PROTECT YOUR DNA WITH QUANTUM TECHNOLOGY
Orgo-Life the new way to the future Advertising by AdpathwayOpen-access content Jack Loughran —
Tue 25 Feb 2025
An AI tool designed to keep bridges in operation for longer has been developed by ETH Zurich researchers.
The team developed a model for railway bridges made of reinforced concrete – which are particularly common in Switzerland – that provides an initial assessment of structural safety. They also developed an AI assistant that helps engineers to design new bridges.
The average lifespan of a bridge is typically considered to be around 75-100 years depending on the materials used and the amount of traffic, although this can be extended through careful maintenance.
“Switzerland is facing a situation in which a considerable proportion of its infrastructure is nearing the end of its planned service life and must be inspected and strengthened if necessary,” said doctoral researcher Sophia Kuhn. “We’re developing a tool that helps to keep bridges in operation for as long as possible and therefore [conserves] resources without running a disproportionate risk of [an] accident.”
The AI model is able to provide an initial assessment of the structural safety of “rigid frame bridges” and predict whether they are reaching a critical state. This allows bridges with the highest risk factors to be prioritised for structural assessment.
The last few years have seen some major bridge collapses that were deemed to have occurred due to age and lack of proper maintenance. This includes the collapse of a motorway bridge in Genoa in 2018, which led to 43 deaths, and a collapsed tram bridge over the River Elbe in Dresden in September 2024.
The AI tool allows users to input the features of existing bridges to provide forecasts within seconds of how well specific bridges can withstand static loads.
“We looked at lots of examples – how they’re built, how variable they are – and developed a parametric simulation pipeline based on them,” Kuhn explained. The team then built an artificial neural network – an algorithm that learns from the data in a similar way to a human brain.
This gave rise to a machine learning-based model that delivers the desired predictions for many existing rigid frame bridges, even if these have not been calculated by experts or by the simulation pipeline. The researchers have validated their model on a test dataset and with real bridge examples to ensure its efficacy.