Action-first AI for facilities management: prioritising post-flood asset repairs to restore access to essential services

Abstract

Flood recovery in developing countries is often hindered by fragmented information and ad hoc repair prioritisation, delaying restoration of essential services. This study proposes an Action-First AI approach for post-flood facilities management that prioritises asset repairs based on service criticality, population impact, accessibility, and operational feasibility. Using a design science research approach, a conceptual framework was developed from the literature and evaluated through qualitative expert review involving disaster management, facilities management, and digital systems professionals. Expert insights were then used to derive a conceptual flow model and a prototype decision-support dashboard demonstrating how prioritisation logic could be integrated within existing FM systems. Rather than presenting a technical AI implementation, the study provides a conceptual blueprint for embedding AI-inspired decision support into FM workflows to enhance transparency, coordination, and effectiveness of post-flood recovery planning.