摘要
This study proposes an approach of leveraging information gathered from multiple traffic data sources at different resolutions to obtain approximate inference on the traffic distribution of Chicago's O'Hare Airport area.Specifically,it proposes the ingestion of traffic datasets at different resolutions to build spatiotemporal models for predicting the distribution of traffic volume on the road network.Due to its good adaptability and flexibility for spatiotemporal data,the Gaussian process(GP)regression was employed to provide short-term forecasts using data collected by loop detectors(sensors)and supplemented by telematics data.The GP regression is used to make predictions of the distribution of the proportion of sensor data traffic volume represented by the telematics data for each location of the sensors.Consequently,the fitted GP model can be used to determine the approximate traffic distribution for a testing location outside of the training points.Policymakers in the transportation sector can find the results of this work helpful for making informed decisions relating to current and future transportation conditions in the area.
基金
supported by the U.S.Department of Energy,Office of Vehicle Technologies,under contract DE-AC02-06CH11357。
作者简介
Damola M.Akinlana Dr.had her PhD in mathematics(concentration in statistics)from the University of South Florida.During her PhD program,she had internship for two summers at Argonne National Laboratory where she worked on projects that involved spatio-temporal modelling of traffic data for the prediction of traffic conditions.Damola's research interests has been in design of experiments,statistical process control,spatial statistics,change point detection and machine learning.E-mail addresses:oyindamolaakinlana@gmail.com;Corresponding author:Arindam Fadikar Dr.is an assistant computational statistician in the Decision and Infrastructure Sciences Division at Argonne National Laboratory.His primary research interest is in the area of design and analysis,calibration,uncertainty quantification of computer models under input dependent noise with application in epidemiology,material science,urban traffic network systems and cosmology.E-mail addresses:afadikar@anl.gov;Stefan M.Wild Dr.is senior scientist and director of the Applied Mathematics and Computational Research Division at Lawrence Berkeley National Laboratory,and an adjunct faculty member in Industrial Engineering and Management Sciences and a senior fellow in NAISE at Northwestern University.Wild's primary research interests focus on developing model-based algorithms and software for challenging numerical optimization problems and automated learning.E-mail addresses:wild@lbl.gov;Natalia Zuniga-Garcia Dr.is a computational transportation engineer in the Vehicle and Mobility Systems team at Argonne National Laboratory.In 2020,she earned her PhD in civil engineering with a focus on transportation systems from The University of Texas at Austin,where she also obtained an M.Sc.in statistics and data sciences in 2018.In her research work,she uses statistics and econometric models,AI/machine learning&stochastic methods,and simulation tools to understand the mobility and energy impact of new transportation technologies.E-mail addresses:nzuniga@anl.gov;Joshua Auld Dr.is the manager of the Transportation Systems and Mobility Group in the Vehicle and Mobility Simulations Department at the Argonne National Laboratory.He is an expert in agent-based modeling,behavioral analysis,transportation simulation and travel data collection and is the lead designer of the POLARIS transportation simulation system.Dr.Auld received his doctorate from the University of Illinois at Chicago in 2011 in Civil and Materials Engineering.E-mail addresses:jauld@anl.gov。
