Cyber-Physical Digital Twin for Smart City Transportation Systems: A Real-Time Simulation and Optimization Framework

Advanced traffic management systems are required as a result of the increasing traffic congestion caused by cities' rapid growth. Using a Cyber-Physical Digital Twin (CPDT) architecture, this research aims to improve urban traffic optimization and real-time simulation for smart city transportation systems. To gather data in real-time, the CPDT architecture incorporates a number of data sources, including as GPS devices, traffic cameras, and Internet of Things sensors. Ant Colony Optimization (ACO) and Genetic Algorithms (GA) are used in a unique way to improve route planning and traffic signal timings, respectively. To forecast traffic patterns and events, the framework also makes use of machine learning models, such as CNNs and LSTM networks. The suggested methods greatly increase traffic management metrics and forecast accuracy. The CPDT framework achieves a prediction accuracy of 94.2% by reducing the MAE to 2.89 and the RMSE to 4.67, according to the results. Furthermore, the overall optimization strategy lowers fuel consumption by 24.8%, improves traffic flow by 29.3%, and reduces average delays by 38.2%. These results highlight how well the CPDT framework works to improve urban mobility and support the development of sustainable, productive smart cities.