Big data driven assessment of probe-sourced data

dc.contributor.advisor Anuj Sharma
dc.contributor.author Ahsani, Vesal
dc.contributor.department Civil, Construction, and Environmental Engineering
dc.date 2019-08-21T10:11:43.000
dc.date.accessioned 2020-06-30T03:14:47Z
dc.date.available 2020-06-30T03:14:47Z
dc.date.copyright Wed May 01 00:00:00 UTC 2019
dc.date.embargo 2001-01-01
dc.date.issued 2019-01-01
dc.description.abstract <p>Presently, there is an expanding interest among transportation agencies and state Departments of Transportation to consider augmenting traffic data collection with probe-based services, such as INRIX. The objective is to decrease the cost of deploying and maintaining sensors and increase the coverage under constrained budgets. This dissertation documents a study evaluating the opportunities and challenges of using INRIX data in Midwest. The objective of this study is threefold: (1) quantitative analysis of probe data characteristics: coverage, speed bias, and congestion detection precision (2) improving probe based congestion performance metrics accuracy by using change point detection, and (3) assessing the impact of game day schedule and opponents on travel patterns and route choice.</p> <p>The first study utilizes real-time and historical traffic data which are collected through two different data sources; INRIX and Wavetronix. The INRIX probe data stream is compared to a benchmarked Wavetronix sensor data source in order to explain some of the challenges and opportunities associated with using wide area probe data. In the following, INRIX performance is thoroughly evaluated in three major criteria: coverage and penetration, speed bias, congestion detection precision.</p> <p>The second study focuses on the number of congested events and congested hour as two important performance measures. To improve the accuracy and reliability of performance measures, this study addresses a big issue in calculating performance measures by comparing Wavetronix against INRIX. We examine the very traditional and common method of congestion detection and congested hour calculation which utilized a fixed-threshold and we show how unreliable and erroneous that method can be. After that, a novel traffic congestion identification method is proposed in this paper and in the following the number of congested events and congested hour are computed as two performance measures.</p> <p>After evaluating the accuracy and reliability of INRIX probe data in chapter 2 and 3, the purpose of the last study in chapter 4 is to assess the impacts of game day on travel pattern and route choice behaviors using INRIX, the accurate and reliable data source. It is shown that the impacts vary depending on the schedule and also the opponents. Also, novel methods are proposed for hotspot detection and prediction.</p> <p>Overall, this dissertation evaluates probe-sourced streaming data from INRIX, to study its characteristics as a data source, challenges and opportunities associated with using wide area probe data, and finally make use of INRIX as a reliable data source for travel behavior analysis.</p>
dc.format.mimetype application/pdf
dc.identifier archive/lib.dr.iastate.edu/etd/16952/
dc.identifier.articleid 7959
dc.identifier.contextkey 14820751
dc.identifier.s3bucket isulib-bepress-aws-west
dc.identifier.submissionpath etd/16952
dc.identifier.uri https://dr.lib.iastate.edu/handle/20.500.12876/31135
dc.language.iso en
dc.source.bitstream archive/lib.dr.iastate.edu/etd/16952/Ahsani_iastate_0097E_17861.pdf|||Fri Jan 14 21:08:29 UTC 2022
dc.subject.disciplines Civil Engineering
dc.subject.disciplines Computer Sciences
dc.subject.disciplines Transportation
dc.title Big data driven assessment of probe-sourced data
dc.type article
dc.type.genre dissertation
dspace.entity.type Publication
thesis.degree.level dissertation
thesis.degree.name Doctor of Philosophy
File
Original bundle
Now showing 1 - 1 of 1
Name:
Ahsani_iastate_0097E_17861.pdf
Size:
2.88 MB
Format:
Adobe Portable Document Format
Description: