Advanced and hybrid machine learning techniques for predicting compressive strength in palm oil fuel ash-modified concrete with SHAP analysis DOI Creative Commons

Tariq Ali,

Kennedy C. Onyelowe, Muhammad Sarmad Mahmood

et al.

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Feb. 10, 2025

The increasing demand for sustainable construction materials has led to the incorporation of Palm Oil Fuel Ash (POFA) into concrete reduce cement consumption and lower CO₂ emissions. However, predicting compressive strength (CS) POFA-based remains challenging due variability input factors. This study addresses this issue by applying advanced machine learning models forecast CS POFA-incorporated concrete. A dataset 407 samples was collected, including six parameters: content, POFA dosage, water-to-binder ratio, aggregate superplasticizer curing age. divided 70% training 30% testing. evaluated include Hybrid XGB-LGBM, ANN, Bagging, LSSVM, GEP, XGB LGBM. performance these assessed using key metrics, coefficient determination (R2), root mean square error (RMSE), normalized means (NRMSE), absolute (MAE) Willmott index (d). XGB-LGBM model achieved maximum R2 0.976 lowest RMSE, demonstrating superior accuracy, followed ANN with an 0.968. SHAP analysis further validated identifying most impactful factors, ratio emerging as influential. These predictive offer industry a reliable framework evaluating concrete, reducing need extensive experimental testing, promoting development more eco-friendly, cost-effective building materials.

Language: Английский

Comparative Analysis of Machine Learning Models for Predicting the Compressive Strength of Ultra-High-Performance Steel Fiber Reinforced Concrete DOI Creative Commons
Md Sohel Rana,

Md Minaz Hossain,

Fangyuan Li

et al.

Journal of Engineering Research, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 1, 2025

Language: Английский

Citations

1
Microstructural assessment and supervised machine learning-aided modeling to explore the potential of quartz powder as an alternate binding material in concrete DOI Creative Commons
Md. Habibur Rahman Sobuz,

Md. Kawsarul Islam Kabbo,

M.R. Khatun

et al.

Case Studies in Construction Materials, Journal Year: 2025, Volume and Issue: unknown, P. e04568 - e04568

Published: March 1, 2025

Language: Английский

Citations

1
Multi-performance collaborative optimization of existing residential building retrofitting in extremely arid and hot climate zone: A case study in Turpan, China DOI

Guangchao Shi,

Shanshan Yao,

Junkang Song

et al.

Journal of Building Engineering, Journal Year: 2024, Volume and Issue: 89, P. 109304 - 109304

Published: April 16, 2024

Language: Английский

Citations

8
Forecasting the strength of preplaced aggregate concrete using interpretable machine learning approaches DOI Creative Commons
Muhammad Faisal Javed, Muhammad Fawad, Rida Hameed Lodhi

et al.

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

Published: April 10, 2024

Abstract Preplaced aggregate concrete (PAC) also known as two-stage (TSC) is widely used in construction engineering for various applications. To produce PAC, a mixture of Portland cement, sand, and admixtures injected into mold subsequent to the deposition coarse aggregate. This process complicates prediction compressive strength (CS), demanding thorough investigation. Consequently, emphasis this study on enhancing comprehension PAC using machine learning models. Thirteen models are evaluated with 261 data points eleven input variables. The result depicts that xgboost demonstrates exceptional accuracy correlation coefficient 0.9791 normalized determination (R 2 ) 0.9583. Moreover, Gradient boosting (GB) Cat boost (CB) perform well due its robust performance. In addition, Adaboost, Voting regressor, Random forest yield precise predictions low mean absolute error (MAE) root square (RMSE) values. sensitivity analysis (SA) reveals significant impact key parameters overall model sensitivity. Notably, gravel takes lead substantial 44.7% contribution, followed by sand at 19.5%, cement 15.6%, Fly ash GGBS 5.9% 5.1%, respectively. best fit i.e., XG-Boost model, was employed SHAP assess relative importance contributing attributes optimize unveiled water-to-binder (W/B) ratio, superplasticizer, most factors influencing CS PAC. Furthermore, graphical user interface (GUI) have been developed practical applications predicting strength. simplifies offers valuable tool leveraging model's potential field civil engineering. comprehensive evaluation provides insights researchers practitioners, empowering them make informed choices projects. By reliability applicability predictive models, contributes preplaced prediction.

Language: Английский

Citations

7
Intelligent multi-objective optimization of 3D printing low-carbon concrete for multi-scenario requirements DOI

Songyuan Geng,

Qiling Luo,

Boyuan Cheng

et al.

