Large-scale and Long-term Forecasting of Performance Measurement of Public Transportation Systems

Accurate forecasting of public transportation metrics is critical towards the high reliability and efficiency of the public transportation system. However, deploying a forecasting system to serve city-level public transportation with long-term forecasting is challenging. In this project, the authors develop the capability for processing the entire Los Angeles Metropolitan Area (LAMA) for long-term forecasting of a variety of public transportation system performance metrics. First, the authors explore both spatial statistical methods and machine learning methods to estimate traffic flows for the road segments that do not have traffic sensors. Second, the authors develop methods to enable traffic forecasting with a deep learning model designed for small networks for the entire LAMA road network. The authors also study various training strategies (e.g., teacher forcing) to enable accurate long-term forecasting of traffic flows and bus arrival times. Lastly, the authors develop an end-to-end deep learning approach that combines the estimation and forecasting of traffic flow with data imputation methods for estimating bus arrival time for each stop in individual bus routes in LAMA. Using the real-world data in the University of Southern California Archived Transportation Data Management System (ADMS), the authors show that the proposed approach and system are capable of predicting bus arrival times with a city-level spatial coverage and a route-level temporal forecasting horizon. The authors also demonstrate the overall result of the bus arrival time estimation in a web dashboard. This dashboard enables users at all levels of technical skills to benefit from the developed machine learning approach and access to valuable information for trip planning, vehicle management, and policymaking.

• Record URL:
  https://www.metrans.org/assets/research/psr-20-21_shahabi_final-report.pdf
• 
• Summary URL:
  https://rosap.ntl.bts.gov/view/dot/67532
• Record URL:
  https://rosap.ntl.bts.gov/view/dot/58494
• Supplemental Notes:
  • This document was sponsored by the U.S. Department of Transportation, University Transportation Centers Program.
• Corporate Authors:

  METRANS Transportation Center

  University of Southern California
  Los Angeles, CA  United States  90089-0626

  Pacific Southwest Region University Transportation Center

  University of Southern California
  Los Angeles, CA  United States  90089

  California Department of Transportation

  Sacramento, CA  United States  95819

  Office of the Assistant Secretary for Research and Technology

  University Transportation Centers Program
  Department of Transportation
  Washington, DC  United States  20590
• Authors:
  • Tran, Luan
  • Chiang, Yao-Yi
  • 0000-0002-8923-0130
  • Shahabi, Cyrus
  • 0000-0001-9118-0681
• Publication Date: 2021-7-26

Language

• English

Media Info

• Media Type: Digital/other
• Edition: Final Report
• Features: Figures; References; Tables;
• Pagination: 35p

Subject/Index Terms

• TRT Terms: Arrivals and departures; Bus routes; Machine learning; Performance measurement; Public transit; Traffic forecasting
• Geographic Terms: Los Angeles Metropolitan Area
• Subject Areas: Operations and Traffic Management; Planning and Forecasting; Public Transportation;

Filing Info

• Accession Number: 01785223
• Record Type: Publication
• Report/Paper Numbers: PSR-20-21 TO-038
• Contract Numbers: USDOT Grant 69A3551747109; Caltrans 65A0674 TO 038
• Files: UTC, NTL, TRIS, ATRI, USDOT, STATEDOT
• Created Date: Oct 22 2021 9:18AM