{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/106479"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/106479","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Quantitative analyses of freight train derailments","abstract":"Railroad safety analysis encompasses several areas, one of which is freight train risk analysis. There are various types of freight train accidents including derailments, collisions, highway-rail grade crossing incidents, and others. The mainline freight train derailment rate of the major U.S. Railroads (Class I) declined almost 50% over the past decade. Nevertheless, derailments remain the most frequent type of major railroad accident. They can damage infrastructure and rolling stock, delay transportation, and may cause casualties and the release of hazardous materials. Therefore, further reduction in derailments remains an ongoing priority of the rail industry and government. Freight train derailment analysis involves several areas: derailment rate estimation, identification of factors affecting derailment rate, root cause analysis, hazardous material transportation risk management, and consequence analysis. Statistical and risk analysis techniques can be used to evaluate progress in rail safety and prioritize areas for improvement. Past research has analyzed various factors affecting freight train derailments and introduced different models for derailment rate estimation and consequence analysis. The principal objective of this dissertation encompasses three derailment research areas: quantitative analysis of derailments, cause-specific derailment rate estimation, and severity analysis. The first part provides a quantitative analysis of several aspects of recent derailment trends and new approaches to assessing cause-specific changes. It also improves upon the methodology for derailment rate estimation and presents new, up-to-date derailment rate estimates for Class I railroad mainlines using statistical techniques. The second part develops a statistical methodology to understand the effect of two metrics of traffic exposure, train-miles and car-miles and which accident causes are related to these two metrics. Properly accounting for train-mile and car-mile causes has implications for derailment rate analysis in general, and the effect of train length in particular. A machine learning methodology is introduced to improve the classification of derailment causes as being a function of train-miles or car-miles. The last section presents a new methodology to identify factors affecting derailment severity and use of statistical learning techniques to predict derailment severity. The new approaches to derailment trend and cause analysis will provide new insights for industry, government, and researchers in assessing risk and prioritizing derailment prevention resources. The incorporation of the effect of train-mile and car-mile caused derailments will enable more accurate safety and risk analyses than was previously possible. This dissertation also introduces new methodologies for use of applied machine learning techniques to address several freight train accident analysis questions.","abstract_html":"Railroad safety analysis encompasses several areas, one of which is freight train risk analysis. There are various types of freight train accidents including derailments, collisions, highway-rail grade crossing incidents, and others. The mainline freight train derailment rate of the major U.S. Railroads (Class I) declined almost 50% over the past decade. Nevertheless, derailments remain the most frequent type of major railroad accident. They can damage infrastructure and rolling stock, delay transportation, and may cause casualties and the release of hazardous materials. Therefore, further reduction in derailments remains an ongoing priority of the rail industry and government. Freight train derailment analysis involves several areas: derailment rate estimation, identification of factors affecting derailment rate, root cause analysis, hazardous material transportation risk management, and consequence analysis. Statistical and risk analysis techniques can be used to evaluate progress in rail safety and prioritize areas for improvement. Past research has analyzed various factors affecting freight train derailments and introduced different models for derailment rate estimation and consequence analysis. The principal objective of this dissertation encompasses three derailment research areas: quantitative analysis of derailments, cause-specific derailment rate estimation, and severity analysis. The first part provides a quantitative analysis of several aspects of recent derailment trends and new approaches to assessing cause-specific changes. It also improves upon the methodology for derailment rate estimation and presents new, up-to-date derailment rate estimates for Class I railroad mainlines using statistical techniques. The second part develops a statistical methodology to understand the effect of two metrics of traffic exposure, train-miles and car-miles and which accident causes are related to these two metrics. Properly accounting for train-mile and car-mile causes has implications for derailment rate analysis in general, and the effect of train length in particular. A machine learning methodology is introduced to improve the classification of derailment causes as being a function