AI-ASSISTED SEISMIC PERFORMANCE ASSESSMENT OF SUSTAINABLE HIGH-RISE STRUCTURES USING SMART MATERIAL TECHNOLOGIES IN PAKISTAN

Abstract

The increasing vulnerability of high-rise structures to seismic hazards, particularly in developing countries such as Pakistan, necessitates advanced, adaptive, and intelligent structural assessment frameworks. This study developed an AI-assisted seismic performance evaluation model for sustainable high-rise structures incorporating smart material technologies, including shape memory alloys (SMA), fiber-reinforced polymers (FRP), and damping systems. A quantitative simulation-based research design was employed, integrating finite element modeling with machine learning algorithms to predict seismic response parameters such as inter-story drift ratio, base shear, damage index, and energy dissipation capacity. Multiple AI models, including ANN, CNN, LSTM, and hybrid ANN–LSTM architectures, were trained and validated using k-fold cross-validation. The results indicated that the hybrid ANN–LSTM model achieved the highest predictive accuracy (R² = 0.96), outperforming conventional machine learning approaches. Furthermore, structures integrated with smart materials exhibited significant improvements in seismic resilience, including reduced inter-story drift (35.7% reduction), lower damage indices (39.0% reduction), and enhanced energy dissipation capacity (28.1% increase). Sustainability performance also improved substantially in smart material-based systems due to enhanced material efficiency and reduced lifecycle impact. The study concludes that the integration of AI-driven predictive analytics with smart material technologies provides a robust and scalable framework for seismic performance assessment of high-rise structures. This integrated approach enhances structural safety, sustainability, and decision-making capabilities, particularly in seismic-prone urban regions of Pakistan