Journal of Cleaner Production, Journal Year: 2024, Volume and Issue: 445, P. 141361 - 141361

Published: Feb. 20, 2024

Language: Английский

Citations

6
A novel compressive strength estimation approach for 3D printed fiber-reinforced concrete: integrating machine learning and gene expression programming DOI
Md Nasir Uddin, Junhong Ye, M. Aminul Haque

et al.

Multiscale and Multidisciplinary Modeling Experiments and Design, Journal Year: 2024, Volume and Issue: 7(5), P. 4889 - 4910

Published: April 15, 2024

Language: Английский

Citations

5
Performance evaluation of conductive materials in conductive mortar based on machine learning DOI
Shuxian Hong,

Jie Wu,

Biqin Dong

et al.

Journal of Building Engineering, Journal Year: 2024, Volume and Issue: 92, P. 109695 - 109695

Published: May 24, 2024

Language: Английский

Citations

5
Advancing mix design prediction in 3D printed concrete: Predicting anisotropic compressive strength and slump flow DOI Creative Commons
Umair Jalil Malik, Raja Dilawar Riaz, Saif Ur Rehman

et al.

Case Studies in Construction Materials, Journal Year: 2024, Volume and Issue: 21, P. e03510 - e03510

Published: July 11, 2024

Introducing 3D-concrete printing has started a revolution in the construction industry, presenting unique opportunities alongside undeniable challenges. Among these, major challenge is iterative process associated with mix design formulation, which results significant material and time consumption. This research uses machine learning (ML) techniques such as Extreme Gradient Boosting (XGBoost), Support Vector Machine (SVM), Decision Tree Regression (DTR), Gaussian Process (GPR), Artificial Neural Network (ANN) to overcome these A dataset containing 21 constituent features 4 output properties (cast printed compressive strength, slump flow) was extracted from literature investigate relationship between performance. The models were assessed using range of evaluation metrics, including Mean Absolute Error (MAE), Root Squared (RMSE), (MSE), R-squared value. (GPR) yielded more favorable results. In case cast GPR achieved an R2 value 0.9069, along RMSE, MSE, MAE values 13.04, 170.12, 9.40, respectively. similar trend observed for strengths directions 1, 2, 3. exceeding 0.91 all directions, accompanied by significantly lower RMSE (below 4.1). also validated four designs. These mixes 3D tested strength flow. GPR's average error 10.55 %, while SVM slightly 9.38 %. Overall, this work presents novel approach optimizing 3D-printed concrete enabling prediction flow directly design. can facilitate fabrication structures that fulfill necessary printability requirements.

Language: Английский

Citations

5
Study on printability of 3D printing carbon fiber reinforced eco-friendly concrete: Characterized by fluidity and consistency DOI Creative Commons

Wen Xu,

Dengjie Jiang,

Qian Zhao

et al.

Case Studies in Construction Materials, Journal Year: 2024, Volume and Issue: 21, P. e03589 - e03589

Published: July 30, 2024

Carbon fibers have often been added to concrete as reinforcement. Eco-friendly with industrial by-products has widely studied and applied green building materials. Studying the workability printability of eco-friendly carbon is worthwhile. The are dominating factors that ensure printing can be carried out smoothly. This study uses a combination experiments numerical simulations performance fiber-reinforced (CFREFC). 9 mixes 3D CFREFC under various combinations different water-binder (w/b) ratio levels superplasticizer (SP) dosages were tested. Two methods, namely consistency fluidity tests, used characterize printability. After mortar mixtures printed get performance. Finally, relationship between was established. condition numbered M7 (w/b = 0.4, SP 0.5) in selected experimental group source its simulation parameter. result shows it feasible using workability, i.e., fluidity, which increase w/b dosage. Under parameters HC1008 printer determined, 20 mm nozzle size, 50 mm/s speed, 30 rpm material extrusion 14 layer height, fresh not suitable for 3DP application when less than 48.99 more 81.96 mm, or 166.72 200.93 mm. When from 56.34 65.61 172.18 183.30 best same parameters. results indicated increased number layers, bottom model would deformed by gradual pressure, specific height loss occur, consistent results.

Language: Английский

Citations

5
Machine learning models for estimating the compressive strength of rubberized concrete subjected to elevated temperature: Optimization and hyper-tuning DOI
Turki S. Alahmari, Irfan Ullah, Furqan Farooq

et al.

Sustainable Chemistry and Pharmacy, Journal Year: 2024, Volume and Issue: 42, P. 101763 - 101763

Published: Sept. 3, 2024

Language: Английский

Citations

5