Prediction of Road Traffic Congestion using Deep Neural Network Technique
Keywords:
Prediction, Traffic Congestion, Deep Neural Network, Deep Learning, TechniqueAbstract
Traffic congestion remains one of the major challenges in many parts of the world, negatively impacting urban communities in various ways. These effects include the loss of valuable time, delayed deliveries, financial losses, fuel wastage, vehicle wear and tear, and increased stress levels. The current practice of manually controlling traffic by traffic wardens or managers lacks analytical capabilities. This study focuses on developing predictive model for road traffic congestion to enhance traffic management in Lagos State. The historical dataset was collected for a period of six months (February-July, 2024) on Oshodi-Badagry expressway. The dataset consists of 4600 instances with carefully selected 13 attributes which were splitted into 80% 20%; 60% 40% for training and for testing. Deep Neural Network (DNN) was employed to formulate the predictive model. The formulation and simulation of the predictive models were carried out using Python Google Colab environment. The performance of model was assessed through validating metrics such as accuracy, F1 score, precision, specificity and recall. The results showed Accuracy (%) 98, Precision of 0.98, Recall of 0.98, F1-Score of 0.98, and Specificity of 0.97. The model is recommended to support decision-making in the transportation sector so as to reduce traffic congestion in metropolitan cities.
