Hybrid machine learning approach to prediction of the compressive and flexural strengths of UHPC and parametric analysis with shapley additive explanations

Case Studies in Construction Materials, Journal Year: 2023, Volume and Issue: 20, P. e02723 - e02723

Published: Nov. 28, 2023

Ultra-high-performance concrete (UHPC) is a sustainable construction material; it can be applied as substitute for cement concrete. Artificial intelligence methods have been used to evaluate composites reduce time and money in the industries. So, this study machine learning (ML) hybrid ML approaches predict compressive flexural strength of UHPC. A dataset 626 317 data points was collected from published research articles, where fourteen important variables were selected input parameters analysis algorithms. This XGBoost, LightGBM, XGBoost- LightGBM algorithms UHPC materials. Grid search (GS) techniques adjust model hyper-parameters improved high accuracy efficiency. models train, test stage utilized statistical assessments such R-square, root mean square error (RMSE), absolute (MAE), coefficient efficiency (CE). The results presented algorithm superior XGBoost terms R-square RMSE values both prediction. two showed CS considerable above 0.94 at testing stages just over 0.97 training phase. Hybrid performance prediction value found that almost 0.996 0.963 phases. At same time, FS result traditional founded 0.95 phase around 0.81 But among them, XGB-LGB lowest trained its hyperparameters optimized. Additionally, SHAP investigation reveals impact relationship with output variables. outcome curing age steel fiber content parameter had highest positive on

Language: Английский

---

Machine learning and interactive GUI for concrete compressive strength prediction

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: July 19, 2024

Concrete compressive strength (CS) is a crucial performance parameter in concrete structure design. Reliable prediction reduces costs and time design prevents material waste from extensive mixture trials. Machine learning techniques solve structural engineering challenges such as CS prediction. This study used Learning (ML) models to enhance the of CS, analyzing 1030 experimental data ranging 2.33 82.60 MPa previous research databases. The ML included both non-ensemble ensemble types. were regression-based, evolutionary, neural network, fuzzy-inference-system. Meanwhile, consisted adaptive boosting, random forest, gradient boosting. There eight input parameters: cement, blast-furnace-slag, aggregates (coarse fine), fly ash, water, superplasticizer, curing days, with output. Comprehensive evaluations include visual quantitative methods k-fold cross-validation assess study's reliability accuracy. A sensitivity analysis using Shapley-Additive-exPlanations (SHAP) was conducted understand better how each variable affects CS. findings showed that Categorical-Gradient-Boosting (CatBoost) model most accurate during testing stage. It had highest determination-coefficient (R

Language: Английский

---

Compressive strength prediction of sustainable concrete incorporating rice husk ash (RHA) using hybrid machine learning algorithms and parametric analyses

Case Studies in Construction Materials, Journal Year: 2024, Volume and Issue: 20, P. e03030 - e03030

Published: March 5, 2024

The construction industry is making efforts to reduce the environmental impact of cement production in concrete by incorporating alternative and supplementary cementitious materials, as well lowering carbon emissions. One such material that has gained popularity this context rice husk ash (RHA) due its pozzolanic reactions. This study aims forecast compressive strength (CS) RHA-based (RBC) examining effects several factors cement, RHA content, curing age, water usage, aggregate amount, superplasticizer content. To accomplish this, collected analyzed data from literature, resulting a dataset 1404 observations. Several machine learning (ML) models, light gradient boosting (LGB), extreme (XGB), random forest (RF), hybrid (HML) approaches like XGB-LGB XGB-RF were employed thoroughly analyze these parameters assess their on strength. was split into training testing groups, statistical analyses performed determine relationships between input CS. Moreover, performance all models evaluated using various evaluation criteria, including mean absolute percentage error (MAPE), coefficient efficiency (CE), root square (RMSE), determination (R2). model found have higher precision (R2 = 0.95, RMSE 5.255 MPa) compared other models. SHAP (SHapley Additive exPlanations) analysis revealed RHA, had positive effect Overall, study's findings suggest with identified can be used accurately predict CS RBC. application technologies sector facilitate rapid low-cost identification qualities parameters.

Language: Английский

---

A comparative study of machine learning models for construction costs prediction with natural gradient boosting algorithm and SHAP analysis

Asian Journal of Civil Engineering, Journal Year: 2024, Volume and Issue: 25(4), P. 3301 - 3316

Published: Feb. 9, 2024

Language: Английский

---

Improved forecasting of the compressive strength of ultra‐high‐performance concrete (UHPC) via the CatBoost model optimized with different algorithms

Structural Concrete, Journal Year: 2024, Volume and Issue: unknown

Published: May 19, 2024

Abstract This paper focuses on the applicability of CatBoost models constructed using various optimization techniques for improved forecasting compressive strength ultra‐high‐performance concrete (UHPC). Phasor particle swarm (PPSO), dwarf mongoose (DMO), and atom search (ASO), which have been very popular recently, are preferred as algorithms. A comprehensive reliable data set is used to develop models, include 785 test results with 15 input features. The performance (PPSO‐CatBoost, DMO‐CatBoost, ASO‐CatBoost) optimized different algorithms thoroughly assessed by means statistical metrics error analysis determine model best capability, this compared obtained from previous studies. In addition, Shapley additive exPlanations (SHAP) ensure interpretability overcome “black box” problem machine learning (ML) models. demonstrate that all outstandingly forecast UHPC. Among these DMO‐CatBoost stands out other in metrics, such high coefficient determination ( R 2 ) values, low root mean squared (RMSE), absolute percentage (MAPE), (MAE) along a smaller ratio. words, RMSE, , MAPE, MAE values training 3.67, 0.993, 0.019, 2.35, respectively, whereas those 6.15, 0.978, 0.038, 4.51. Additionally, ranking optimize hyperparameters follows: DMO > PPSO ASO. On hand, SHAP showed age, fiber dosage, cement dosage significantly influence These findings can guide structural engineers design UHPC, thus assisting them developing strategies improve properties material. Finally, based developed work, graphical user interface has easily UHPC practical applications without additional tools or software.

Language: Английский

---

Shear behavior of reinforced concrete beams comprising a combination of crumb rubber and rice husk ash

Engineering Structures, Journal Year: 2024, Volume and Issue: 319, P. 118862 - 118862

Published: Sept. 1, 2024

Language: Английский

---

Prediction model for compressive strength of rice husk ash blended sandcrete blocks using a machine learning models

Asian Journal of Civil Engineering, Journal Year: 2024, Volume and Issue: 25(6), P. 4745 - 4758

Published: May 24, 2024

Language: Английский

---

Tree-based machine learning models for predicting the bond strength in reinforced recycled aggregate concrete

Asian Journal of Civil Engineering, Journal Year: 2024, Volume and Issue: unknown

Published: Sept. 2, 2024

Language: Английский

---

Prediction of compressive strength of high-performance concrete using optimization machine learning approaches with SHAP analysis

Journal of Building Pathology and Rehabilitation, Journal Year: 2024, Volume and Issue: 9(2)

Published: May 24, 2024

Language: Английский

---

Compressive Strength Prediction of Basalt Fiber Reinforced Concrete Based on Interpretive Machine Learning Using SHAP Analysis

Iranian Journal of Science and Technology Transactions of Civil Engineering, Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 28, 2024

Language: Английский

---