Estimating historical hourly traffic volumes via machine learning and vehicle probe data: A Maryland case study. Sekuła, P., Marković, N., Vander Laan, Z., & Sadabadi, K. F. Transportation Research Part C: Emerging Technologies, 97:147–158, December, 2018. ![Estimating historical hourly traffic volumes via machine learning and vehicle probe data: A Maryland case study [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex This paper focuses on the problem of estimating historical traffic volumes between sparsely-located traffic sensors, which transportation agencies need to accurately compute statewide performance measures. To this end, the paper examines applications of vehicle probe data, automatic traffic recorder counts, and neural network models to estimate hourly volumes in the Maryland highway network, and proposes a novel approach that combines neural networks with an existing profiling method. On average, the proposed approach yields 24% more accurate estimates than volume profiles, which are currently used by transportation agencies across the US to compute statewide performance measures. The paper also quantifies the value of using vehicle probe data in estimating hourly traffic volumes, which provides important managerial insights to transportation agencies interested in acquiring this type of data. For example, results show that volumes can be estimated with a mean absolute percent error of about 21% at locations where average number of observed probes is between 30 and 47 vehicles/h, which provides a useful guideline for assessing the value of probe vehicle data from different vendors.
@article{sekula_estimating_2018,
title = {Estimating historical hourly traffic volumes via machine learning and vehicle probe data: {A} {Maryland} case study},
volume = {97},
issn = {0968-090X},
shorttitle = {Estimating historical hourly traffic volumes via machine learning and vehicle probe data},
url = {http://www.sciencedirect.com/science/article/pii/S0968090X18314773},
doi = {10.1016/j.trc.2018.10.012},
abstract = {This paper focuses on the problem of estimating historical traffic volumes between sparsely-located traffic sensors, which transportation agencies need to accurately compute statewide performance measures. To this end, the paper examines applications of vehicle probe data, automatic traffic recorder counts, and neural network models to estimate hourly volumes in the Maryland highway network, and proposes a novel approach that combines neural networks with an existing profiling method. On average, the proposed approach yields 24\% more accurate estimates than volume profiles, which are currently used by transportation agencies across the US to compute statewide performance measures. The paper also quantifies the value of using vehicle probe data in estimating hourly traffic volumes, which provides important managerial insights to transportation agencies interested in acquiring this type of data. For example, results show that volumes can be estimated with a mean absolute percent error of about 21\% at locations where average number of observed probes is between 30 and 47 vehicles/h, which provides a useful guideline for assessing the value of probe vehicle data from different vendors.},
urldate = {2018-12-12},
journal = {Transportation Research Part C: Emerging Technologies},
author = {Sekuła, Przemysław and Marković, Nikola and Vander Laan, Zachary and Sadabadi, Kaveh Farokhi},
month = dec,
year = {2018},
keywords = {Multi-layered neural networks, Regression, Traffic volume estimation, Vehicle probe data},
pages = {147--158}
}
Downloads: 0
{"_id":"vvKamdu6ou2tZJvaq","bibbaseid":"sekua-markovi-vanderlaan-sadabadi-estimatinghistoricalhourlytrafficvolumesviamachinelearningandvehicleprobedataamarylandcasestudy-2018","authorIDs":[],"author_short":["Sekuła, P.","Marković, N.","Vander Laan, Z.","Sadabadi, K. F."],"bibdata":{"bibtype":"article","type":"article","title":"Estimating historical hourly traffic volumes via machine learning and vehicle probe data: A Maryland case study","volume":"97","issn":"0968-090X","shorttitle":"Estimating historical hourly traffic volumes via machine learning and vehicle probe data","url":"http://www.sciencedirect.com/science/article/pii/S0968090X18314773","doi":"10.1016/j.trc.2018.10.012","abstract":"This paper focuses on the problem of estimating historical traffic volumes between sparsely-located traffic sensors, which transportation agencies need to accurately compute statewide performance measures. To this end, the paper examines applications of vehicle probe data, automatic traffic recorder counts, and neural network models to estimate hourly