Research article Open Access Logo

Designing the framework for predictive maintenance method for road surface damage management System to resilience and sustainability

Son Hong Vu Pham 1, 2
khoi Tien Van Nguyen 1, 2, *
  1. Faculty of Civil Engineering, Ho Chi Minh City University of Technology (HCMUT), Ho Chi Minh City, Vietnam
  2. Vietnam National University (VNU-HCM), Ho Chi Minh City, Vietnam
Correspondence to: khoi Tien Van Nguyen, Faculty of Civil Engineering, Ho Chi Minh City University of Technology (HCMUT), Ho Chi Minh City, Vietnam; Vietnam National University (VNU-HCM), Ho Chi Minh City, Vietnam. Email: [email protected].
Volume & Issue: Vol. 9 No. 3 (2026) | Page No.: 3216-3235 | DOI: 10.32508/vnuhcmj-et.v9i3.1530
Published: 2026-08-25

Online metrics


Statistics from the website

  • Abstract Views: 0
  • Galley Views: 0

Statistics from Dimensions

This article is published with open access by Viet Nam National University, Ho Chi Minh City, Viet Nam. This article is distributed under the terms of the Creative Commons Attribution License (CC-BY 4.0) which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited. 

Abstract

Climate change not only increases the frequency and intensity of extreme weather events but also poses significant challenges to transportation infrastructure, particularly road networks. These impacts accelerate pavement deterioration, increase uncertainty in predicting infrastructure service life, and directly affect traffic safety and operational efficiency. In response to these challenges, this study proposes an intelligent automated pavement quality management solution designed to preserve the value of road infrastructure assets throughout their operational lifecycle. The primary objective is to develop an intelligent pavement damage management framework that enhances the predictive capability and climate resilience of road transportation systems under the increasing impacts of climate change. The proposed system integrates key Industry 4.0 technologies, combining machine learning, Global Positioning System (GPS) localization, and a Decision Support System (DSS). Specifically, the framework employs the YOLOv8 model trained on a localized dataset containing more than 11,000 annotated pavement damage images collected in Vietnam, enabling real-time detection and geospatial localization of multiple pavement distress types. Experimental results demonstrate that the proposed model achieves high detection accuracy, with an [email protected] exceeding 93.2%, robust recognition performance under diverse pavement conditions, and efficient inference speed of 0.91 seconds per frame. Beyond pavement damage detection, the system incorporates a predictive analytics module to estimate pavement deterioration trends and a DSS to prioritize maintenance interventions based on damage severity, associated risks, and climate change impacts. Compared with previous studies that primarily focused on damage detection or sensor-based monitoring, this research contributes a comprehensive pavement management framework that transforms conventional reactive inspection practices into proactive, predictive, and climate-adaptive infrastructure management. The proposed framework provides an effective approach to improving the sustainability, resilience, and long-term performance of road transportation infrastructure under changing climatic conditions.

Comments