Organisational Readiness for AI in the Front-end Planning of Public Construction Projects

Abstract

The Kingdom of Saudi Arabia (KSA) Vision 2030 initiatives aim to diversify the economy and enhance the public sector. This led to an increase in public projects. However, many projects suffer from underperformance and failure, with these issues frequently arising during Front-end Planning (FEP), which is a crucial initial phase of project definition. Artificial Intelligence (AI) has been identified as having the potential to lower the barrier of carrying out FEP and improve decision- making for better overall project outcomes. The adoption of AI at the FEP stage can significantly improve practices, however, the readiness of public construction organisations to adopt AI remains under-explored. This research established foundational knowledge (exploratory) and test hypotheses (explanatory), defined as collective capability, culture, and governance structure required for AI integration in KSA public construction. The research employed a sequential exploratory mixed-methods approach grounded in the Technology-Organisation-Environment (TOE) Framework and Theory of Planned Behaviour (TPB). The research first conducted 30 semi-structured interviews with key stakeholders (government officials, engineers, and industry experts) to explore their perspectives on FEP and enablers and barriers to adopt AI, while an online survey of 234 professionals validated these insights. Purposive and snowball sampling ensured relevance, while the demographic profile reflected the structure of the Saudi construction workforce, enhancing the representativeness of the sample. Findings revealed a robust model (R²=0.995, p<.001) where organisational absorptive capacity (β =0.261) and organisational maturity (β= 0.235) emerged as key factors for this readiness. Followed by technological readiness (β= 0.216), environmental support (β= 0.143), and senior management support (β= 0.126). reinforced by government support, senior management engagement, and technological readiness. Survey results showed 82.9% identified team competence as the most critical failure factor at FEP. These insights extended the theory by integrating TOE with TPB, showing that structural enablers such as process and resources, must align with behavioural dimensions to achieve readiness. Overall, this research makes a novel theoretical contribution by demonstrating how the intersection of mixed-methods, context specific (KSA), multi-level frameworks (TOE+TPB), specific project phases (FEP), and industry specificity (construction) creates unique adoption dynamics absent from Western- centric models. This research contributes to knowledge by identifying the interconnected role of organisational absorptive capacity and organisational maturity in determining organisational readiness to adopt AI. Theoretically, it extends the TOE framework by integrating individual-level behavioural factors, offering a contextualised perspective and provides the first empirical examination of AI adoption in FEP in the KSA construction industry. This framework provides a contextualised approach to AI adoption tailored to the KSA public construction sector, highlighting the need to reduce bureaucratic rigidity, enhance managerial communication, and promote learning in organisational culture. It also addresses employee concerns related to job security to ensure readiness for successful AI adoption. Future research should explore the integration of dynamic capabilities theory to address aspects of readiness development, which can guide policymakers and industry practitioners in improving organisational readiness for AI.