Compressive strength test was performed on cubic and cylindrical samples, having various sizes. It's hard to think of a single factor that adds to the strength of concrete. Adv. Build. SI is a standard error measurement, whose smaller values indicate superior model performance. In SVR, \(\{ x_{i} ,y_{i} \} ,i = 1,2,,k\) is the training set, where \(x_{i}\) and \(y_{i}\) are the input and output values, respectively. Nguyen-Sy, T. et al. Eventually, 63 mixes were omitted and 176 mixes were selected for training the models in predicting the CS of SFRC. 16, e01046 (2022). The CivilWeb Compressive Strength to Flexural Strength Conversion spreadsheet is included in the CivilWeb Flexural Strength of Concrete suite of spreadsheets. The same results are also reported by Kang et al.18. The presented paper aims to use machine learning (ML) and deep learning (DL) algorithms to predict the CS of steel fiber reinforced concrete (SFRC) incorporating hooked ISF based on the data collected from the open literature. Constr. Khan et al.55 also reported that RF (R2=0.96, RMSE=3.1) showed more acceptable outcomes than XGB and GB with, an R2 of 0.9 and 0.95 in the prediction CS of SFRC, respectively. The flexural properties and fracture performance of UHPC at low-temperature environment ( T = 20, 30, 60, 90, 120, and 160 C) were experimentally investigated in this paper. Therefore, owing to the difficulty of CS prediction through linear or nonlinear regression analysis, data-driven models are put into practice for accurate CS prediction of SFRC. 3-point bending strength test for fine ceramics that partially complies with JIS R1601 (2008) [Testing method for flexural strength of fine ceramics at room temperature] (corresponding part only). Mater. In many cases it is necessary to complete a compressive strength to flexural strength conversion. Concr. As can be seen in Fig. Moreover, among the three proposed ML models here, SVR demonstrates superior performance in estimating the influence of the W/C ratio on the predicted CS of SFRC with a correlation of R=0.999, followed by CNN with a correlation of R=0.96. Setti, F., Ezziane, K. & Setti, B. However, this parameter decreases linearly to reach a minimum value of 0.75 for concrete strength of 103 MPa (15,000 psi) or above. Moreover, it is essential to mention that only 26% of the presented mixes contained fly-ash, and the results obtained were according to these mixes. volume13, Articlenumber:3646 (2023) Your IP: 103.74.122.237, Requested URL: www.concreteconstruction.net/how-to/correlating-compressive-and-flexural-strength_o, User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/103.0.0.0 Safari/537.36. Mater. 2018, 110 (2018). The result of this analysis can be seen in Fig. 6(4) (2009). 34(13), 14261441 (2020). Finally, results from the CNN technique were consistent with the previous studies, and CNN performed efficiently in predicting the CS of SFRC. consequently, the maxmin normalization method is adopted to reshape all datasets to a range from \(0\) to \(1\) using Eq. sqrt(fck) Where, fck is the characteristic compressive strength of concrete in MPa. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Limit the search results modified within the specified time. 2, it is obvious that the CS increased with increasing the SP (R=0.792) followed by fly ash (R=0.688) and C (R=0.501). Based on the developed models to predict the CS of SFRC (Fig. Moreover, Nguyen-Sy et al.56 and Rathakrishnan et al.57, after implementing the XGB, noted that the XGB was the best model for predicting the CS of NC. The capabilities of ML algorithms were demonstrated through a sensitivity analysis and parametric analysis. Dubai, UAE Google Scholar. This is a result of the use of the linear relationship in equation 3.1 of BS EN 1996-1-1 and was taken into account in the UK calibration. As can be seen in Table 4, the performance of implemented algorithms was evaluated using various metrics. 33(3), 04019018 (2019). The air content was found to be the most significant fresh field property and has a negative correlation with both the compressive and flexural strengths. PubMed The linear relationship between two variables is stronger if \(R\) is close to+1.00 or 1.00. Mater. & Liew, K. Data-driven machine learning approach for exploring and assessing mechanical properties of carbon nanotube-reinforced cement composites. 2020, 17 (2020). Martinelli, E., Caggiano, A. Where flexural strength is critical to the design a correlation specific to the concrete mix should be developed from testing and this relationship used for the specification and quality control. STANDARDS, PRACTICES and MANUALS ON FLEXURAL STRENGTH AND COMPRESSIVE STRENGTH ACI CODE-350-20: Code Requirements for Environmental Engineering Concrete Structures (ACI 350-20) and Commentary (ACI 350R-20) ACI PRC-441.1-18: Report on Equivalent Rectangular Concrete Stress Block and Transverse Reinforcement for High-Strength Concrete Columns Lee, S.