Return to Projects List

[NTC2016-MU-R-01] Urban Transportation System Analytics and Optimization: A Sensor Data-Driven Approach

P.I.: 
George List, Xuesong Zhou
North Carolina State University, Arizona State University
Year: 
2016
Website: 
http://ntc.umd.edu/node/166
Subject Area: 
Collaborative Research


Description: 

Sensor data, from vehicles and facilities, is revolutionizing the manner in which urban transportation systems can operate. Pre-trip route choices can be informed by network status, en-route path choices can be made, prices can influence path choice decisions, and more robust network operating conditions can be obtained. Zhou and List (2010) have demonstrated that careful sensor placement can enhance the estimation of Origin-Destination (OD) matrices. Ma, Smith, and Zhou (2015) have shown that an agent-based approach can enhance individual user real-time decision-making. Lei, List and Taylor (2015) have illustrated how probe data can be used to characterize corridor-level travel time distributions. Mahmoudi and Zhou (2015) have shown that real-time data can improve freight- based vehicle routing choices. Chen, Zhou, and List (2011) have demonstrated that time-varying tolls can be used to improve truck arrival patterns at port facilities. Cetin, List and Zhou (2005) have explored the number of probes required to develop credible network travel time estimates. And Eisenman and List (2004) explored the ways in which probe data can be used to enhance trip matrix estimation.

Building on these and other efforts, this project will explore new and creative ways to use sensor data to 1) enhance freight-related path choice decision making, both pre-trip and en-route, and 2) to improve network performance from a freight perspective. While a significant body of literature has begun to emerge on general path choice decision making and network performance management, the authors believe that they have new and creative ways to address freight-focused methodologies predicated on statistical analysis of historical and real-time data. These ideas will lead to better freight-focused routing decisions and network operating conditions whose performance for freight can be improved and statistically can be assessed. Moreover, the authors have a wealth of probe data from many locations that can be used to develop and inform the network performance assessment process and test the validity of the methods developed.