Technology Artificial intelligence checks railroad bridges for stability
Zurich researchers are using artificial intelligence (AI) to check the structural safety of railroad bridges. They have developed a new AI model that predicts whether a bridge is structurally critical or not.
24.02.2025, 09:12
SDA
This makes it possible to set priorities for structural inspections, the Swiss Federal Institute of Technology in Zurich (ETH Zurich) announced on Monday. This makes it possible to determine which bridges require a structural inspection and, if necessary, structural measures in the near future.
According to ETH Zurich, reinforced concrete railroad bridges are particularly common in Switzerland. "In Switzerland, too, a considerable proportion of infrastructure structures are approaching the end of their planned service life and need to be checked and, if necessary, reinforced," Sophia Kuhn, the developer of the AI model, was quoted as saying in the press release.
Machine learning
The ETH researchers developed the program together with the Swiss Federal Railways (SBB). The system consists of an artificial neural network, i.e. an algorithm that learns from data like a brain. It was trained using a huge database with information on existing bridges.
According to the university, a major advantage of the program is that it not only says whether a bridge is stable, but also indicates how reliable this assessment is.