-C., Oh, J.-H. & Cho, J.-Y. Cloudflare is currently unable to resolve your requested domain. Materials 8(4), 14421458 (2015). Eng. : Investigation, Conceptualization, Methodology, Data Curation, Formal analysis, WritingOriginal Draft; N.R. Azimi-Pour, M., Eskandari-Naddaf, H. & Pakzad, A. Ati, C. D. & Karahan, O. According to the results obtained from parametric analysis, among the developed models, SVR can accurately predict the impact of W/C ratio, SP, and fly-ash on the CS of SFRC, followed by CNN. Tensile strength - UHPC has a tensile strength over 1,200 psi, while traditional concrete typically measures between 300 and 700 psi. Date:11/1/2022, Publication:IJCSM However, the understanding of ISFs influence on the compressive strength (CS) behavior of concrete is still questioned by the scientific society. Also, a significant difference between actual and predicted values was reported by Kang et al.18 in predicting the CS of SFRC (RMSE=18.024). Civ. Date:2/1/2023, Publication:Special Publication Until now, fibers have been used mainly to improve the behavior of structural elements for serviceability purposes. Res. East. Civ. Mater. Eng. CNN model is a new architecture for DL which is comprised of several layers that process and transform an input to produce an output. Build. Beyond limits of material strength, this can lead to a permanent shape change or structural failure. Internet Explorer). As per IS 456 2000, the flexural strength of the concrete can be computed by the characteristic compressive strength of the concrete. Mater. Build. Flexural strength is measured by using concrete beams. A., Hall, A., Pilon, L., Gupta, P. & Sant, G. Can the compressive strength of concrete be estimated from knowledge of the mixture proportions? Constr. As can be seen in Fig. Mansour Ghalehnovi. Han, J., Zhao, M., Chen, J. Flexural test evaluates the tensile strength of concrete indirectly. A 9(11), 15141523 (2008). fck = Characteristic Concrete Compressive Strength (Cylinder) h = Depth of Slab 232, 117266 (2020). Gler, K., zbeyaz, A., Gymen, S. & Gnaydn, O. Therefore, as can be perceived from Fig. Flexural strength calculator online - We'll provide some tips to help you select the best Flexural strength calculator online for your needs. The proposed regression equations exhibit small errors when compared to the experimental results, which allow for efficient and accurate predictions of the flexural strength. Equation(1) is the covariance between two variables (\(COV_{XY}\)) divided by their standard deviations (\(\sigma_{X}\), \(\sigma_{Y}\)). Google Scholar, Choromanska, A., Henaff, M., Mathieu, M., Arous, G. B. 2021, 117 (2021). Based on the results obtained from the implementation of SVR in predicting the CS of SFRC and outcomes from previous studies in using the SVR to predict the CS of NC and SFRC, it was concluded that in some research, SVR demonstrated acceptable performance. | Copyright ACPA, 2012, American Concrete Pavement Association (Home). Case Stud. 36(1), 305311 (2007). Fax: 1.248.848.3701, ACI Middle East Regional Office Step 1: Estimate the "s" using s = 9 percent of the flexural strength; or, call several ready mix operators to determine the value. To perform the parametric analysis to analyze the influence of one specific parameter (for example, W/C ratio) on the predicted CS of SFRC, the actual values of that parameter (W/C ratio) were considered, while the mean values for all the other input parameters values were introduced. In LOOCV, the number of folds is equal the number of instances in the dataset (n=176). From the open literature, a dataset was collected that included 176 different concrete compressive test sets. Kang et al.18 observed that KNN predicted the CS of SFRC with a great difference between actual and predicted values. SVR model (as can be seen in Fig. You do not have access to www.concreteconstruction.net. Cite this article. ADS Also, it was concluded that the W/C ratio and silica fume content had the most impact on the CS of SFRC. ; The values of concrete design compressive strength f cd are given as . & Nitesh, K. S. Study on the effect of steel and glass fibers on fresh and hardened properties of vibrated concrete and self-compacting concrete. . For materials that deform significantly but do not break, the load at yield, typically measured at 5% deformation/strain of the outer surface, is reported as the flexural strength or flexural yield strength. & Maerefat, M. S. Effects of fiber volume fraction and aspect ratio on mechanical properties of hybrid steel fiber reinforced concrete. Date:11/1/2022, Publication:Structural Journal Mech. ADS The factors affecting the flexural strength of the concrete are generally similar to those affecting the compressive strength. Duan, J., Asteris, P. G., Nguyen, H., Bui, X.