A Comparative Study of Random Forest and Genetic Engineering Programming for the Prediction of Compressive Strength of High Strength Concrete (HSC)

Farooq, Furqan; Amin, Muhammad; Khan, Kaffayatullah; Sadiq, Muhammad Rehan; Javed, Muhammad Faisal; Aslam, Fahid; Alyousef, Rayed

doi:10.3390/app10207330

Cited by 158 publications

(53 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, a perfect fit can be achieved when the RRSE value is zero. Farooq et al [ 70 ] developed a model to predict the compressive strength of high performance concrete. Their model exhibited high performance with RSE values of 0.092 and 0.023 for both validation and test data respectively.…”

Section: Resultsmentioning

confidence: 99%

Optimisation of GBFS, Fly Ash, and Nano-Silica Contents in Alkali-Activated Mortars

et al. 2021

View full text Add to dashboard Cite

Although free-cement-based alkali-activated paste, mortar, and concrete have been recognised as sustainable and environmental-friendly materials, a considerable amount of effort is still being channeled to ascertain the best binary or ternary binders that would satisfy the requirements of strength and durability as well as environmental aspects. In this study, the mechanical properties of alkali-activated mortar (AAM) made with binary binders, involving fly ash (FA) and granulated blast-furnace slag (GBFS) as well as bottle glass waste nano-silica powder (BGWNP), were opti-mised using both experimentally and optimisation modelling through three scenarios. In the first scenario, the addition of BGWNP varied from 5% to 20%, while FA and GBFS were kept constant (30:70). In the second and third scenarios, BGWNP (5–20%) was added as the partial replacement of FA and GBFS, separately. The results show that the combination of binary binders (FA and GBFS) and BGWNP increased AAM’s strength compared to that of the control mixture for all scenarios. In addition, the findings also demonstrated that the replacement of FA by BGWNP was the most significant, while the effect of GBFS replacement by BGWNP was less significant. In particular, the highest improvement in compressive strength was recorded when FA, GBFS, and BGWNP were 61.6%, 30%, and 8.4%, respectively. Furthermore, the results of ANOVA (p values < 0.0001 and high F-values) as well as several statistical validation methods (R > 0.9, RAE < 0.1, RSE < 0.013, and RRSE < 0.116) confirmed that all the models were robust, reliable, and significant. Similarly, the data variation was found to be less than 5%, and the difference between the predicted R2 and adj. R2 was very small (<0.2), thus confirming that the proposed non-linear quadratic equations had the capability to predict for further observation. In conclusion, the use of BGWNP in AAM could act as a beneficial and sustainable strategy, not only to address environmental issues (e.g., landfill) but to also enhance strength properties.

show abstract

Section: Resultsmentioning

confidence: 99%

Optimisation of GBFS, Fly Ash, and Nano-Silica Contents in Alkali-Activated Mortars

et al. 2021

View full text Add to dashboard Cite

show abstract

“…For the effective evaluation of the performance of concrete according to the advanced design technologies, its mechanical properties must be examined [14]. One of its supreme mechanical properties is its compressive strength, which is alternately the sign of structural safety throughout life [15]. This remarkable property of concrete can be affected by numerous factors, like particle size, water-tocement ratio, waste composition, and use of chemicals.…”

Section: Introductionmentioning

confidence: 99%

“…The result indicates better prediction by employing an individual algorithm based on 653 data samples. Farooq et al [ 15 ] predict the compressive nature of HSC by developing two models with random forest (RF) and GEP. RF gives a robust performance with precise correlation with strong predicted values.…”

Section: Introductionmentioning

confidence: 99%

Prediction of Compressive Strength of Fly Ash Based Concrete Using Individual and Ensemble Algorithm

et al. 2021

Self Cite

View full text Add to dashboard Cite

Machine learning techniques are widely used algorithms for predicting the mechanical properties of concrete. This study is based on the comparison of algorithms between individuals and ensemble approaches, such as bagging. Optimization for bagging is done by making 20 sub-models to depict the accurate one. Variables like cement content, fine and coarse aggregate, water, binder-to-water ratio, fly-ash, and superplasticizer are used for modeling. Model performance is evaluated by various statistical indicators like mean absolute error (MAE), mean square error (MSE), and root mean square error (RMSE). Individual algorithms show a moderate bias result. However, the ensemble model gives a better result with R2 = 0.911 compared to the decision tree (DT) and gene expression programming (GEP). K-fold cross-validation confirms the model’s accuracy and is done by R2, MAE, MSE, and RMSE. Statistical checks reveal that the decision tree with ensemble provides 25%, 121%, and 49% enhancement for errors like MAE, MSE, and RMSE between the target and outcome response.

show abstract

“…The statistical results from both models’ data clearly show that the boosting approach had fewer error values than the ANN model. The following Equations (1)–(3) in accordance with the literature [ 46 , 57 ] were used to evaluate the response of each parameter.

where: n = total number of data samples,

= reference values in the data sample,

= predicted values from models.…”

Section: Resultsmentioning

confidence: 99%

Prediction of Geopolymer Concrete Compressive Strength Using Novel Machine Learning Algorithms

Ahmad

Chaiyasarn

et al. 2021

Polymers

Self Cite

View full text Add to dashboard Cite

The innovation of geopolymer concrete (GPC) plays a vital role not only in reducing the environmental threat but also as an exceptional material for sustainable development. The application of supervised machine learning (ML) algorithms to forecast the mechanical properties of concrete also has a significant role in developing the innovative environment in the field of civil engineering. This study was based on the use of the artificial neural network (ANN), boosting, and AdaBoost ML approaches, based on the python coding to predict the compressive strength (CS) of high calcium fly-ash-based GPC. The performance comparison of both the employed techniques in terms of prediction reveals that the ensemble ML approaches, AdaBoost, and boosting were more effective than the individual ML technique (ANN). The boosting indicates the highest value of R2 equals 0.96, and AdaBoost gives 0.93, while the ANN model was less accurate, indicating the coefficient of determination value equals 0.87. The lesser values of the errors, MAE, MSE, and RMSE of the boosting technique give 1.69 MPa, 4.16 MPa, and 2.04 MPa, respectively, indicating the high accuracy of the boosting algorithm. However, the statistical check of the errors (MAE, MSE, RMSE) and k-fold cross-validation method confirms the high precision of the boosting technique. In addition, the sensitivity analysis was also introduced to evaluate the contribution level of the input parameters towards the prediction of CS of GPC. The better accuracy can be achieved by incorporating other ensemble ML techniques such as AdaBoost, bagging, and gradient boosting.

show abstract

A Comparative Study of Random Forest and Genetic Engineering Programming for the Prediction of Compressive Strength of High Strength Concrete (HSC)

Cited by 158 publications

References 56 publications

Optimisation of GBFS, Fly Ash, and Nano-Silica Contents in Alkali-Activated Mortars

Optimisation of GBFS, Fly Ash, and Nano-Silica Contents in Alkali-Activated Mortars

Prediction of Compressive Strength of Fly Ash Based Concrete Using Individual and Ensemble Algorithm

Prediction of Geopolymer Concrete Compressive Strength Using Novel Machine Learning Algorithms

Contact Info

Product

Resources

About