Please use this identifier to cite or link to this item: http://repositorio.lnec.pt:8080/jspui/handle/123456789/1012908
Full metadata record
DC FieldValueLanguage
dc.contributor.authorMarcelino, P.pt_BR
dc.contributor.authorAntunes, M. L.pt_BR
dc.contributor.authorFortunato, E.pt_BR
dc.contributor.authorGomes, M.pt_BR
dc.date.accessioned2020-08-26T13:04:06Zpt_BR
dc.date.accessioned2020-09-03T11:00:59Z-
dc.date.available2020-08-26T13:04:06Zpt_BR
dc.date.available2020-09-03T11:00:59Z-
dc.date.issued2020-03pt_BR
dc.identifier.citation10.1007/s42947-019-0096-zpt_BR
dc.identifier.issn1996-6814pt_BR
dc.identifier.urihttp://repositorio.lnec.pt:8080/jspui/handle/123456789/1012908-
dc.description.abstractAccurate pavement performance prediction models are essential to ensure optimal allocation of resources in maintenance management. These models are developed using inventory and monitoring data regarding pavement structure, climate, traffic, and condition. However, numerous road agencies have limited pavement data. Due to the inexistence of historical data, data collection frequency, and/or quality issues, the amount of data available for the development of performance models is reduced. As a result, the resource allocation process is significantly undermined. This paper proposes a transfer learning approach to develop pavement performance prediction models in limited data contexts. The proposed transfer learning approach is based on a boosting algorithm. In particular, a modified version of the popular TrAdaBoost learning algorithm was used. To test the proposed transfer learning approach, a case study was developed using data from the Long-Term Pavement Performance (LTPP) database and from the Portuguese road administration database. The results of this work show that it is possible to develop accurate performance prediction models in limited data contexts when a transfer learning approach is applied. All the models resulting from this approach outperformed baseline models, especially in what regards long-term forecasts. The results also showed that the transfer learning models perform consistently over different time frames, with minor performance losses from one-step to multi-step forecasts. The findings of this study should be of interest to road agencies facing limited data contexts and aiming to develop accurate prediction models that can improve their pavement management practice.pt_BR
dc.language.isoengpt_BR
dc.publisherSpringerpt_BR
dc.rightsrestrictedAccesspt_BR
dc.subjectTransfer learningpt_BR
dc.subjectPavement performance modelspt_BR
dc.subjectInternational Roughness Index (IRI)pt_BR
dc.subjectMachine learningpt_BR
dc.subjectPavement Management Systems (PMS)pt_BR
dc.subjectPredictive maintenancept_BR
dc.titleTransfer Learning for Pavement Performance Predictionpt_BR
dc.typeworkingPaperpt_BR
dc.description.pages154–167pt_BR
dc.description.volume13pt_BR
dc.description.sectorDT/NITpt_BR
dc.description.magazineInternational Journal of Pavement Research and Technologypt_BR
dc.contributor.peer-reviewedSIMpt_BR
dc.contributor.academicresearchersSIMpt_BR
dc.contributor.arquivoNAOpt_BR
Appears in Collections:DT/NIT - Comunicações a congressos e artigos de revista

Files in This Item:
There are no files associated with this item.


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.