Optimize pavement paving with artificial intelligence

Expensive test pavement installations can now be saved and asphalt formulations can be checked conveniently from the office. Benefit from the statistics of over 100 SDA 4 pavements installed. Test whether your mix composition actually produces a low-noise pavement or whether it is worth optimizing an existing pavement recipe. In simple terms, asphalt pavements consist of a combination of different sized aggregates mixed with bitumen - together this produces the mixture. When paving low-noise pavements, high demands are placed on such a mixture. We at G+P have now developed a new AI tool that can be used to estimate the acoustic effect of a mixture composition.

Optimize pavement paving with artificial intelligence
Optimize pavement paving with artificial intelligence
Optimize pavement paving with artificial intelligence
Optimize pavement paving with artificial intelligence

--> https://acoustic-mixture-testing.gundp.ch/ <--
(only in german)

The tool can be used to determine the expected acoustic effect of SDA4 pavements with just a few clicks. The tool is particularly suitable for:

  • Noise control authorities: Ensure that the low-noise pavements ordered are actually low-noise.
  • Civil Engineering Offices: Determine the exact specifications of low-noise pavements.
  • Mixture plants: Checking the pavement formulation and optimizing it according to the plant's own capabilities.

With the tool, expensive test installations can thus be dispensed with and the acoustic effect can be determined based on the algorithm. A calculation of the expected void content is also presented within the tool.

How does the tool work and how do I get access?

The tool is implemented as a web tool on the site https://acoustic-mixture-testing.gundp.ch/. Access to the model can be obtained by registering a user. After that, one can test his rubbers with the model. If required, a report of the tested rubber can be generated and downloaded. als Webtool umgesetzt. 

Modeling the acoustics of the road surface

The basic ingredient of a pavement is a grain structure. This aggregate mixture is mixed with a bituminous binder and heated. The asphalt can then be applied to the road and compacted (rolled). For low-noise pavements in particular, the composition of the mixture is extremely crucial.

For more than ten years, low-noise pavements have been used intensively in Switzerland. The acoustic effect can be measured and checked using the CPX method. By digitizing the paving records of more than 100 different pavements and about 850 measurement points, an AI model was trained to predict the acoustic effect of a pavement mixture.

Validation of the model

The model was validated with a portion of the data set (20% of the data points) to test the predictive power of the model. The validation showed that the model has a very good predictive performance. This is because the agreement between the measured and the modeled points is very high. This is true from the best values (-9 dB) to the worst values (-3.5 dB). This also shows the importance of optimizing the mix. This is because the range of acoustic performance in a pavement installation can be large.

Similar articles

3D texture measurements - visual method for determining the surface properties of road pavements

3D texture measurements - visual method for determining the surface properties of road pavements

Low-noise road pavements are used as effective noise abatement measures in Switzerland. To determine the actual effect of low-noise road surfaces, G+P uses numerous measurement methods. While the most common measurement methods record the properties of road pavements acoustically (e.g. with microphones), 3D texture measurements can be used to measure and assess the visual properties of road pavements that are relevant for noise generation. G+P has been carrying out such 3D texture measurements with its own photogrammetry scanner since 2022.
How sustainable are low-noise road surfaces?

How sustainable are low-noise road surfaces?

Low-noise road surfaces are an established measure for noise protection at source, as they reduce road noise emissions over the entire service life of the road. However, their shorter service life results in higher costs for developers and increases their ecological footprint from construction activity. How do these advantages and disadvantages weigh up in a holistic comparative analysis?