Bus arrival time prediction for dynamic operations control and passenger information systems.

摘要

Automatic Vehicle Location (AVL) and Automatic Passengers Counters (APC) systems have been increasingly implemented by transit agencies for the real time monitoring of transit vehicles and automatic counting of passengers boarding and alighting at bus stops. As a result, a vast amount of potentially online data related to transit operation could be obtained from these systems.; Four different models were developed based on Artificial Neural Networks, Kalman Filter, Regression and Historical Average techniques using empirical data from AVL and APC systems obtained for a specific bus route in Downtown Toronto. The performance of these models was tested using data representing different scenarios of bus operation using the “VISSIM” microsimulation software package.; Finally, Kalman Filter model was used to develop an automated dynamic bus arrival information system. This system has the ability to evaluate real-time bus performance and implement proactive control strategies to prevent any expected disturbance at transit operations.