Traffic Density Estimation with IoT Edge Computing In Smart City: A Case Study

Authors : Hande Okutucu, Şükrü Mustafa Kaya

Pages : 189-215

View : 41 | Download : 97

Publication Date : 2025-12-12

Article Type : Research Paper

Abstract :In the digitalizing world, smart city management and applications have become an integral part of our lives in recent years. With the increase in innovative sensor-based devices, concepts such as smart environment, energy, transportation, healthcare, and traffic have emerged within smart cities, improving the quality of life for citizens through smart city management. This study focuses on the concept of smart transportation and traffic management, which is a subcategory of smart cities. The concept of traffic management in smart cities has advanced thanks to the integration of IoT (Internet of Things) and Edge Computing technologies. This system provides more realistic traffic density predictions. IoT devices, traffic sensors, cameras, and GPS-enabled devices collect real-time data such as traffic density, vehicle speeds, and road conditions in smart cities. In our study, we aim to predict hourly vehicle density for a specific day. A three-year dataset was utilized, consisting of 24-hour vehicle density data for each day of the year. Using time series algorithms, hourly vehicle density predictions were made for a future date. Algorithms such as ANN (Artificial Neural Network), KNN (K-Nearest Neighbors), LSTM (Long Short-Term Memory), Random Forest, Prophet, and XGBoost (Extreme Gradient Boosting) were employed for predictions. The error rates of the algorithms were analyzed to identify the most accurate prediction method. The vehicle density prediction data produced by this algorithm was considered the closest to reality. The results were discussed and evaluated in the final section of the article.
Keywords : Nesnelerin İnterneti, Uç Bilişim, Akıllı Şehirler, Trafik Yoğunluğu, Makine Öğrenmesi

ORIGINAL ARTICLE URL