{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/115481"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/115481","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Methodology to estimate crash frequency at highway-rail grade crossings in a connected vehicle environment","abstract":"Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2022-11-11 without embargo terms","abstract_html":"Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2022-11-11 without embargo terms","abstract_has_math":false,"creators":["Mathew, Jacob"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"Ph.D.","degree_level":"Dissertation","degree_discipline":"Civil Engineering","degree_department":null,"school":null,"contributors":["Benekohal, Rahim F.","Barkan, Christopher P.L.","Ouyang, Yanfeng","Liang, Feng"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2022,"date_issued":"2022-05","date_published":"2022-05","updated_at":"2026-07-22T22:24:54Z","subjects":["Accident Prediction","Highway-Rail Grade Crossing","Connected Vehicle Environment","Zero-Inflated Negative Binomial","Empirical Bayes","Machine Learning"],"languages":["en","eng"],"rights":["Copyright 2022 Jacob Mathew"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/2142/115481","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Benekohal, Rahim F.","Barkan, Christopher P.L.","Ouyang, Yanfeng","Liang, Feng"]},{"key":"dc:creator","label":"Author","values":["Mathew, Jacob"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2022-05","2022-04-19"]},{"key":"dc:type","label":"Dc Type","values":["text","Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Civil Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Ph.D."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois at Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Accident Prediction","Highway-Rail Grade Crossing","Connected Vehicle Environment","Zero-Inflated Negative Binomial","Empirical Bayes","Machine Learning"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en","eng"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2022 Jacob Mathew"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://hdl.handle.net/2142/115481"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2022-11-11 without embargo terms","The student, Jacob Mathew, accepted the attached license on 2022-04-18 at 12:15.","The student, Jacob Mathew, submitted this Dissertation for approval on 2022-04-18 at 18:45.","This Dissertation was approved for publication on 2022-04-19 at 13:44.","DSpace SAF Submission Ingestion Package generated from Vireo submission #17749 on 2022-11-11 at 13:15:33","As the transportation industry is heading towards the Internet of Things, new safety technologies are designed to increase situational awareness of the vehicles. These technologies can reduce or eliminate crashes and can improve the safety of transportation systems. While there have been studies to estimate the potential benefits of Connected and Autonomous Vehicles (CAV) safety applications, the quantification of safety benefits at a highway-rail grade crossing in a connected vehicle environment has not been done before. This dissertation presents a methodology to estimate the expected number of crashes at a highway-rail grade crossing in a connected vehicle environment. For over four decades, the state of practice to estimate accident counts at a railroad grade crossing was to use the USDOT accident prediction formula. Recently the Federal Railroad Administration released a new model for accident prediction at railroad grade crossings using a Zero-Inflated Negative Binomial (ZINB) model with Empirical Bayes (EB) adjustments for accident history. This new model is adopted from the work that was conducted previously by me. I completed the following tasks to develop my methodology to estimate the expected number of crashes at a highway-rail grade crossing in a connected vehicle environment. First, I developed the ZINEBS model (Zero Inflated Negative binomial with Empirical Bayes adjustment System). The ZINEBS model gives three different equations depending on the type of warning device used at the crossings (gates, flashing lights, and crossbucks). For crossings with gates, a model including the variables, total train, aadt, total tracks, number of highway lanes and the angle is recommended. For crossings with flashing lights, a model including the variables, total train, aadt, and posted highway speed is recommended. For crossings with crossbucks, a model including the variables total train, aadt, and type of crossing surface is recommended. The predicted values for the ZINEBS model show a closer agreement with the field data than the new FRA model. This observation was true for all three warning device types analyzed. Then, I analyze the accident data to identify crashes that can be prevented in a connected vehicle environment to create a labelled dataset to be used in a supervised machine learning model. I list out the connected vehicle technologies considered in this analysis and list the incident characteristics used to identify accidents that could be preventable (or not preventable). Based on the assumptions, a two-class label was created. Next, I develop machine learning models to estimate the likelihood of a preventable crash in a connected vehicle environment. Based on my analysis, I recommend the Random Forest model. I then develop an Inventory Based (IB) model which uses inventory variables to estimate the likelihood of a preventable crash. Finally, I put together the ZINEBS and IB models to show the application to estimate the crash reduction at a highway-rail grade crossing in a connected vehicle environment. The most significant contribution of this research work is the development of a framework that can estimate the expected number of crashes at highway-rail grade crossing. The research also gives an in-depth analysis of how the type of warning device or crossing characteristics can affect the crash types that happen at the grade crossing. The framework developed would allow its users to identify crossings posing a high safety risk based on the expected number of accidents. This framework can be used along with other non-safety factors like economic, environmental, etc., to identify crossings for upgrades or grade separation. The framework also allows a user to estimate the expected accident count at a crossing when there are connected vehicles in the traffic stream. This research will open new opportunities to innovate new V2V and V2I technologies for railroad risk reduction, leading to a safer operating environment for railroads, rail passengers, highway users, and the general public alike."