-N. & Moayedi, H. A novel artificial intelligence technique to predict compressive strength of recycled aggregate concrete using ICA-XGBoost model. Where the modulus of elasticity of the concrete is required to complete a design there is a correlation equation relating flexural strength with the modulus of elasticity, shown below. On the other hand, MLR shows the highest MAE in predicting the CS of SFRC. : New insights from statistical analysis and machine learning methods. Artif. It is a measure of the maximum stress on the tension face of an unreinforced concrete beam or slab at the point of. An. This can refer to the fact that KNN considers all characteristics equally, even if they all contribute differently to the CS of concrete6. Flexural tensile strength can also be calculated from the mean tensile strength by the following expressions. Materials 13(5), 1072 (2020). Midwest, Feedback via Email Appl. This method converts the compressive strength to the Mean Axial Tensile Strength, then converts this to flexural strength and includes an adjustment for the depth of the slab. Invalid Email Address. 4: Flexural Strength Test. The focus of this paper is to present the data analysis used to correlate the point load test index (Is50) with the uniaxial compressive strength (UCS), and to propose appropriate Is50 to UCS conversion factors for different coal measure rocks. Therefore, based on tree-based technique outcomes in predicting the CS of SFRC and compatibility with previous studies in using tree-based models for predicting the CS of various concrete types (SFRC and NC), it was concluded that tree-based models (especially XGB) showed good performance. The primary sensitivity analysis is conducted to determine the most important features. Ly, H.-B., Nguyen, T.-A. Effects of steel fiber content and type on static mechanical properties of UHPCC. Technol. 313, 125437 (2021). 267, 113917 (2021). Polymers 14(15), 3065 (2022). Select Baseline, Compressive Strength, Flexural Strength, Split Tensile Strength, Modulus of Determine mathematic problem I need help determining a mathematic problem. Marcos-Meson, V. et al. Mater. Source: Beeby and Narayanan [4]. The spreadsheet is also included for free with the CivilWeb Rigid Pavement Design suite. Flexural strength may range from 10% to 15% of the compressive strength depending on the concrete mix. The CS of SFRC was predicted through various ML techniques as is described in section "Implemented algorithms". Rathakrishnan, V., Beddu, S. & Ahmed, A. N. Comparison studies between machine learning optimisation technique on predicting concrete compressive strength (2021). Deng et al.47 also observed that CNN was better at predicting the CS of recycled concrete (average relative error=3.65) than other methods. Concr. Mater. The correlation of all parameters with each other (pairwise correlation) can be seen in Fig. Mater. This method converts the compressive strength to the Mean Axial Tensile Strength, then converts this to flexural strength and includes an adjustment for the depth of the slab. The stress block parameter 1 proposed by Mertol et al. Moreover, some others were omitted because of lacking the information of mixing components (such as FA, SP, etc.). Date:4/22/2021, Publication:Special Publication 1.1 This test method provides guidelines for testing the flexural strength of cured geosynthetic cementitious composite mat (GCCM) products in a three (3)-point bend apparatus. Fluctuations of errors (Actual CSpredicted CS) for different algorithms. Google Scholar. Phone: 1.248.848.3800, Home > Topics in Concrete > topicdetail, View all Documents on flexural strength and compressive strength , Publication:Materials Journal In contrast, others reported that SVR showed weak performance in predicting the CS of concrete. Gupta, S. Support vector machines based modelling of concrete strength. The test jig used in this video has a scale on the receiver, and the distance between the external fulcrums (distance between the two outer fulcrums . Caggiano, A., Folino, P., Lima, C., Martinelli, E. & Pepe, M. On the mechanical response of hybrid fiber reinforced concrete with recycled and industrial steel fibers. Compressive strength result was inversely to crack resistance. Is there such an equation, and, if so, how can I get a copy? Explain mathematic . InInternational Conference on Applied Computing to Support Industry: Innovation and Technology 323335 (Springer, 2019). Mater. The sensitivity analysis demonstrated that, among different input variables, W/C ratio, fly ash, and SP had the most contributing effect on the CS behavior of SFRC, followed by the amount of ISF. Res. Regarding Fig. Area and Volume Calculator; Concrete Mixture Proportioner (iPhone) Concrete Mixture Proportioner (iPad) Evaporation Rate Calculator; Joint Noise Estimator; Maximum Joint Spacing Calculator J. Comput. 