Development of artificial neural network and multiple linear regression models in the prediction process of the hot mix asphalt properties

Androjić, Ivica; Marović, Ivan

doi:10.1139/cjce-2017-0300

Cited by 20 publications

(17 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although the mechanistic constitutive methods allow a rational and in-depth analysis of the material response to be performed [10][11][12][13][14][15][16][17], statistical or machine learning approaches are gaining considerable success in the academic community due to the good reliability of their predictions [18][19][20][21], even if they are not physically based. Nevertheless, it has also been reported that statistical regressions of experimental data can produce less accurate predictions than machine learning methods, specifically with regard to Artificial Neural Networks (ANNs) [22][23][24][25][26][27]. Recently, Lam et al [28] have found that a multiple regression analysis (coefficient of determination, R 2 , equal to 0.790) is less reliable than an ANN approach (R 2 = 0.925) searching the analytical model to infer compressive strength of roller-compacted concrete pavement from steel slags aggregate and fly ash levels replacing cement.…”

Section: Introductionmentioning

confidence: 99%

Bituminous Mixtures Experimental Data Modeling Using a Hyperparameters-Optimized Machine Learning Approach

et al. 2021

View full text Add to dashboard Cite

This study introduces a machine learning approach based on Artificial Neural Networks (ANNs) for the prediction of Marshall test results, stiffness modulus and air voids data of different bituminous mixtures for road pavements. A novel approach for an objective and semi-automatic identification of the optimal ANN’s structure, defined by the so-called hyperparameters, has been introduced and discussed. Mechanical and volumetric data were obtained by conducting laboratory tests on 320 Marshall specimens, and the results were used to train the neural network. The k-fold Cross Validation method has been used for partitioning the available data set, to obtain an unbiased evaluation of the model predictive error. The ANN’s hyperparameters have been optimized using the Bayesian optimization, that overcame efficiently the more costly trial-and-error procedure and automated the hyperparameters tuning. The proposed ANN model is characterized by a Pearson coefficient value of 0.868.

show abstract

Section: Introductionmentioning

confidence: 99%

Bituminous Mixtures Experimental Data Modeling Using a Hyperparameters-Optimized Machine Learning Approach

et al. 2021

View full text Add to dashboard Cite

show abstract

“…In another study, the k-nearest neighbor model—an effective machine learning algorithm—was applied to predict the moisture susceptibility of HMA [ 14 ], which suggested that this machine learning model should be used for accurate prediction of the vital properties of the HMA. Androjić et al [ 15 ] studied and compared two popular machine learning models namely ANN and multiple linear regression (MLR) for prediction of different HMA properties (air void and soluble binder content), and concluded that both applied machine learning models are suitable for prediction of the HMA properties. However, ANN is better than MLR at predicting the air void content, whereas MLR is better than ANN in predicting soluble binder content.…”

Section: Introductionmentioning

confidence: 99%

Cost-Effective Approaches Based on Machine Learning to Predict Dynamic Modulus of Warm Mix Asphalt with High Reclaimed Asphalt Pavement

Dao

Nguyen

et al. 2020

Materials

View full text Add to dashboard Cite

Warm mix asphalt (WMA) technology, taking advantage of reclaimed asphalt pavements, has gained increasing attention from the scientific community. The determination of technical specifications of such a type of asphalt concrete is crucial for pavement design, in which the asphalt concrete dynamic modulus (E*) of elasticity is amongst the most critical parameters. However, the latter could only be determined by complicated, costly, and time-consuming experiments. This paper presents an alternative cost-effective approach to determine the dynamic elastic modulus (E*) of WMA based on various machine learning-based algorithms, namely the artificial neural network (ANN), support vector machine (SVM), Gaussian process regression (GPR), and ensemble boosted trees (Boosted). For this, a total of 300 samples were fabricated by warm mix asphalt technology. The mixtures were prepared with 0%, 20%, 30%, 40%, and 50% content of reclaimed asphalt pavement (RAP) and modified bitumen binder using Sasobit and Zycotherm additives. The dynamic elastic modulus tests were conducted by varying the temperature from 10 °C to 50 °C at different frequencies from 0.1 Hz to 25 Hz. Various common quantitative indications, such as root mean square error (RMSE), mean absolute error (MAE), and correlation coefficient (R) were used to validate and compare the prediction capability of different models. The results showed that machine learning models could accurately predict the dynamic elastic modulus of WMA using up to 50% RAP and fabricated by warm mix asphalt technology. Out of these models, the Boosted algorithm (R = 0.9956) was found as the best predictor compared with those obtained by ANN-LMN (R = 0.9954), SVM (R = 0.9654), and GPR (R= 0.9865). Thus, it could be concluded that Boosted is a promising cost-effective tool for the prediction of the dynamic elastic modulus (E*) of WMA. This study might help in reducing the cost of laboratory experiments for the determination of the dynamic modulus (E*).

show abstract

“…Researches such as Loia et al (2000), Quintero et al (2005), Šelih et al (2008), and Jajac et al (2009) can be highlighted as the one where ANNs were used for solving problems on both tactical and strategic management levels based upon Turban's DSS concept. Such approach bloomed in 2010s when more and more researchers very efficiently applied ANNs for solving structured problems that occur on operational level (Gesoglu et al, 2010;Xiao et al, 2010;Ozgan, 2011;Ciresan et al, 2012;Singh et al, 2013;Ozturk and Kutay, 2014;Zavratnik et al, 2016;Androji and Marovi , 2017;Zhang et al, 2018;Gong et al, 2019) as well as for solving semi-structured and unstructured problems that occur on tactical and strategic levels as a part of various DSS'es (Durduran, 2010;Coutinho-Rodrigues et al, 2011;Dahal et al, 2013;Du et al, 2014;Jajac et al, 2014;Jajac et al, 2015;Han et al, 2016;Marovi et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

Possible Applications of Neural Networks in Managing Urban Road Networks

Marović¹

2020

XV International Conference on Durability of Building Materials and Components. eBook of Proceedings

View full text Add to dashboard Cite

Life-cycle management of urban road networks as a part of an urban system is a very complex process from the management standpoint of social, technical and economic aspects. The complexity and multidisciplinarity of such a problem suggest the need for using soft computing tools as well as multicriteria analysis and group decision-making. Recently, there is a significant increase in using various soft computing tools, especially neural networks, for different prediction purposes in the field of road construction planning and management. Along with known advantages of such a prediction method, yet some applications showed the shortcomings. In that sense, the focus of this research is on possible applications of neural networks related to the life-cycle phases during the management of urban road projects. This is done in both horizontal (projects' life-cycle phases) and vertical (hierarchical decisionmaking levels) approach. The final aim of the research is to compare and highlight the possible applications of neural networks as a prediction tool and support for decision-making in urban road management.

show abstract

Development of artificial neural network and multiple linear regression models in the prediction process of the hot mix asphalt properties

Cited by 20 publications

References 20 publications

Bituminous Mixtures Experimental Data Modeling Using a Hyperparameters-Optimized Machine Learning Approach

Bituminous Mixtures Experimental Data Modeling Using a Hyperparameters-Optimized Machine Learning Approach

Cost-Effective Approaches Based on Machine Learning to Predict Dynamic Modulus of Warm Mix Asphalt with High Reclaimed Asphalt Pavement

Possible Applications of Neural Networks in Managing Urban Road Networks

Contact Info

Product

Resources

About