SWUTC Ph.D. Candidate Assistantship Project Description
Link Travel Time Estimation Based on Network Entry/Exit Time Stamps of Trips
University: Texas A&M University
Principal Investigator:
Wen Wang
Zachry Department of Civil Engineering
Faculty Supervisor:
Bruce X. Wang
Zachry Department of Civil Engineering
(979) 845-9901
Funding Source: USDOT Funds
Total Project Cost: $36,279
Project Number: 600451-00033
Date Started: 9/1/13
Estimated Completion Date: 12/31/14
Project Summary
Project Abstract:
Travel time for a road trip is the most important factor when the driver plans a route from an origin to a destination. Network link travel time is also essential to traffic operations and control as well as planning. Accurate travel time estimation on a transportation network can provide supportive information for decision making in both private sectors and public agencies. Prior studies on travel time estimation mostly focus on a line structure network, such as arterial roadway with signalized intersections, or a highway corridor comprising of segments with varying traffic conditions. This research proposes to extend the existing travel time estimation techniques to a general network, and develop methodologies to estimate the link travel time based on network connectivity and trips’ entry/exit locations and time stamps.
Data is often abundantly available such as the metro riders’ check in/out information every day. The wide use of blue tooth and Automated Vehicle Identification (AVI) technologies has also created a situation in which vehicles’ entry and exit information on a network is known. How to efficiently utilize this abundant information is a meaningful question. Furthermore, efficient utilization of this data may give rise to methodologies about how to most economically obtain necessary data.
The entry/exit time stamp of each trip tells the time duration of the entire itinerary of a trip, making this problem different from most literatures. We propose to develop methods to utilize this entry/exit information to infer the link travel times on the entire network. Note that the travel time estimation methods may also be used to assess the performance of networks of other modes such as subway or bus systems with passengers’ entry and exit information tracked at check in/out points. Major freight corridors where trucks are tracked with GPS periodically with a certain time interval (e.g. 15 minutes) may also be considered a special case of a network where itineraries overlap with each other on the road sections.
Project Objectives:
This research will accomplish the following objectives:
- To estimate the link performance on a network with known itinerary travel times for vehicles. Prior research on link travel time estimation along a line structure of major arterial corridor has suggested estimation approaches substantially dependent on data collection techniques. This research will relax the dependence on specific collection techniques, but develop methods that extend the link performance estimation to a general transportation network, given vehicles’itinerary travel times tracked on the network.
- To apply the link travel time estimation methods to different networks, and numerically validate using simulation on representative network structures. Various network structures will be tested in simulation framework.
- To develop a sound model to incorporate vehicles’ route choices into the formulation of link travel time estimation. The existence and uniqueness of solution will also be examined in this research.
Task Descriptions:
Task 1– Review literature on travel time estimation, vehicle route choices and assignment techniques
Task 2– Develop analytical methods for link travel time estimation with known routes and known entry/exit information
Task 3– Develop a model for link travel time estimation with only known entry/exit information but with no known routes
Task 4– Application networks
Task 5 – Build a simulation framework and conduct numerical tests
Task 6 – Document the dissertation and final report