Traffic State Prediction: A Traveler Equity and Multi-model Perspective
Traffic congestion has become one of the modern life problems in many urban areas. Urban traffic congestion has the potential to be relieved by developing tools that plan multi-mode trips to encourage more people to ride public transportation and provide better alternatives to less affluent citizens. Traffic state prediction is the key component to plan multi-mode trips in complex transportation network. This research attempts to address transportation system state prediction problems considering private vehicle, transit, and bike share services within the context of a multimodal transportation system. For public transit service, the proposed effort focuses on developing real-time passenger demand prediction models using multiple data sources to enhance prediction accuracy. For bike share service, the proposed effort focuses on developing prediction models for the number of bikes and travel times of bikes. Finally, for the automobile effort this research will develop a comprehensive traffic prediction tool by including different categories of prediction models. The proposed prediction algorithms and tools will be evaluated by the field data collected in multimodal transportation system by comparing with existing methods.
Universities and Sponsoring Organizations Involved
Hesham A. Rakha (VT) Email: email@example.com
Funding Sources and Amounts (Split By Organization and Type of Funding) Format
June 1, 2017
Expected Completion Date
May 31, 2019
Expected Research Outcomes
The proposed research will develop prediction models for passenger demand by incorporating the traditional APC data and the newly developed user data from mobile APP to improve prediction accuracy. The proposed research will also develop state-of-art prediction models for bike share system based on the latest popular machine learning algorithms.
Expected Equity Impacts and Benefits of Implementation
The proposed research effort will improve traveler experience, public transit operations and bike share services by operating more efficiently and reliably, which is beneficial for the overall operation of multimodal transportation systems. The accurate prediction for passenger demand is helpful for public transit agencies to minimize operational costs and improve bus service quality by properly allocating limited resources. The proposed work of developing prediction models for bike share system will be very useful for practitioners to design and manage bike share system.
Multimodal Transportation System, Transit Passenger Demand Prediction, Bike Share System, Travel Time Prediction.