Abstract:

Bus arrival time prediction is a key factor in passenger satisfaction and bus usage. Bus arrival time information reduces both passenger anxiety and their waiting time at the bus stop. Therefore, giving passengers accurate bus arrival time information is very important in public transportation. Various time series prediction methods are used for bus arrival time in this paper. Moreover, five different performance measurements are considered to assess the accuracy of the prediction models. A case study is presented using real data from Istanbul, Turkey for the proposed models. The models predict bus arrival time on a route for its different segments. The results of the proposed models are compared according to performance measures. The model with the best accuracy result among the eight prediction models can support service operators and the authorities in obtaining better passenger satisfaction.

Keywords: