Navjot Kaur
Computer Science
March 2025
The rising effects of climate change present considerable challenges to urban planning and development, calling for innovative prediction and mitigation solutions. This research investigated the use of machine learning models—Random Forest, Gradient Boosting, Long Short-Term Memory (LSTM), and Support Vector Machine (SVM)—to predict key urban indicators, such as urban heat island intensity, resource requirements, and flood hazards. Different data sources provided climate and urban information which followed preprocessing before the models utilized these datasets for evaluation through Mean Absolute Error (MAE) combined with Root Mean Squared Error (RMSE) alongside R² scores. The forecasting accuracy of climate-related variables reached its highest point when using the LSTM model. The research identified significant intensifications of urban heat islands alongside increased resource consumption and flood hazards in major cities which demand proper sustainable planning measures by 2050. The incorporation of machine learning-based forecasting into urban planning systems along with development of infrastructure resilience and use of renewable energy and enhanced green cover represents key suggestions
560- 573
© Airo Journals 2021