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Methodology to estimate crash frequency at highway-rail grade crossings in a connected vehicle environment"]}]}],"canonical_facts":{"dc:contributor":["Benekohal, Rahim F.","Barkan, Christopher P.L.","Ouyang, Yanfeng","Liang, Feng"],"dc:creator":["Mathew, Jacob"],"dc:date":["2022-05","2022-04-19"],"dc:description":["Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2022-11-11 without embargo terms","The student, Jacob Mathew, accepted the attached license on 2022-04-18 at 12:15.","The student, Jacob Mathew, submitted this Dissertation for approval on 2022-04-18 at 18:45.","This Dissertation was approved for publication on 2022-04-19 at 13:44.","DSpace SAF Submission Ingestion Package generated from Vireo submission #17749 on 2022-11-11 at 13:15:33","As the transportation industry is heading towards the Internet of Things, new safety technologies are designed to increase situational awareness of the vehicles. These technologies can reduce or eliminate crashes and can improve the safety of transportation systems. While there have been studies to estimate the potential benefits of Connected and Autonomous Vehicles (CAV) safety applications, the quantification of safety benefits at a highway-rail grade crossing in a connected vehicle environment has not been done before. This dissertation presents a methodology to estimate the expected number of crashes at a highway-rail grade crossing in a connected vehicle environment. For over four decades, the state of practice to estimate accident counts at a railroad grade crossing was to use the USDOT accident prediction formula. Recently the Federal Railroad Administration released a new model for accident prediction at railroad grade crossings using a Zero-Inflated Negative Binomial (ZINB) model with Empirical Bayes (EB) adjustments for accident history. This new model is adopted from the work that was conducted previously by me. I completed the following tasks to develop my methodology to estimate the expected number of crashes at a highway-rail grade crossing in a connected vehicle environment. First, I developed the ZINEBS model (Zero Inflated Negative binomial with Empirical Bayes adjustment System). The ZINEBS model gives three different equations depending on the type of warning device used at the crossings (gates, flashing lights, and crossbucks). For crossings with gates, a model including the variables, total train, aadt, total tracks, number of highway lanes and the angle is recommended. For crossings with flashing lights, a model including the variables, total train, aadt, and posted highway speed is recommended. For crossings with crossbucks, a model including the variables total train, aadt, and type of crossing surface is recommended. The predicted values for the ZINEBS model show a closer agreement with the field data than the new FRA model. This observation was true for all three warning device types analyzed. Then, I analyze the accident data to identify crashes that can be prevented in a connected vehicle environment to create a labelled dataset to be used in a supervised machine learning model. I list out the connected vehicle technologies considered in this analysis and list the incident characteristics used to identify accidents that could be preventable (or not preventable). Based on the assumptions, a two-class label was created. Next, I develop machine learning models to estimate the likelihood of a preventable crash in a connected vehicle environment. Based on my analysis, I recommend the Random Forest model. I then develop an Inventory Based (IB) model which uses inventory variables to estimate the likelihood of a preventable crash. Finally, I put together the ZINEBS and IB models to show the application to estimate the crash reduction at a highway-rail grade crossing in a connected vehicle environment. The most significant contribution of this research work is the development of a framework that can estimate the expected number of crashes at highway-rail grade crossing. The research also gives an in-depth analysis of how the type of warning device or crossing characteristics can affect the crash types that happen at the grade crossing. The framework developed would allow its users to identify crossings posing a high safety risk based on the expected number of accidents. This framework can be used along with other non-safety factors like economic, environmental, etc., to identify crossings for upgrades or grade separation. The framework also allows a user to estimate the expected accident count at a crossing when there are connected vehicles in the traffic stream. This research will open new opportunities to innovate new V2V and V2I technologies for railroad risk reduction, leading to a safer operating environment for railroads, rail passengers, highway users, and the general public alike."],"dc:format":["application/pdf"],"dc:identifier":["https://hdl.handle.net/2142/115481"],"dc:language":["en","eng"],"dc:rights":["Copyright 2022 Jacob Mathew"],"dc:subject":["Accident Prediction","Highway-Rail Grade Crossing","Connected Vehicle Environment","Zero-Inflated Negative Binomial","Empirical Bayes","Machine Learning"],"dc:title":["Methodology to estimate crash frequency at highway-rail grade crossings in a connected vehicle environment"],"dc:type":["text","Thesis"],"thesis:degree_discipline":["Civil Engineering"],"thesis:degree_level":["Dissertation"],"thesis:degree_name":["Ph.D."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:24:54Z"}