Real-time RSET prediction across three types of geometries and simulation training dataset: A comparative study of machine learning models

Simple item page
dc.contributor.authorUhlík, Ondřejcs
dc.contributor.authorOkřinová, Petracs
dc.contributor.authorTokarevskikh, Artemcs
dc.contributor.authorApeltauer, Tomášcs
dc.contributor.authorApeltauer, Jiřícs
dc.coverage.issue100461cs
dc.coverage.volume18cs
dc.date.issued2024-05-24cs
dc.description.abstractAgent-based evacuation models provide useful data of the evacuation process, but they are not primarily designed for use during an emergency. The paper aims to test predicting RSET using a surrogate ML model trained on a simulation dataset with 60 samples. A total of 9 machine learning algorithms were tested on 3 simple geometries: bottleneck, stairway and walkway. A set of 7 spatial features was used to train the surrogate models. The results showed a relatively good ability of Artificial Neural Network to learn in scenarios involving bottlenecks and stairways, with an R2: 0.99 on the testing dataset. In the walkway scenario, all models experienced a significant drop in performance, with Gradient Boost performing the best (R2: 0.92). The paper demonstrated ability to generalize effectively in bottleneck-type tasks with training on a relatively small dataset containing spatial parameters obtainable in real-time from camera systems.en
dc.formattextcs
dc.format.extent13cs
dc.format.mimetypeapplication/pdfcs
dc.identifier.citationDevelopments in the Built Environment. 2024, vol. 18, issue 100461, 13 p.en
dc.identifier.doi10.1016/j.dibe.2024.100461cs
dc.identifier.issn2666-1659cs
dc.identifier.orcid0000-0001-9047-4784cs
dc.identifier.orcid0000-0003-0300-8967cs
dc.identifier.orcid0000-0003-3186-2175cs
dc.identifier.orcid0000-0002-9791-4655cs
dc.identifier.other188671cs
dc.identifier.researcheridAAG-2349-2021cs
dc.identifier.researcheridQ-2414-2015cs
dc.identifier.researcheridP-9965-2015cs
dc.identifier.scopus57197712362cs
dc.identifier.scopus35067734400cs
dc.identifier.scopus56566688300cs
dc.identifier.urihttp://hdl.handle.net/11012/249353
dc.language.isoencs
dc.publisherElseviercs
dc.relation.ispartofDevelopments in the Built Environmentcs
dc.relation.urihttps://www.sciencedirect.com/science/article/pii/S266616592400142X?via%3Dihubcs
dc.rightsCreative Commons Attribution 4.0 Internationalcs
dc.rights.accessopenAccesscs
dc.rights.sherpahttp://www.sherpa.ac.uk/romeo/issn/2666-1659/cs
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/cs
dc.subjectEvacuation modelingen
dc.subjectMachine learningen
dc.subjectRequired safe egress timeen
dc.subjectAgent-based modelsen
dc.titleReal-time RSET prediction across three types of geometries and simulation training dataset: A comparative study of machine learning modelsen
dc.type.driverarticleen
dc.type.statusPeer-revieweden
dc.type.versionpublishedVersionen
sync.item.dbidVAV-188671en
sync.item.dbtypeVAVen
sync.item.insts2025.03.06 11:53:36en
sync.item.modts2025.03.05 12:32:06en
thesis.grantorVysoké učení technické v Brně. Fakulta stavební. Ústav pozemních komunikacícs
thesis.grantorVysoké učení technické v Brně. Fakulta stavební. Ústav automatizace inženýrských úloh a informatikycs
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
1s2.0S266616592400142Xmain.pdf
Size:
5.81 MB
Format:
Adobe Portable Document Format
Description:
file 1s2.0S266616592400142Xmain.pdf
Download
Collections
Ústav automatizace inženýrských úloh a informatiky
Ústav pozemních komunikací