Machine learning applied to road safety modeling : a systematic literature review
Saved in:
| 主要作者: | |
|---|---|
| Publication Date: | 2020 |
| 其他作者: | , |
| 格式: | Article |
| 語言: | eng |
| Source: | Repositório Institucional da UnB |
| Download full: | http://repositorio.unb.br/handle/10482/51304 https://doi.org/10.1016/j.jtte.2020.07.004 https://orcid.org/0000-0003-3249-9603 https://orcid.org/0000-0001-7469-3186 |
總結: | Road safety modeling is a valuable strategy for promoting safe mobility, enabling the development of crash prediction models (CPM) and the investigation of factors contributing to crash occurrence. This modeling has traditionally used statistical techniques despite acknowledging the limitations of this kind of approach (specific assumptions and prior definition of the link functions), which provides an opportunity to explore alternatives such as the use of machine learning (ML) techniques. This study reviews papers that used ML techniques for the development of CPM. A systematic literature review protocol was conducted, that resulted in the analysis of papers and their systematization. Three types of models were identified: crash frequency, crash classification by severity, and crash frequency and severity. The first is a regression problem, the second, a classificatory one and the third can be approached either as a combination of the preceding two or as a regression model for the expected number of crashes by severity levels. The main groups of techniques used for these purposes are nearest neighbor classification, decision trees, evolutionary algorithms, support-vector machine, and artificial neural networks. The last one is used in many kinds of approaches given the ability to deal with both regression and classification problems, and also multivariate response models. This paper also presents the main performance metrics used to evaluate the models and compares the results, showing the clear superiority of the ML-based models over the statistical ones. In addition, it identifies the main explanatory variables used in the models, which shows the predominance of road-environmental aspects as the most important factors contributing to crash occurrence. The review fulfilled its objective, identifying the various approaches and the main research characteristics, limitations, and opportunities, and also highlighting the potential of the usage of ML in crash analyses. |
| _version_ | 1871442236319727616 |
|---|---|
| author | Silva, Philippe Barbosa |
| author2 | Andrade, Michelle Ferreira, Sara |
| author2_role | author author |
| author_browse | Andrade, Michelle Ferreira, Sara Silva, Philippe Barbosa |
| author_facet | Silva, Philippe Barbosa Andrade, Michelle Ferreira, Sara |
| author_role | author |
| bitstream.checksum.fl_str_mv | aed4704d04bb260d4decd80db311aaa5 53012aa3ce3d747176c5bbc82b3a9fa6 |
| bitstream.checksumAlgorithm.fl_str_mv | MD5 MD5 |
| bitstream.url.fl_str_mv | http://repositorio.unb.br/bitstream/10482/51304/2/license.txt http://repositorio.unb.br/bitstream/10482/51304/1/ARTIGO_MachineLearningApplied.pdf |
| collection | Repositório Institucional da UnB |
| dc.contributor.affiliation.pt_BR.fl_str_mv | Goiano Federal Institute-Rio Verde Campus, Department of Civil Engineering University of Brasilia, Department of Civil and Environmental Engineering University of Brasilia, Department of Civil and Environmental Engineering University of Porto, Research Centre for Territory, Transports and Environment |
| dc.contributor.author.fl_str_mv | Silva, Philippe Barbosa Andrade, Michelle Ferreira, Sara |
| dc.date.accessioned.fl_str_mv | 2025-01-08T12:40:10Z |
| dc.date.available.fl_str_mv | 2025-01-08T12:40:10Z |
| dc.date.issued.fl_str_mv | 2020-10-31 |
| dc.identifier.citation.fl_str_mv | SILVA, Philippe Barbosa; ANDRADE, Michelle; FERREIRA, Sara. Machine learning applied to road safety modeling: a systematic literature review. Journal of Traffic and Transportation Engineering, [S. l.], v. 7, n. 6, p. 775-790, Dec. 2020. DOI: https://doi.org/10.1016/j.jtte.2020.07.004. Disponível em: https://www.sciencedirect.com/science/article/pii/S2095756420301410?via%3Dihub. Acesso em: 08 jan. 2025. |
| dc.identifier.doi.pt_BR.fl_str_mv | https://doi.org/10.1016/j.jtte.2020.07.004 |
| dc.identifier.orcid.pt_BR.fl_str_mv | https://orcid.org/0000-0003-3249-9603 https://orcid.org/0000-0001-7469-3186 |
| dc.identifier.uri.fl_str_mv | http://repositorio.unb.br/handle/10482/51304 |
| dc.language.iso.fl_str_mv | eng |
| dc.publisher.none.fl_str_mv | Elsevier B.V. on behalf of Owner |
| dc.rights.driver.fl_str_mv | info:eu-repo/semantics/openAccess |
