SMART URBAN PARKING OPTIMIZATION USING INTEGRATED MACHINE LEARNING MODELS

Abstract

Urban areas across the globe face growing challenges in parking management due to rising vehicle density, limited space availability, and inefficient manual systems. This research presents a smart urban parking optimization framework that leverages integrated machine learning models to predict parking availability, enhance utilization, and reduce congestion. The proposed system combines realtime sensor data, historical occupancy records, and environmental parameters to build predictive models using ensemble learning techniques such as Random Forest, Gradient Boosting, and LSTM networks. Experimental results from real-world parking datasets demonstrate that the integrated model significantly improves prediction accuracy and parking allocation efficiency compared to standalone models. The framework provides actionable insights for city planners and can be deployed as part of intelligent transportation systems to promote sustainable urban mobility.