The use of machine learning algorithms in traffic control has garnered a lot of interest recently. In order to better understand the design and development of prediction algorithms for machine learning-based real-time traffic management, In this paper, we summarize the deep learning and machine learning methods for data processing, like image processing, SVM, KNN, CNN, HOG, big data, ANN, regression, Yolov3 algorithms, LSTM, GRU, etc. We are attempting to identify these methods’ limitations and potential future applications. This study helps to choose the best prediction algorithms. Here, we tried to use congestion theory in a real-time traffic application.