volumes in the Maryland highway network, and proposes a novel approach that combines neural networks with an existing profiling method. On average, the proposed approach yields 24% more accurate estimates than volume profiles, which are currently used by transportation agencies across the US to compute statewide performance measures. The paper also quantifies the value of using vehicle probe data in estimating hourly traffic volumes, which provides important managerial insights to transportation agencies interested in acquiring this type of data. For example, results show that volumes can be estimated with a mean absolute percent error of about 21% at locations where average number of observed probes is between 30 and 47 vehicles/h, which provides a useful guideline for assessing the value of probe vehicle data from different vendors.","urldate":"2018-12-12","journal":"Transportation Research Part C: Emerging Technologies","author":[{"propositions":[],"lastnames":["Sekuła"],"firstnames":["Przemysław"],"suffixes":[]},{"propositions":[],"lastnames":["Marković"],"firstnames":["Nikola"],"suffixes":[]},{"propositions":[],"lastnames":["Vander","Laan"],"firstnames":["Zachary"],"suffixes":[]},{"propositions":[],"lastnames":["Sadabadi"],"firstnames":["Kaveh","Farokhi"],"suffixes":[]}],"month":"December","year":"2018","keywords":"Multi-layered neural networks, Regression, Traffic volume estimation, Vehicle probe data","pages":"147–158","bibtex":"@article{sekula_estimating_2018,\n\ttitle = {Estimating historical hourly traffic volumes via machine learning and vehicle probe data: {A} {Maryland} case study},\n\tvolume = {97},\n\tissn = {0968-090X},\n\tshorttitle = {Estimating historical hourly traffic volumes via machine learning and vehicle probe data},\n\turl = {http://www.sciencedirect.com/science/article/pii/S0968090X18314773},\n\tdoi = {10.1016/j.trc.2018.10.012},\n\tabstract = {This paper focuses on the problem of estimating historical traffic volumes between sparsely-located traffic sensors, which transportation agencies need to accurately compute statewide performance measures. To this end, the paper examines applications of vehicle probe data, automatic traffic recorder counts, and neural network models to estimate hourly volumes in the Maryland highway network, and proposes a novel approach that combines neural networks with an existing profiling method. On average, the proposed approach yields 24\\% more accurate estimates than volume profiles, which are currently used by transportation agencies across the US to compute statewide performance measures. The paper also quantifies the value of using vehicle probe data in estimating hourly traffic volumes, which provides important managerial insights to transportation agencies interested in acquiring this type of data. For example, results show that volumes can be estimated with a mean absolute percent error of about 21\\% at locations where average number of observed probes is between 30 and 47 vehicles/h, which provides a useful guideline for assessing the value of probe vehicle data from different vendors.},\n\turldate = {2018-12-12},\n\tjournal = {Transportation Research Part C: Emerging Technologies},\n\tauthor = {Sekuła, Przemysław and Marković, Nikola and Vander Laan, Zachary and Sadabadi, Kaveh Farokhi},\n\tmonth = dec,\n\tyear = {2018},\n\tkeywords = {Multi-layered neural networks, Regression, Traffic volume estimation, Vehicle probe data},\n\tpages = {147--158}\n}\n\n","author_short":["Sekuła, P.","Marković, N.","Vander Laan, Z.","Sadabadi, K. F."],"key":"sekula_estimating_2018","id":"sekula_estimating_2018","bibbaseid":"sekua-markovi-vanderlaan-sadabadi-estimatinghistoricalhourlytrafficvolumesviamachinelearningandvehicleprobedataamarylandcasestudy-2018","role":"author","urls":{"Paper":"http://www.sciencedirect.com/science/article/pii/S0968090X18314773"},"keyword":["Multi-layered neural networks","Regression","Traffic volume estimation","Vehicle probe data"],"downloads":0},"bibtype":"article","biburl":"https://api.zotero.org/users/4548732/collections/9BA2I5LW/items?key=GSzhzZJMwfFi2xy72gVJaMBi&format=bibtex&limit=100","creationDate":"2019-05-29T01:52:43.991Z","downloads":0,"keywords":["multi-layered neural networks","regression","traffic volume estimation","vehicle probe data"],"search_terms":["estimating","historical","hourly","traffic","volumes","via","machine","learning","vehicle","probe","data","maryland","case","study","sekuła","marković","vander laan","sadabadi"],"title":"Estimating historical hourly traffic volumes via machine learning and vehicle probe data: A Maryland case study","year":2018,"dataSources":["7quxtgr6iDS68E7jv"]}