ensuring scalability. By integrating natural language processing with domain-specific expertise, the study reveals critical issues in the construction industry by identifying research gaps.