of train-miles or car-miles. The last section presents a new methodology to identify factors affecting derailment severity and use of statistical learning techniques to predict derailment severity. The new approaches to derailment trend and cause analysis will provide new insights for industry, government, and researchers in assessing risk and prioritizing derailment prevention resources. The incorporation of the effect of train-mile and car-mile caused derailments will enable more accurate safety and risk analyses than was previously possible. This dissertation also introduces new methodologies for use of applied machine learning techniques to address several freight train accident analysis questions.","abstract_has_math":false,"creators":["Wang, Brandon Zhao"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"Ph.D.","degree_level":"Dissertation","degree_discipline":"Civil Engineering","degree_department":null,"school":null,"contributors":["Barkan, Christopher PL","Benekohal, Rahim F","Liang, Feng","Saat, Rapik"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2020,"date_issued":"2020-03-02T22:38:52Z","date_published":"2020-03-02T22:38:52Z","updated_at":"2026-07-22T22:24:47Z","subjects":["railroad engineering, railway safety, railway risk analysis, freight train, accidents, derailments, machine learning, statistics, applied statistics"],"languages":["en"],"rights":["Copyright 2019 Brandon Zhao Wang"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/2142/106479","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Barkan, Christopher PL","Benekohal, Rahim F","Liang, Feng","Saat, Rapik"]},{"key":"dc:creator","label":"Author","values":["Wang, Brandon Zhao"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2020-03-02T22:38:52Z","2022-03-03T10:15:13Z","2019-12-03","2019-12"]},{"key":"dc:type","label":"Dc Type","values":["text"]},{"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":["railroad engineering, railway safety, railway risk analysis, freight train, accidents, derailments, machine learning, statistics, applied statistics"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2019 Brandon Zhao Wang"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/2142/106479"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Railroad safety analysis encompasses several areas, one of which is freight train risk analysis. There are various types of freight train accidents including derailments, collisions, highway-rail grade crossing incidents, and others. The mainline freight train derailment rate of the major U.S. Railroads (Class I) declined almost 50% over the past decade. Nevertheless, derailments remain the most frequent type of major railroad accident. They can damage infrastructure and rolling stock, delay transportation, and may cause casualties and the release of hazardous materials. Therefore, further reduction in derailments remains an ongoing priority of the rail industry and government. Freight train derailment analysis involves several areas: derailment rate estimation, identification of factors affecting derailment rate, root cause analysis, hazardous material transportation risk management, and consequence analysis. Statistical and risk analysis techniques can be used to evaluate progress in rail safety and prioritize areas for improvement. Past research has analyzed various factors affecting freight train derailments and introduced different models for derailment rate estimation and consequence analysis. The principal objective of this dissertation encompasses three derailment research areas: quantitative analysis of derailments, cause-specific derailment rate estimation, and severity analysis. The first part provides a quantitative analysis of several aspects of recent derailment trends and new approaches to assessing cause-specific changes. It also improves upon the methodology for derailment rate estimation and presents new, up-to-date derailment rate estimates for Class I railroad mainlines using statistical techniques. The second part develops a statistical methodology to understand the effect of two metrics of traffic exposure, train-miles and car-miles and which accident causes are related to these two metrics. Properly accounting for train-mile and car-mile causes has implications for derailment rate analysis in general, and the effect of train length in particular. A machine learning methodology is introduced to improve the classification of derailment causes as being a function of train-miles or car-miles. The last section presents a new methodology to identify factors affecting derailment severity and use of statistical learning techniques to predict derailment severity. The new approaches to derailment trend and cause analysis will provide new insights for industry, government, and researchers in assessing risk and prioritizing derailment prevention resources. The incorporation of the effect of train-mile and car-mile caused derailments will enable more accurate safety and risk analyses than was previously possible. This dissertation also introduces new methodologies for use of applied machine learning techniques to address several freight train accident analysis questions.","Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2021-12-01","The student, Brandon Wang, accepted the attached license on 2019-12-03 at 14:46.","The student, Brandon Wang, submitted this Dissertation for approval on 2019-12-03 at 14:53.","This Dissertation was approved for publication on 2019-12-03 at 16:56.","DSpace SAF Submission Ingestion Package generated from Vireo submission #14672 on 2020-02-28 at 17:37:30","Made available in DSpace on 2020-03-02T22:38:52Z (GMT). No. of bitstreams: 2 WANG-DISSERTATION-2019.pdf: 1829336 bytes, checksum: 575889f9c82527da69abf33e8ee07128 (MD5) LICENSE.txt: 4209 bytes, checksum: 28e8ba80eb25bfeb708add0ba2fdf8ad (MD5) Previous issue date: 2019-12-03","Embargo set by: Seth Robbins for item 114023 Lift date: 2022-03-02T22:39:04Z Reason: Author requested closed access (OA after 2yrs) in Vireo ETD system","Limited Restriction Lifted for Item 114023 on 2022-03-03T10:15:13Z."