Unravelling the Blockades in Implementing Large Language Models in Construction Sector

Abstract

The adoption of text-based Generative AI (GenAI), especially Large Language Models (LLMs), offers promising opportunities for improving documentation, coordination, compliance, and decision-making in construction Operations and Supply Chain Management (OSCM). Yet multiple constraints limit effective deployment in the Indian construction context. This study investigates the blockades hindering GenAI adoption and models their interrelationships using expert evaluations and a Grey-DEMATEL approach. Ten interconnected blockades are identified across technological, data-integration, and organizational–institutional domains. Real-time data processing, model accuracy and domain validity, and computational resource requirements emerge as the most influential cause blockades, shaping downstream challenges in collaboration, regulatory alignment, and workflow management. By revealing how these blockades interact, the study provides a structured framework for understanding GenAI adoption barriers and offers theoretically grounded implications for policymakers, contractors, SMEs, and technology providers. The findings deliver a focused evidence base to guide targeted interventions aimed at strengthening readiness for GenAI-enabled transformation in India’s construction supply chains.

Ähnliche Arbeiten