Incorporating the effectiveness of preservation and rehabilitation techniques on flexible pavement service life predictions using machine learning approach

Citir, NazikCeylan, HalilKim, SunghwanKaya, Orhan2025-01-062025-01-062021978-078448351-010.1061/9780784483510.0332-s2.0-85108079631https://doi.org/10.1061/9780784483510.033https://hdl.handle.net/20.500.14669/1401The Transportation and Development Institute (T and DI) of the American Society of Civil Engineers (ASCE)International Airfield and Highway Pavements 2021: Pavement Materials and Sustainability -- 8 June 2021 through 10 June 2021 -- Virtual, Online -- 169470A flexible pavement network may be exposed to traffic changes over time because of detours that may cause unusually high traffic loading on some parts of the pavement sections. Such sections, therefore, need to be preserved earlier than their expected service life. This study focuses on incorporating the effectiveness of preservation and rehabilitation techniques, e.g., functional thin overlay and structural asphalt overlay on flexible pavement service life predictions using a machine learning approach. The developed models are capable of estimating the current and future international roughness index (IRI) and of detecting performance recovery after application of preservation treatment, thereby forecasting the remaining service life. The traffic, pavement structural, and performance data were obtained from the pavement management information system (PMIS) of the Iowa Department of Transportation (DOT). This study describes realistic what-if scenarios considering various treatment techniques to provide a better understanding of the effectiveness of treatments on extending flexible pavement service life. Such outcomes will help facilitate better pavement preservation management strategies. © ASCE.All right reserved.eninfo:eu-repo/semantics/closedAccessForecastingHighway administrationInformation managementMachine learningPredictive analyticsService lifeSustainable developmentTraffic surveysTuring machinesDepartment of TransportationInternational roughness indexMachine learning approachesPavement management information systemsPavement preservationPerformance recoveryRehabilitation techniquesTreatment techniquesPavementsIncorporating the effectiveness of preservation and rehabilitation techniques on flexible pavement service life predictions using machine learning approachConference Object377365