Assessing the Suitability of Different Machine Learning Approaches for Smart Traffic Mobility

One of the most critical traffic management issues is congestion in modern and big smart cities. The first task is to accurately forecast traffic patterns to reduce congestion and accidents due to rapid economic development and rising number of vehicles. It is essential for Intelligent Transportation Systems to accurately anticipate future traffic circumstances (such as traffic flow, speed, and traffic time) so that administrators may take proper preventative actions against congestion and travelers can make better-informed judgments. Better trip planning, more efficient traffic operations, lower carbon emissions, and less congestion are all possible outcomes of this forecast. This paper explores different deep-learning time-series forecasting methods such as LSTM, BiLSTM, Prophet, and Transformer models for making short-term predictions regarding traffic flows to ensure smart mobility. The next step is to analyze traffic patterns to provide convenient transportation mobility. Then, we evaluated several performance matrices and computational loads of the proposed methods in this paper,