12 illustrates the impact of SP on the predicted CS of SFRC. 6(5), 1824 (2010). This can be due to the difference in the number of input parameters. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. Eur. the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in October 18, 2022. It concluded that the addition of banana trunk fiber could reduce compressive strength, but could raise the concrete ability in crack resistance Keywords: Concrete . Table 3 displays the modified hyperparameters of each convolutional, flatten, hidden, and pooling layer, including kernel and filter size and learning rate. Also, the CS of SFRC was considered as the only output parameter. Deng, F. et al. The user accepts ALL responsibility for decisions made as a result of the use of this design tool. What factors affect the concrete strength? Adv. 12, the SP has a medium impact on the predicted CS of SFRC. Performance comparison of SVM and ANN in predicting compressive strength of concrete (2014). Statistical characteristics of input parameters, including the minimum, maximum, average, and standard deviation (SD) values of each parameter, can be observed in Table 1. KNN (R2=0.881, RMSE=6.477, MAE=4.648) showed lower accuracy compared with MLR in predicting the CS of SFRC. Flexural strength is an indirect measure of the tensile strength of concrete. 324, 126592 (2022). In other words, the predicted CS decreases as the W/C ratio increases. Karahan et al.58 implemented ANN with the LevenbergMarquardt variant as the backpropagation learning algorithm and reported that ANN predicted the CS of SFRC accurately (R2=0.96). : Validation, WritingReview & Editing. Date:7/1/2022, Publication:Special Publication Normal distribution of errors (Actual CSPredicted CS) for different methods. Table 3 provides the detailed information on the tuned hyperparameters of each model. Despite the enhancement of CS of normal strength concrete incorporating ISF, no significant change of CS is obtained for high-performance concrete mixes by increasing VISF14,15. (3): where \(\hat{y}\), \(x_{n}\), and \(\alpha\) are the dependent parameter, independent parameter, and bias, respectively18. Limit the search results with the specified tags. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. Email Address is required Also, a specific type of cross-validation (CV) algorithm named LOOCV (Fig. Comparing ML models with regard to MAE and MAPE, it is seen that CNN performs superior in predicting the CS of SFRC, followed by GB and XGB. To try out a fully functional free trail version of this software, please enter your email address below to sign up to our newsletter. All these mixes had some features such as DMAX, the amount of ISF (ISF), L/DISF, C, W/C ratio, coarse aggregate (CA), FA, SP, and fly ash as input parameters (9 features). For CEM 1 type cements a very general relationship has often been applied; This provides only the most basic correlation between flexural strength and compressive strength and should not be used for design purposes. A parametric analysis was carried out to determine how well the developed ML algorithms can predict the effect of various input parameters on the CS behavior of SFRC. For example compressive strength of M20concrete is 20MPa. However, regarding the Tstat, the outcomes show that CNN performance was approximately 58% lower than XGB. Anyone you share the following link with will be able to read this content: Sorry, a shareable link is not currently available for this article. