Prediction of Bus Travel Time Using Artificial Neural Network

The objective of this study is to apply artificial neural network (ANN) for development of bus travel time prediction model. The bus travel time prediction model was developed to give real time bus arrival information to the passenger and transit agencies for applying proactive strategies. For development of ANN model, dwell time, delays and distance between the bus stops was taken as input data. Arrivals/departure times, delays, average speed between the bus stop and distance between the bus stops were collected for two urban routes in Delhi. Model was developed, validated and tested using GPS (Global Positioning System) data collected from field study. Comparative study reveals that ANN model outperformed the regression model in terms of accuracy and robustness.