CBPM Conference 2026 · Vol. 2026, Issue 1

Leveraging Artificial Intelligence Driven Predictive Analytics for Sustainable Airfield Maintenance: A Project Management Perspective from Air Force Base Waterkloof, South Africa

Ayodeji Olatunji Aiyetan, Dillip Kumar Das, Mbuyiseni Mdakaneu

Keywords: Artificial Intelligence; Airfield Maintenance; Construction Project Management; Infrastructure; Predictive Maintenance.

Abstract

Airfield infrastructure in the defence industry is increasingly hindered by outdated facilities, reactive maintenance strategies, and disconnected data management systems, compromising future operational readiness. This paper analyses the main operational challenges related to airfield maintenance in Air Force Base Waterkloof in South Africa, as well as stakeholder readiness to embrace AI-driven predictive analytics. A quantitative survey-based research design was employed with the selected 316 maintenance actors. Data analysis involved descriptive statistics and significance tests. Findings suggest that existing maintenance activities are carried out using an unconnected hybrid strategy, with a focus on reactive maintenance and frequent emergency repairs. Despite certain structural challenges (such as ageing facilities, disconnected databases, and a lack of technical skills), stakeholders showed a high level of readiness to embrace AI predictive analytics. High levels of receptivity were found for machine learning (predictive maintenance and scheduling), computer vision (automatic defect detection), and intelligent decision-making tools that can facilitate the shift to proactive asset management. The study offers actionable insights for construction and project managers by identifying priority areas for adopting AI-driven predictive analytics in maintenance systems. It highlights the need for improved data integration, workforce upskilling, and strategic alignment to facilitate the transition from reactive to predictive maintenance practices in complex, resource-constrained environments.