Modeling Remaining Service Life and Structural Health Monitoring of Roads with Machine Learning and Deep Learning

EasyChair Preprint 15645

Abstract

The integration of machine learning (ML) and deep learning (DL) in structural health monitoring (SHM) and remaining service life (RSL) has revolutionized the ability to assess and maintain critical infrastructure. This review looks at the current state of SHM methods that use ML and DL. This is done by providing a detailed taxonomy that groups these methods into groups based on algorithmic strategies, data sources, and specific SHM and RSL applications. Using Scopus as the primary source for the literature, we conducted a systematic review following PRISMA guidelines to ensure thorough screening and quality assessment of most relevant studies. The review covers key areas that include supervised and unsupervised learning techniques, neural networks, and their applications to structural damage detection, failure prediction, improving precision in monitoring. Based on the trend analysis and highlighting of some of the challenges in this context, this review has identified a few future opportunities for applying advanced learning techniques to SHM to improve infrastructure safety and management.

Keyphrases: Remaining service life, Structural Health Monitoring, deep learning, machine learning

@booklet{EasyChair:15645,
  author    = {Mohammed Mudabbir and Amir Mosavi and Felde Imre and Natacha Moniz and Kazizat Iskakov and Azodinia Mohamadreza},
  title     = {Modeling Remaining Service Life and Structural Health Monitoring of Roads with Machine Learning and Deep Learning},
  howpublished = {EasyChair Preprint 15645},
  year      = {EasyChair, 2025}}