Multiple Extreme Learning Machines Based Arrival Time Prediction for Public Bus Transport

Due to fast-growing urbanization, the traffic management system becomes a crucial problem owing rapid growth in number of vehicles The research proposes an Intelligent public transportation where information regarding all buses connecting city will be gathered, processed and accurate bus arrival time prediction presented user. Various linear time-varying parameters such as distance, waiting at stops, red signal duration signal, density, turning rush hours, weather conditions, passengers on bus, type day, road type, average vehicle speed limit, current affecting are used for analysis. proposed model exploits feasibility applicability ELM travel forecasting area. Multiple ELMs (MELM) explicitly training dynamic, trajectory approach. A large-scale dataset (historical data) obtained from Kerala State Road Transport Corporation is training. Simulations carried out by using MATLAB R2021a. experiments revealed that efficiency MELM independent day week. It can manage huge volumes data with less human intervention greater learning speeds. found yields accuracy range 96.7% 99.08%. MAE value between 0.28 1.74 minutes study there could regularity usage daily rides predictable better degree accuracy. has proved superior predictions terms error, compared other approaches.