| dc.source.none.fl_str_mv | reponame:Repositório Institucional da UnB instname:Universidade de Brasília (UnB) instacron:UNB |
| dc.subject.keyword.pt_BR.fl_str_mv | Engenharia de transportes Modelagem de segurança viária Modelos de previsão de acidentes Gravidade dos ferimentos em acidentes Aprendizado de máquina |
| dc.title.pt_BR.fl_str_mv | Machine learning applied to road safety modeling : a systematic literature review |
| dc.type.driver.fl_str_mv | info:eu-repo/semantics/article |
| dc.type.status.fl_str_mv | info:eu-repo/semantics/publishedVersion |
| description | Road safety modeling is a valuable strategy for promoting safe mobility, enabling the development of crash prediction models (CPM) and the investigation of factors contributing to crash occurrence. This modeling has traditionally used statistical techniques despite acknowledging the limitations of this kind of approach (specific assumptions and prior definition of the link functions), which provides an opportunity to explore alternatives such as the use of machine learning (ML) techniques. This study reviews papers that used ML techniques for the development of CPM. A systematic literature review protocol was conducted, that resulted in the analysis of papers and their systematization. Three types of models were identified: crash frequency, crash classification by severity, and crash frequency and severity. The first is a regression problem, the second, a classificatory one and the third can be approached either as a combination of the preceding two or as a regression model for the expected number of crashes by severity levels. The main groups of techniques used for these purposes are nearest neighbor classification, decision trees, evolutionary algorithms, support-vector machine, and artificial neural networks. The last one is used in many kinds of approaches given the ability to deal with both regression and classification problems, and also multivariate response models. This paper also presents the main performance metrics used to evaluate the models and compares the results, showing the clear superiority of the ML-based models over the statistical ones. In addition, it identifies the main explanatory variables used in the models, which shows the predominance of road-environmental aspects as the most important factors contributing to crash occurrence. The review fulfilled its objective, identifying the various approaches and the main research characteristics, limitations, and opportunities, and also highlighting the potential of the usage of ML in crash analyses. |
| eu_rights_str_mv | openAccess |
| format | article |
| id | UNB_2075ac5979746182ec00d7f35c114fd9 |
| identifier_str_mv | SILVA, Philippe Barbosa; ANDRADE, Michelle; FERREIRA, Sara. Machine learning applied to road safety modeling: a systematic literature review. Journal of Traffic and Transportation Engineering, [S. l.], v. 7, n. 6, p. 775-790, Dec. 2020. DOI: https://doi.org/10.1016/j.jtte.2020.07.004. Disponível em: https://www.sciencedirect.com/science/article/pii/S2095756420301410?via%3Dihub. Acesso em: 08 jan. 2025. |
| instacron_str | UNB |
| institution | UNB |
| instname_str | Universidade de Brasília (UnB) |
| language | eng |
| network_acronym_str | UNB |
| network_name_str | Repositório Institucional da UnB |
| oai_identifier_str | oai:repositorio.unb.br:10482/51304 |
| publishDate | 2020 |
| publishDateSort | 2020 |
| publisher.none.fl_str_mv | Elsevier B.V. on behalf of Owner |
| reponame_str | Repositório Institucional da UnB |
| repository.mail.fl_str_mv | repositorio@unb.br |
| repository.name.fl_str_mv | Repositório Institucional da UnB - Universidade de Brasília (UnB) |
| repository_id_str | |
| spelling | Silva, Philippe BarbosaAndrade, MichelleFerreira, SaraGoiano Federal Institute-Rio Verde Campus, Department of Civil EngineeringUniversity of Brasilia, Department of Civil and Environmental EngineeringUniversity of Brasilia, Department of Civil and Environmental EngineeringUniversity of Porto, Research Centre for Territory, Transports and Environment2025-01-08T12:40:10Z2025-01-08T12:40:10Z2020-10-31SILVA, Philippe Barbosa; ANDRADE, Michelle; FERREIRA, Sara. Machine learning applied to road safety modeling: a systematic literature review. Journal of Traffic and Transportation Engineering, [S. l.], v. 7, n. 6, p. 775-790, Dec. 2020. DOI: https://doi.org/10.1016/j.jtte.2020.07.004. Disponível em: https://www.sciencedirect.com/science/article/pii/S2095756420301410?via%3Dihub. Acesso em: 08 jan. 2025.http://repositorio.unb.br/handle/10482/51304https://doi.org/10.1016/j.jtte.2020.07.004https://orcid.org/0000-0003-3249-9603https://orcid.org/0000-0001-7469-3186engElsevier