SYDNEY, Australia — Recently, machine-learning technology has paved way for significant advancements in almost all fields and areas of research. Now, a team of University of Technology Sydney researchers has generated new technology to improve road conditions and reduce potholes and maintenance costs in road construction.
This groundbreaking work uses the “Goldilocks” principle that focuses on making something “just right.”
Normally, roads consist of three or more rolled and compacted layers of soil and a mixture of materials such as crushed up rock with asphalt or concrete on top. Since soil can vary in nature and moisture, it can lead to an imbalance in compaction quality.
“Like Goldilocks, the compaction needs to be ‘just right’ to provide the correct structural integrity and strength. Over-compaction can break down the material and change its composition, and under-compaction can lead to uneven settlement,” says Associate Professor Behzad Fatahi, head of geotechnical and transport engineering, in a university release.
Rain and flooding, mixed with poor quality road construction, results in weak and unsafe roads that are more prone to tire blowouts and car accidents. This emphasizes the need for quality improvement measures to be in place. To address this, the team has developed an “intelligent compaction” technology, which they incorporate into a road roller to assess road base compaction quality in real-time by processing the data from a sensor attached to the roller.
“We have developed an advanced computer model that incorporates machine-learning and big data from construction sites to predict the stiffness of compacted soil with a high degree of accuracy in a fraction of second, so roller operators can make adjustments,” says Fatahi.
A well-balanced road is a safer road
A well-balanced, layered, and compacted road can increase the road’s ability to bear the weight of heavy loads and become more resilient. Such a road doesn’t degrade as easily over time in response to heavy use and harsh weather. The team points out that trucks can be incredibly heavy even without carrying lots of material inside as they normally do, so it’s incredibly important for roads to be able to withstand the wear and not weaken as quickly.
Future directions of this research involve testing this new machine-based technique on different ground and roller conditions across an array of projects on roads, railways, and even dam construction sites. Through this, the team hopes to study new methods for measuring soul density and moisture content in real-time during the construction process to promote safer and more efficient outcomes.
The findings are published in the journal Engineering Structures.