A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Prognosis of flow of fly ash and blast furnace slag-based concrete: leveraging advanced machine learning algorithms
In the field of construction, the workability of concrete, specifically its ability to flow, is one of the most concerned parameters. In recent times, the integration of artificial intelligence (AI) has brought about a significant transformation in the construction industry, resulting in enhanced efficiency, precision, and innovation. Considering these aspects, the present study has been carried out on a large dataset comprising 1103 data points while taking the ten input parameters into account to predict the flow of concrete. In this regard, six distinct models such as multilayer perceptron (MLP), K-nearest neighbors (KNN), gradient boosting (GB), M5P regression, backpropagation neural networks (BPNN), and lasso regressor have been used to forecast the flow. Along with that, various visualization and evaluation techniques, including scatter plots, histograms, heatmap, coefficient of correlation, errors, SHAP, Taylor’s diagram, have been utilized to illustrate the data availability and performance of models. Based on the output of the study, it has been noticed that the KNN, M5P, and GB models demonstrated exceptional accuracy with negligible errors and high R-squared values (R2 ≤ 0.98), whereas other models encountered difficulties in achieving satisfactory performance. This study highlights the significance of water content, coarse aggregates, and fine aggregates as crucial factors that directly affect the flow characteristics of concrete.
Prognosis of flow of fly ash and blast furnace slag-based concrete: leveraging advanced machine learning algorithms
In the field of construction, the workability of concrete, specifically its ability to flow, is one of the most concerned parameters. In recent times, the integration of artificial intelligence (AI) has brought about a significant transformation in the construction industry, resulting in enhanced efficiency, precision, and innovation. Considering these aspects, the present study has been carried out on a large dataset comprising 1103 data points while taking the ten input parameters into account to predict the flow of concrete. In this regard, six distinct models such as multilayer perceptron (MLP), K-nearest neighbors (KNN), gradient boosting (GB), M5P regression, backpropagation neural networks (BPNN), and lasso regressor have been used to forecast the flow. Along with that, various visualization and evaluation techniques, including scatter plots, histograms, heatmap, coefficient of correlation, errors, SHAP, Taylor’s diagram, have been utilized to illustrate the data availability and performance of models. Based on the output of the study, it has been noticed that the KNN, M5P, and GB models demonstrated exceptional accuracy with negligible errors and high R-squared values (R2 ≤ 0.98), whereas other models encountered difficulties in achieving satisfactory performance. This study highlights the significance of water content, coarse aggregates, and fine aggregates as crucial factors that directly affect the flow characteristics of concrete.
Prognosis of flow of fly ash and blast furnace slag-based concrete: leveraging advanced machine learning algorithms
Asian J Civ Eng
Kumar, Rahul (author) / Rathore, Ayush (author) / Singh, Rajwinder (author) / Mir, Ajaz Ahmad (author) / Tipu, Rupesh Kumar (author) / Patel, Mahesh (author)
Asian Journal of Civil Engineering ; 25 ; 2483-2497
2024-04-01
15 pages
Article (Journal)
Electronic Resource
English
Blast furnace slag as concrete aggregate
Engineering Index Backfile | 1931
Blast furnace Slag as Concrete Aggregate
Engineering Index Backfile | 1930
Blast furnace slag as concrete aggregate
Engineering Index Backfile | 1911
Blast furnace slag as concrete aggregate
Engineering Index Backfile | 1931