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. Convert. Please enter search criteria and search again, Informational Resources on flexural strength and compressive strength, Web Pages on flexural strength and compressive strength, FREQUENTLY ASKED QUESTIONS ON FLEXURAL STRENGTH AND COMPRESSIVE STRENGTH. Flexural strength is about 10 to 15 percent of compressive strength depending on the mixture proportions and type, size and volume of coarse aggregate used. Constr. Second Floor, Office #207 However, it is depicted that the weak correlation between the amount of ISF in the SFRC mix and the predicted CS. The impact of the fly-ash on the predicted CS of SFRC can be seen in Fig. Struct. Compressive strength, Flexural strength, Regression Equation I. The least contributing factors include the maximum size of aggregates (Dmax) and the length-to-diameter ratio of hooked ISFs (L/DISF). MathSciNet Build. Flexural strength of concrete = 0.7 . Hence, various types of fibers are added to increase the tensile load-bearing capability of concrete. Also, the characteristics of ISF (VISF, L/DISF) have a minor effect on the CS of SFRC. ML is a computational technique destined to simulate human intelligence and speed up the computing procedure by means of continuous learning and evolution. Unquestionably, one of the barriers preventing the use of fibers in structural applications has been the difficulty in calculating the FRC properties (especially CS behavior) that should be included in current design techniques10. Google Scholar. Southern California Xiamen Hongcheng Insulating Material Co., Ltd. View Contact Details: Product List: 73, 771780 (2014). The flexural strengths of all the laminates tested are significantly higher than their tensile strengths, and are also higher than or similar to their compressive strengths. Hameed et al.52 developed an MLR model to predict the CS of high-performance concrete (HPC) and noted that MLR had a poor correlation between the actual and predicted CS of HPC (R=0.789, RMSE=8.288). 9, the minimum and maximum interquartile ranges (IQRs) belong to AdaBoost and MLR, respectively. Scientific Reports Properties of steel fiber reinforced fly ash concrete. This is particularly common in the design and specification of concrete pavements where flexural strengths are critical while compressive strengths are often specified. This highlights the role of other mixs components (like W/C ratio, aggregate size, and cement content) on CS behavior of SFRC. Development of deep neural network model to predict the compressive strength of rubber concrete. In addition, the studies based on ML techniques that have been done to predict the CS of SFRC are limited since it is difficult to collect inclusive experimental data to develop models regarding all contributing features (such as the properties of fibers, aggregates, and admixtures). Meanwhile, the CS of SFRC could be enhanced by increasing the amount of superplasticizer (SP), fly ash, and cement (C). Caution should always be exercised when using general correlations such as these for design work. Whereas, it decreased by increasing the W/C ratio (R=0.786) followed by FA (R=0.521). Based on this, CNN had the closest distribution to the normal distribution and produced the best results for predicting the CS of SFRC, followed by SVR and RF. Comparing implemented ML algorithms in terms of Tstat, it is observed that XGB shows the best performance, followed by ANN and SVR in predicting the CS of SFRC. Hu, H., Papastergiou, P., Angelakopoulos, H., Guadagnini, M. & Pilakoutas, K. Mechanical properties of SFRC using blended manufactured and recycled tyre steel fibres. Build. Schapire, R. E. Explaining adaboost. However, their performance in predicting the CS of SFRC was superior to that of KNN and MLR. Technol. The minimum performance requirements of each GCCM Classification Type have been defined within ASTM D8364, defining the appropriate GCCM specific test standards to use, such as: ASTM D8329 for compressive strength and ASTM D8058 for flexural strength. Mater. Constr. Scientific Reports (Sci Rep) The flexural strength is the strength of a material in bending where the top surface is tension and the bottom surface. In these cases, an SVR with a non-linear kernel (e.g., a radial basis function) is used. On the other hand, K-nearest neighbor (KNN) algorithm with R2=0.881, RMSE=6.477, and MAE=4.648 results in the weakest performance. [1] ; Compressive Strength - UHPC's advanced compressive strength is particularly significant when . You are using a browser version with limited support for CSS. 147, 286295 (2017). http://creativecommons.org/licenses/by/4.0/. Comput. The use of an ANN algorithm (Fig. Plus 135(8), 682 (2020). Flexural strength, also known as modulus of rupture, or bend strength, or transverse rupture strengthis a material property, defined as the stressin a material just before it yieldsin a flexure test. The result of compressive strength for sample 3 was 105 Mpa, for sample 2 was 164 Mpa and for sample 1 was 320 Mpa. Where as, Flexural strength is the behaviour of a structure in direct bending (like in beams, slabs, etc.) Distributions of errors in MPa (Actual CSPredicted CS) for several methods. Sign up for the Nature Briefing newsletter what matters in science, free to your inbox daily. In Artificial Intelligence and Statistics 192204. Mater. Dumping massive quantities of waste in a non-eco-friendly manner is a key concern for developing nations. However, there are certain commonalities: Types of cement that may be used Cement quantity, quality, and brand Setti et al.12 also introduced ISF with different volume fractions (VISF) to the concrete and reported the improvement of CS of SFRC by increasing the content of ISF. Constr. Article PubMed & Arashpour, M. Predicting the compressive strength of normal and High-Performance Concretes using ANN and ANFIS hybridized with Grey Wolf Optimizer. 