.png){kind=icon}
.png){kind=icon}
Return
Unveiling interdependencies across phases: MICMAC analysis of BIM and AI integration challenges in construction
.png){kind=icon}
OA
AI
S
E
DOI：10.1080/17452007.2026.2658647.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The integration of Building Information Modeling (BIM) and Artificial Intelligence (AI) in the architecture, engineering, and construction (AEC) industry faces complex challenges across project phases, hindering adoption. This study pioneers the use of Cross-Impact Matrix Multiplication Applied to Classification (MICMAC) to systematically map direct and indirect interdependencies among 17 key challenges, extracted from a systematic review of 65 sources and validated by 10 industry experts. Employing a sociotechnical systems lens, the analysis reveals bivariate instability drivers—‘Data Integration Issues’ and ‘Data Management Concerns’—that propagate risk from design to operation via causal chains and feedback loops (e.g. design-phase data fragmentation → execution-phase real-time processing failure). While prior BIM–AI integration studies predominantly employ isolated barrier enumeration, ISM for hierarchy building, or DEMATEL for cause–effect mapping, they rarely reveal indirect causal propagation and bivariate instability amplifiers that span project phases. This study applies MICMAC—which uniquely classifies variables by direct + indirect influence/dependence power and exposes feedback loops and systemic instability—to uncover phase-crossing vulnerabilities and bivariate strategic variables (not identifiable via ISM’s hierarchy alone or DEMATEL’s pairwise causality without multiplication iterations). The resulting model provides the first phase-sensitive, predictive systemic framework for BIM+AI adoption risks in the AEC industry.
Keywords:
Building information modeling (BIM)
Artificial intelligence (AI)
Cross-impact matrix analysis (MICMAC)
System interactions
AEC industry
Journal
IF:
2.5
Papers:
183
Citations:
1.3K
Organization
Cited Papers