AI-Based approaches for post-earthquake damage assessment

Abstract

Rapid and reliable post-earthquake damage assessment is essential for effective disaster response, recovery planning, and resilient reconstruction. Recent advances in artificial intelligence (AI) and machine learning have enabled automated damage detection and classification through the integration of diverse data sources, including ground-level photographs, UAV imagery, satellite remote sensing data, and structural response measurements. Despite these advances, significant challenges remain for practical implementation, including limited interpretability, reliability concerns, data imbalance, and insufficient integration with decision-making and project management processes.

This study systematically reviews and classifies AI-based approaches to post-earthquake damage assessment from both engineering and project management perspectives, with particular emphasis on the role of explainable artificial intelligence. The paper highlights the importance of transparency, reliability, and decision-support integration, and provides a theoretical foundation for the development of next-generation AI-driven post-earthquake damage assessment and decision-support systems.