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Quantitative analyses of freight train derailments"]}]}],"canonical_facts":{"dc:contributor":["Barkan, Christopher PL","Benekohal, Rahim F","Liang, Feng","Saat, Rapik"],"dc:creator":["Wang, Brandon Zhao"],"dc:date":["2020-03-02T22:38:52Z","2022-03-03T10:15:13Z","2019-12-03","2019-12"],"dc:description":["Railroad safety analysis encompasses several areas, one of which is freight train risk analysis. There are various types of freight train accidents including derailments, collisions, highway-rail grade crossing incidents, and others. The mainline freight train derailment rate of the major U.S. Railroads (Class I) declined almost 50% over the past decade. Nevertheless, derailments remain the most frequent type of major railroad accident. They can damage infrastructure and rolling stock, delay transportation, and may cause casualties and the release of hazardous materials. Therefore, further reduction in derailments remains an ongoing priority of the rail industry and government. Freight train derailment analysis involves several areas: derailment rate estimation, identification of factors affecting derailment rate, root cause analysis, hazardous material transportation risk management, and consequence analysis. Statistical and risk analysis techniques can be used to evaluate progress in rail safety and prioritize areas for improvement. Past research has analyzed various factors affecting freight train derailments and introduced different models for derailment rate estimation and consequence analysis. The principal objective of this dissertation encompasses three derailment research areas: quantitative analysis of derailments, cause-specific derailment rate estimation, and severity analysis. The first part provides a quantitative analysis of several aspects of recent derailment trends and new approaches to assessing cause-specific changes. It also improves upon the methodology for derailment rate estimation and presents new, up-to-date derailment rate estimates for Class I railroad mainlines using statistical techniques. The second part develops a statistical methodology to understand the effect of two metrics of traffic exposure, train-miles and car-miles and which accident causes are related to these two metrics. Properly accounting for train-mile and car-mile causes has implications for derailment rate analysis in general, and the effect of train length in particular. A machine learning methodology is introduced to improve the classification of derailment causes as being a function of train-miles or car-miles. The last section presents a new methodology to identify factors affecting derailment severity and use of statistical learning techniques to predict derailment severity. The new approaches to derailment trend and cause analysis will provide new insights for industry, government, and researchers in assessing risk and prioritizing derailment prevention resources. The incorporation of the effect of train-mile and car-mile caused derailments will enable more accurate safety and risk analyses than was previously possible. This dissertation also introduces new methodologies for use of applied machine learning techniques to address several freight train accident analysis questions.","Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2021-12-01","The student, Brandon Wang, accepted the attached license on 2019-12-03 at 14:46.","The student, Brandon Wang, submitted this Dissertation for approval on 2019-12-03 at 14:53.","This Dissertation was approved for publication on 2019-12-03 at 16:56.","DSpace SAF Submission Ingestion Package generated from Vireo submission #14672 on 2020-02-28 at 17:37:30","Made available in DSpace on 2020-03-02T22:38:52Z (GMT). No. of bitstreams: 2 WANG-DISSERTATION-2019.pdf: 1829336 bytes, checksum: 575889f9c82527da69abf33e8ee07128 (MD5) LICENSE.txt: 4209 bytes, checksum: 28e8ba80eb25bfeb708add0ba2fdf8ad (MD5) Previous issue date: 2019-12-03","Embargo set by: Seth Robbins for item 114023 Lift date: 2022-03-02T22:39:04Z Reason: Author requested closed access (OA after 2yrs) in Vireo ETD system","Limited Restriction Lifted for Item 114023 on 2022-03-03T10:15:13Z."],"dc:format":["application/pdf"],"dc:identifier":["http://hdl.handle.net/2142/106479"],"dc:language":["en"],"dc:rights":["Copyright 2019 Brandon Zhao Wang"],"dc:subject":["railroad engineering, railway safety, railway risk analysis, freight train, accidents, derailments, machine learning, statistics, applied statistics"],"dc:title":["Quantitative analyses of freight train derailments"],"dc:type":["text"],"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":"2026-07-22T22:24:47Z"}