B.V. on behalf of OwnerThis is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).info:eu-repo/semantics/openAccessMachine learning applied to road safety modeling : a systematic literature reviewinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleEngenharia de transportesModelagem de segurança viáriaModelos de previsão de acidentesGravidade dos ferimentos em acidentesAprendizado de máquinaRoad safety modeling is a valuable strategy for promoting safe mobility, enabling the development of crash prediction models (CPM) and the investigation of factors contributing to crash occurrence. This modeling has traditionally used statistical techniques despite acknowledging the limitations of this kind of approach (specific assumptions and prior definition of the link functions), which provides an opportunity to explore alternatives such as the use of machine learning (ML) techniques. This study reviews papers that used ML techniques for the development of CPM. A systematic literature review protocol was conducted, that resulted in the analysis of papers and their systematization. Three types of models were identified: crash frequency, crash classification by severity, and crash frequency and severity. The first is a regression problem, the second, a classificatory one and the third can be approached either as a combination of the preceding two or as a regression model for the expected number of crashes by severity levels. The main groups of techniques used for these purposes are nearest neighbor classification, decision trees, evolutionary algorithms, support-vector machine, and artificial neural networks. The last one is used in many kinds of approaches given the ability to deal with both regression and classification problems, and also multivariate response models. This paper also presents the main performance metrics used to evaluate the models and compares the results, showing the clear superiority of the ML-based models over the statistical ones. In addition, it identifies the main explanatory variables used in the models, which shows the predominance of road-environmental aspects as the most important factors contributing to crash occurrence. The review fulfilled its objective, identifying the various approaches and the main research characteristics, limitations, and opportunities, and also highlighting the potential of the usage of ML in crash analyses.Faculdade de Tecnologia (FT)Departamento de Engenharia Civil e Ambiental (FT ENC)Programa de Pós-Graduação em Transportesreponame:Repositório Institucional da UnBinstname:Universidade de Brasília (UnB)instacron:UNBLICENSElicense.txtlicense.txttext/plain102http://repositorio.unb.br/bitstream/10482/51304/2/license.txtaed4704d04bb260d4decd80db311aaa5MD52open accessORIGINALARTIGO_MachineLearningApplied.pdfARTIGO_MachineLearningApplied.pdfapplication/pdf1608591http://repositorio.unb.br/bitstream/10482/51304/1/ARTIGO_MachineLearningApplied.pdf53012aa3ce3d747176c5bbc82b3a9fa6MD51open access10482/513042025-01-08 09:40:11.459open accessoai:repositorio.unb.br:10482/51304U3VibWlzc8OjbyBlZmV0aXZhZGEgZGUgYWNvcmRvIGNvbSBsaWNlbsOnYSBjb25jZWRpZGEgcGVsbyBhdXRvciBlL291IGRldGVudG9yIGRvcyBkaXJlaXRvcyBhdXRvcmFpcy4KRepositório InstitucionalPUBhttps://repositorio.unb.br/oai/requestrepositorio@unb.bropendoar:2025-01-08T12:40:11Repositório Institucional da UnB - Universidade de Brasília (UnB) |
| spellingShingle | Machine learning applied to road safety modeling : a systematic literature review Silva, Philippe Barbosa Engenharia de transportes Modelagem de segurança viária Modelos de previsão de acidentes Gravidade dos ferimentos em acidentes Aprendizado de máquina |
| status_str | publishedVersion |
| title | Machine learning applied to road safety modeling : a systematic literature review |
| title_full | Machine learning applied to road safety modeling : a systematic literature review |
| title_fullStr | Machine learning applied to road safety modeling : a systematic literature review |
| title_full_unstemmed | Machine learning applied to road safety modeling : a systematic literature review |
| title_short | Machine learning applied to road safety modeling : a systematic literature review |
| title_sort | Machine learning applied to road safety modeling : a systematic literature review |
| topic | Engenharia de transportes Modelagem de segurança viária Modelos de previsão de acidentes Gravidade dos ferimentos em acidentes Aprendizado de máquina |
| url | http://repositorio.unb.br/handle/10482/51304 https://doi.org/10.1016/j.jtte.2020.07.004 https://orcid.org/0000-0003-3249-9603 https://orcid.org/0000-0001-7469-3186 |