Geopolymer Concrete Mechanical and Environmental Performance Predictions

Artificial neural network model outcomes predict mechanical properties and CO2 emissions for fly ash and silica fume geopolymer concrete mixes. The dataset contains statistical results from models achieving high prediction accuracy, with R² values between 0.96 and 0.99. Jahanzeb Javed authored this research, which was last updated in April 2026.

Use Cases

• Predict compressive strength from mix design parameters using the ANN model results.
• Analyze the relationship between steel fiber reinforcement and flexural strength based on experimental data.
• Estimate CO2 emission reductions for geopolymer concrete compared to ordinary Portland cement using life cycle assessment outputs.
• Perform sensitivity analysis on key input parameters (e.g., fly ash content, silica fume content) influencing mechanical predictions via SHAP results.
• Validate ANN model performance for predicting ultrasonic pulse velocity measurements in non-destructive testing.

Strengths

• ANN models demonstrated high predictive accuracy with R² values ranging from 0.96 to 0.99.
• Research quantifies performance gains, reporting 20-25% improvement in early-age compressive strength and 15-20% reduction in CO2 emissions for optimized mixes.

Limitations

• The dataset is very small at 5.5 KB, indicating limited scope and likely a summary of results rather than raw experimental data.
• Specific sample size (rows) and input feature columns for the ANN models are not provided, limiting reproducibility.
• Temporal and geographic context for the underlying experimental data is unspecified, which may affect generalizability.

Provenance

Source

Jahanzeb Javed via figshare.

Collection Method

Results derived from an experimental campaign on geopolymer concrete, including destructive testing, non-destructive testing (ultrasonic pulse velocity), and subsequent artificial neural network modeling.

Freshness

Last updated in April 2026.