Concrete Model Making with AI: Predicting Compressive Strength Using Sustainable Materials

Brief 

Learn how AI-driven concrete model making predicts compressive strength using sustainable materials, advancing eco-friendly construction methods.

 

Insight

The construction industry’s focus on sustainability has driven innovations like using artificial intelligence (AI) to predict concrete compressive strength, particularly in concrete made with waste marble (WM). This research examines three AI models developed using artificial neural networks (ANN) combined with ant colony optimization (ACO) and biogeography-based optimization (BBO) techniques.

These AI-based approaches, which bypass traditional, resource-intensive lab testing, provide quick, accurate predictions of concrete strength based on various factors, including cement, water, and specimen age. The ANN-BBO model showed the highest accuracy, achieving R² values of 0.9955, 0.9882, and 0.9867 in training, validation, and testing phases, respectively. Moreover, 94–98% of predictions fell within a narrow error margin, underscoring the robustness of the AI approach.

Notably, this study highlights the effectiveness of WM as a partial cement substitute, enhancing sustainability without compromising concrete’s compressive strength. Key variables like specimen age and cement content significantly impacted strength, while the optimum WM substitution level was around 15% of total cement content.

By integrating AI and sustainable materials, this approach aligns with the industry’s evolving demands for eco-friendly construction methods. The study encourages future research to incorporate diverse variables and evaluate the durability of WM-based concrete in varying environments, promising broader applicability and more sustainable construction practices.

 

Highlight

1. This research examines three AI models developed using artificial neural networks (ANN) combined with ant colony optimization (ACO) and biogeography-based optimization (BBO) techniques.
2. The ANN-BBO model showed the highest accuracy, achieving R² values of 0.9955, 0.9882, and 0.9867 in training, validation, and testing phases, respectively.
3. Notably, this study highlights the effectiveness of WM as a partial cement substitute, enhancing sustainability without compromising concrete’s compressive strength

 

READ MORE

 

 

Related Questions:

• How do you estimate the strength of concrete?
• What is the use of marble waste in concrete?

 

 

Related Publications:

• ASCE
• ICE
• GBO
• GSC