Intelligent Asset Management for Improved Mobility: Technology Transfer for South Carolina

Ziehl, Paul; Ai, Li; Comert, Gurcan · 2024 · ROSA P / Center for Connected Multimodal Mobility, Clemson University

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This report details the development and implementation of a digital twin framework for bridge load rating in South Carolina, addressing the inefficiencies of traditional inspection methods. Conventional approaches are labor-intensive, costly, and rely on static models that fail to capture real-time structural changes, posing risks for aging infrastructure. To mitigate these issues, the researchers developed a graphical user interface (GUI) that integrates real-time sensor data with high-fidelity finite element models (FEM) to provide dynamic, accurate load assessments. The methodology employs a five-step process within the GUI. First, vehicle loads are assessed using acoustic emission (AE) sensors installed on bridges. A probabilistic machine learning algorithm, specifically a random forest model, analyzes AE signals to determine the load level of passing vehicles. Laboratory testing demonstrated this method achieved 97% overall accuracy, with precision scores ranging from 93% to 100% across different load levels. Second, the assessed load is applied to a high-fidelity FEM of the bridge span to simulate mechanical responses, such as strain and displacement. Third, the FEM is validated and updated by comparing its simulated outputs with actual strain measurements from fiber optic and BDI strain gauges. Fourth, drone inspections combined with computer vision algorithms detect surface cracks to update the bridge’s condition factor in the load rating formula. Finally, users input specific truck types and positions into the GUI to calculate the load rating factor. The study found that the digital twin approach accurately simulates bridge responses, with FEM strain predictions closely matching field gauge data. The integration of drone-based crack detection allows for automated updates to condition factors, categorizing bridge states as Good (factor 1.00), Fair (0.95), or Poor (0.85) based on crack spacing. The GUI successfully consolidates these data streams, enabling users to perform load ratings without manual traffic control or extensive labor. Additionally, a workshop involving industry partners like IBM and Structural Monitoring Solutions highlighted the potential for integrating this technology with enterprise asset management platforms, such as IBM Maximo, to facilitate predictive maintenance. The significance of this work lies in its potential to transform infrastructure management from reactive to proactive. By leveraging digital twins, agencies can enhance bridge safety, extend infrastructure lifespan, and reduce maintenance costs through precise, data-driven decision-making. The study demonstrates that combining AE-based load assessment, real-time FEM validation, and automated visual inspections offers a robust, scalable solution for intelligent asset management, addressing the critical need for efficient monitoring of the nation’s aging bridge network.

Key finding

The developed digital twin-based graphical user interface accurately assesses vehicle loads using acoustic emission data and updates finite element models with real-time sensor data to improve bridge load rating precision.

Methodology

modeling

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• telematics crash prediction