Artificial intelligence (AI) has the potential to create smarter infrastructure – from predicting potholes to designing more durable concrete – according to a researcher from the University of Mississippi.
Ali Behnood is Assistant Professor of Civil Engineering at the university with more than 10 years in this field of study under his belt. To date, he has contributed more than 60 published research articles about AI in sustainable infrastructure.
“The goal of our team in the NextGen Infrastructure Lab is to move toward the next generation of sustainable and resilient infrastructure,” said Behnood. “We’re trying to optimise the use of recycled materials, industrial by-products, renewable resources and alternative sustainable materials in construction while reducing not only physical cost, but labour costs, energy costs, environmental impact costs and lifecycle maintenance expense as well.”
In one of his more recent publications, Behnood and Abolfazl Afshin, an Ole Miss doctoral student in civil engineering from Zahedan, Iran, tested different artificial intelligence algorithms, to see how well they could predict asphalt pavements with reclaimed asphalt pavement materials would withstand moisture.
The aim was to understand whether AI has the capability to effectively predict moisture damage. When water leaks into asphalt, it can break the bonds that hold the materials together, and subsequently crack. Weakened pavements cost state and local governments in the US more than $206 billion on maintaining the nation’s roads in 2021. The US Department of Transportation in 2023 reported almost $1 trillion in backlog repairs and maintenance.
“We focused on moisture damage, which is one of the most critical issues in asphalt pavements, particularly for wet and cold regions, because it results in a variety of distresses like stripping, potholes and cracking,” explained Behnood. “We evaluated the effectiveness of four different artificial intelligence algorithms in predicting moisture damage in asphalt mixtures containing (reclaimed asphalt pavement) materials.
“What we found was that these algorithms are able to effectively predict moisture damage in asphalt mixtures with high accuracy. Based on these results, we can optimize material selection and predict failure probability in the pavement’s life cycle.”
Being able to identify the best mixture of reclaimed asphalt pavement and other materials able to withstand wet and cold weather conditions without AI could be time-consuming and cost-intensive, according to Behnood.
“Artificial intelligence-based algorithms offer a cost-effective and efficient alternative to traditional time-consuming and energy-intensive lab-based approaches,” he remarked.
If businesses are interested in developing more sustainable, cost-friendly infrastructure, they can use the procedures Behnood and his team developed.
“The results of all these studies can be used by practicing engineers, by the Department of Transportation, federal agencies, private sectors – whoever who works in this area – to move toward sustainable, cost-effective approaches in the design,” he said. “The tools we develop can be used by any practicing engineers.”
On top of predicting weakened pavements, other aspects of infrastructure can be streamlined using AI and machine learning (ML) including designing better bridges, waste management, and monitoring railroads for faults or breakages.
“AI can also play a crucial role in disaster resilience and risk management,” said Behnood. “In the event of disasters or natural hazards, evacuation becomes critical, and AI can identify optimised routes tailored to various evacuation scenarios, ensuring efficiency and safety.
“There are so many examples of how we can use AI for sustainability in all elements of construction and infrastructure. This is a huge area, and we are doing our little part in this huge area to move toward sustainability and to help society.”
There’s plenty of other editorial on our sister site, Electronic Specifier! Or you can always join in the conversation by commenting below or visiting our LinkedIn page.
