Combining AI and Machine Learning with Building Data Modeling to Detect and Locate Defects in Railway Infrastructure
Year: 2024
Type: article
Abstract: Owing to the elevated need for the railway system nowadays, the velocity and capacity of locomotives are inclined to rise. The impact of the harsh climate is concurrently worsening. This leads to degrading railway infrastructures (RI), resulting in flaws. Deficiencies may impact customer satisfaction and the operational security of the RI. Early detection of faults in RI helps mitigate operational risks, save maintenance costs, and enhance resource handling efficiency. This study integrated Artificial Intelligence (AI) and machine learning (ML) with building data modeling (BDM) to identify and pinpoint faults in RI. The BDM model proposed in the research may be completely integrated with ML and AI to identify problems in RI using the established process. The Convolutional Neural Network (CNN) model yields the optimal results. The study's findings indicate that the amalgamation of BDM with ML and AI is feasible and benefits the railway sector.
Authors: Robert Müller, Nour El Din Sharaf
Institution: ETH Zurich
Language: English
Subfield: Civil and Structural Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4412838934