Friday 28 March 2025
For decades, scientists have been trying to understand how to predict the stability of landfills and prevent devastating failures that can cause environmental disasters and even loss of human life. One crucial aspect of this puzzle is determining the shear strength of municipal solid waste (MSW), which is a complex mixture of organic and inorganic materials.
Recently, researchers from King Mongkut’s University of Technology Thonburi in Thailand developed an innovative artificial intelligence (AI) model that can accurately predict the shear strength of MSW across diverse compositional profiles. This breakthrough has significant implications for waste management practices worldwide.
The team used a combination of machine learning algorithms and large-scale direct shear testing to create their model. The data set consisted of 66 samples collected from a single dumpsite, which is relatively small compared to other studies that have analyzed thousands of samples. However, the researchers demonstrated that their approach can still produce reliable predictions with high accuracy.
The AI model uses a multi-layer perceptron architecture, which is a type of neural network designed for pattern recognition and classification. By analyzing the chemical composition and physical properties of MSW, such as particle size distribution and moisture content, the model can predict the shear strength parameters of friction angle and cohesion.
One of the key findings was that fibrous materials, like food waste and textiles, play a significant role in determining the shear strength of MSW. The researchers also discovered that particle size fractions have a non-linear effect on the mechanical properties of the waste. These insights can help engineers design more effective waste management systems and predict potential failure modes.
The model’s ability to provide transparent explanations for its predictions is another major advantage. By using SHAP (SHapley Additive exPlanations) analysis, the researchers can identify which features or characteristics of MSW contribute most to the predicted shear strength values. This feature attribution capability allows engineers to better understand the relationships between waste composition and mechanical properties, enabling more informed decision-making.
The application potential of this AI model is vast. It can be used to optimize landfill design, predict slope stability, and improve waste management practices. The researchers envision that their approach will become an essential tool for geotechnical engineers, helping them to better manage the complex interactions between waste composition, moisture content, and mechanical properties.
While there are still limitations to this study, such as the small sample size and limited geographic representation, the findings demonstrate the potential of AI in solving real-world problems.
Cite this article: “Artificial Intelligence Model Predicts Shear Strength of Municipal Solid Waste with High Accuracy”, The Science Archive, 2025.
Landfill Stability, Municipal Solid Waste, Shear Strength, Artificial Intelligence, Machine Learning, Waste Management, Geotechnical Engineering, Friction Angle, Cohesion, Shap Analysis