5(7), 113 (2021). The findings show that up to a certain point, adding both HS and SF increases the compressive, tensile, and flexural strength of concrete at all curing ages. Date:9/30/2022, Publication:Materials Journal Recently, ML algorithms have been widely used to predict the CS of concrete. Founded in 1904 and headquartered in Farmington Hills, Michigan, USA, the American Concrete Institute is a leading authority and resource worldwide for the development, dissemination, and adoption of its consensus-based standards, technical resources, educational programs, and proven expertise for individuals and organizations involved in concrete design, construction, and materials, who share a commitment to pursuing the best use of concrete. Experimental study on bond behavior in fiber-reinforced concrete with low content of recycled steel fiber. The flexural strength of UD, CP, and AP laminates was increased by 39-53%, 51-57%, and 25-37% with the addition of 0.1-0.2% MWCNTs. Also, to prevent overfitting, the leave-one-out cross-validation method (LOOCV) is implemented, and 8 different metrics are used to assess the efficiency of developed models. The analyses of this investigation were focused on conversion factors for compressive strengths of different samples. 11(4), 1687814019842423 (2019). 161, 141155 (2018). INTRODUCTION The strength characteristic and economic advantages of fiber reinforced concrete far more appreciable compared to plain concrete. These equations are shown below. Terms of Use The user accepts ALL responsibility for decisions made as a result of the use of this design tool. The CivilWeb Flexural Strength of Concrete suite of spreadsheets is available for purchase at the bottom of this page for only 5. Bending occurs due to development of tensile force on tension side of the structure. & Hawileh, R. A. Date:9/1/2022, Search all Articles on flexural strength and compressive strength », Publication:Concrete International In addition, CNN achieved about 28% lower residual error fluctuation than SVR. Al-Abdaly et al.50 reported that MLR algorithm (with R2=0.64, RMSE=8.68, MAE=5.66) performed poorly in predicting the CS behavior of SFRC. 248, 118676 (2020). A good rule-of-thumb (as used in the ACI Code) is: In contrast, KNN shows the worst performance among developed ML models in predicting the CS of SFRC. Pakzad, S.S., Roshan, N. & Ghalehnovi, M. Comparison of various machine learning algorithms used for compressive strength prediction of steel fiber-reinforced concrete. It was observed that among the concrete mixture properties, W/C ratio, fly-ash, and SP had the most significant effect on the CS of SFRC (W/C ratio was the most effective parameter). Skaryski, & Suchorzewski, J. Question: Are there data relating w/cm to flexural strength that are as reliable as those for compressive View all Frequently Asked Questions on flexural strength and compressive strength», View all flexural strength and compressive strength Events , The Concrete Industry in the Era of Artificial Intelligence, There are no Committees on flexural strength and compressive strength, Concrete Laboratory Testing Technician - Level 1. Constr. Cem. 26(7), 16891697 (2013). One of the drawbacks of concrete as a fragile material is its low tensile strength and strain capacity. Intersect. However, the addition of ISF into the concrete and producing the SFRC may also provide additional strength capacity or act as the primary reinforcement in structural elements. Based upon the results in this study, tree-based models performed worse than SVR in predicting the CS of SFRC. Constr. Kabiru, O. Mater. Constr. Compared to the previous ML algorithms (MLR and KNN), SVRs performance was better (R2=0.918, RMSE=5.397, MAE=4.559). Chou, J.-S. & Pham, A.-D. MLR predicts the value of the dependent variable (\(y\)) based on the value of the independent variable (\(x\)) by establishing the linear relationship between inputs (independent parameters) and output (dependent parameter) based on Eq.