Abstract
Effective prediction of bus arrival times is important to advanced traveler information systems. Here enhanced regression models, based on linear regression and an adaptive algorithm, are presented to predict the arrival times at stops. In the models, the linear regression models predict the baseline travel times between adjacent stops based on traffic conditions of the following segment; the adaptive algorithm uses the most recent bus arrival information, together with the estimated baseline arrival times from linear regression models, to predict arrival times at downstream stops. The adaptive method for bus arrival time prediction is examined with the data of bus No. 23 in Dalian City, China and at last some conclusions are given.
