TY - JOUR
T1 - An empirical validation of network learning with taxi GPS data from Wuhan, China
AU - Xu, Susan Jia
AU - Xie, Qian
AU - Chow, Joseph Y.J.
AU - Liu, Xintao
N1 - Funding Information:
This study was conducted with support from the National Science Foundation Faculty Early Career Development Program award, CMMI-1652735. Xintao Liu was supported by the Hong Kong Polytechnic University Start-Up Research Fund Program under Grant 1-ZE6P and the Area of Excellence under Grant 1-ZE24.
Publisher Copyright:
© 2009-2012 IEEE.
PY - 2021/3/1
Y1 - 2021/3/1
N2 - In prior research, a statistically cheap method was developed to monitor transportation network performance by using only a few groups of agents without having to forecast the population flows. The current study validates this multiagent inverse optimization (MAIO) method using taxi GPS trajectory data from the city of Wuhan, China. Using a controlled 2,062-link network environment and different GPS data processing algorithms, an online monitoring environment was simulated using real data over a 4-h period. Results show that using samples from only one origin-destination (OD) pair, the MAIO method can learn network parameters such that forecasted travel times have a 0.23 correlation with the observed travel times. By increasing the monitoring from just two OD pairs, the correlation improved further, to 0.56.
AB - In prior research, a statistically cheap method was developed to monitor transportation network performance by using only a few groups of agents without having to forecast the population flows. The current study validates this multiagent inverse optimization (MAIO) method using taxi GPS trajectory data from the city of Wuhan, China. Using a controlled 2,062-link network environment and different GPS data processing algorithms, an online monitoring environment was simulated using real data over a 4-h period. Results show that using samples from only one origin-destination (OD) pair, the MAIO method can learn network parameters such that forecasted travel times have a 0.23 correlation with the observed travel times. By increasing the monitoring from just two OD pairs, the correlation improved further, to 0.56.
UR - http://www.scopus.com/inward/record.url?scp=85097375818&partnerID=8YFLogxK
U2 - 10.1109/MITS.2020.3037324
DO - 10.1109/MITS.2020.3037324
M3 - Journal article
AN - SCOPUS:85097375818
SN - 1939-1390
VL - 13
SP - 42
EP - 58
JO - IEEE Intelligent Transportation Systems Magazine
JF - IEEE Intelligent Transportation Systems Magazine
IS - 1
M1 - 9277883
ER -
