A REVIEW ON AI-POWERED MACHINE LEARNING MODEL FOR EARTHQUAKE RESILIENCE: PREDICTIVE MODELLING AND DESIGN OPTIMIZATION FOR EARTHQUAKE RESISTANT STRUCTURES
Keywords:
Earthquake Resilience, Predictive Modeling, Random Forest, Seismic-Resistant Structures, Data Pre-processing, Feature Engineering, AI in Structural Engineering, Flask API Deployment, Multi-Output Regression, Design Optimization.Abstract
This paper presents an integrated, AI-powered approach for enhancing earthquake resilience through predictive modeling and design optimization of seismic-resistant structures. By leveraging a comprehensive dataset that encompasses seismic characteristics (such as magnitude, depth, and peak ground acceleration) and structural attributes (including building height, material type, and reinforcement level), we develop a multi-output Random Forest model that predicts crucial performance parameters: displacement, stress, and damage level. The framework incorporates rigorous data preprocessing, advanced feature engineering, and iterative model training and evaluation. Additionally, the system is designed for deployment within a lightweight Flask-based API, bridging the gap between research and practical, real-world applications. This methodology not only advances structural safety by providing accurate, data-driven insights, but also establishes a versatile platform for integrating real-time sensor data and adaptive response mechanisms in areas vulnerable to seismic events.
