THE DEVELOPMENT AND APPLICATION OF DYNAMIC MODELS FOR PREDICTING TRANSIT ARRIVAL TIMES
This study describes the development of models for dynamically predicting transit arrival times in urban settings, including a basic model, a Kalman filtering model, artificial neural networks (ANNs) and Neural/Dynamic (ND) models. The proposed prediction models are integrated with the enhanced CORSIM microscopic simulation program individually to predict bus arrival times while simulating the operations of a bus transit route in New Jersey. A reliability analysis of the prediction results shows that ANNs are superior to the basic and Kalman filtering models. The study also explores the application of the proposed prediction models to a real-time headway control model, as well as through simulating a high frequency light rail transit route. Results indicate that with the accurately prediction of vehicle arrival information from the proposed models the regularity of headways between any pair of consecutive operating vehicles is improved, while the average passenger wait times at stops are significantly reduced.
- Publication Date: 2000. UMI Company, Ann Arbor MI. Remarks: Thesis (Ph. D.)--New Jersey Institute of Technology, 2000. Abstract also in: Dissertation abstracts international. B. Vol. 61 no. 1 (July 2000), p. 414. Format: website
New Jersey Institute of Technology, NewarkUniversity Heights
Newark, NJ United States 07102
- Ding, Yuqing
- Publication Date: 2000
- Pagination: 149 p.
- TRT Terms: Kalman filtering; Neural networks; Public transit; Simulation
- Subject Areas: Data and Information Technology; Public Transportation;
- Accession Number: 00962344
- Record Type: Publication
- Source Agency: UC Berkeley Transportation Library
- Report/Paper Numbers: AAT 9959684 (UMI order #)
- Files: PATH
- Created Date: Sep 2 2003 12:00AM