An SHM-Integrated and AI-Driven Framework for Multi-Hazard Risk Prediction and Structural Resilience in Critical Aviation Infrastructure: A Case Study of Dubai International Airport

This study presents an AI-driven and SHM-integrated framework for multi-hazard risk prediction and structural resilience assessment in critical aviation infrastructure, with Dubai International Airport (DXB) as the case study. Located in a high-exposure zone subject to dust storms, marine winds, and extreme heat, DXB was selected for its operational sensitivity and regional significance. The proposed framework fuses real-time structural health monitoring (SHM) signals—including vibration, strain, displacement, and temperature—with satellite-derived hazard indicators such as aerosol optical depth (AOD), normalized difference vegetation index (NDVI), and land surface temperature (LST), complemented by ۱۵ years of climatic data from the UAE National Center of Meteorology. Two machine learning models were deployed: a Random Forest (RF) classifier achieved ۹۲.۳% accuracy and ۰.۸۸ F۱-score in categorizing high-, medium-, and low-risk zones; and a Long Short-Term Memory (LSTM) network forecasted compound hazard events with MAE = ۰.۰۶۱ and RMSE = ۰.۰۸۴. Geospatial analysis using GIS identified three high-risk zones—Terminal ۲, the Southern Maintenance Area, and the Western Apron—with cumulative risk indices exceeding ۰.۹۰. SHM simulation data further indicated a ۱۷% increase in axial strain during thermal peaks and ۱۲ mm of uplift in runway asphalt layers under combined dust-heat exposure. The results highlight the framework's capability to accurately forecast environmental hazards and quantify their structural impact in real-time, enabling proactive infrastructure management. This scalable approach offers a replicable model for climate-resilient airport systems across other high-risk regions.

کلیدواژه ها:

Dubai International Airport (DXB) ، Structural Health Monitoring (SHM) ، Multi-Hazard Risk Prediction ، Machine Learning (Random Forest ، LSTM) ، Remote Sensing and GIS ، Climate Resilience ، Aviation Infrastructure ، Dust Storms and Extreme Heat ، Smart Infrastructure ، Critical Facilities under Environmental Stress.