Improving Urban Traffic Management with Big Data-Driven Forecasting Models Using ARIMA and Prophet Algorithms

Transport challenges within urban areas are one of the hardest puzzles for traffic authorities and city developers due to the high traffic rates currently experienced. This paper presents a study in the context of big data-based methods to improve the precision of traffic forecasting for urban areas, using ARIMA and Prophet models. Using real traffic data, climate, and live car density, the research intents to create precise models for traffic flow estimation in advance. The ARIMA model, which has proven to be effective in processing time series is used in capturing of linear movements and trends as well as seasonality. In parallel, the Prophet algorithm used by Facebook fines the discrepancies and nonlinearities of the growth pattern, which is highly suitable in urban settings. The comparative assessment of these sorts of models translates into comparing the efficacy of their forecasts, throughput ability, and computational costs. The preliminary results show that adopting both algorithms has better performance for forecasting compared to single algorithms, which benefits the city officials to get solutions to improve the traffic lights, minimize traffic jams, and enhance the timing of public transport anymore. This research has revealed how big data analytics can help in the provision of mobility in urban areas to pave way for smart and efficient city development. Future work will aim at growing new types of data and incorporating in the system GPS and IoT information for a more precise forecast of the functions.