Predicting the rheological properties of bitumen-filler mastic using artificial neural network methods

Nursyahirah Khamis, Muhamad Razuhanafi Mat Yazid, Asmah Hamim, Sri Atmaja P. Rosyidi, Nur Izzi Md Yusoff, Muhamad Nazri Borhan

Research output: Contribution to journalArticle

Abstract

This study was conducted to develop two types of artificial neural network (ANN) model to predict the rheological properties of bitumen-filler mastic in terms of the complex modulus and phase angle. Two types of ANN models were developed namely; (i) a multilayer feed-forward neural network model and (ii) a radial basis function network model. This study was also conducted to evaluate the accuracy of both types of models in predicting the rheological properties of bitumen-filler mastics by means of statistical parameters such as the coefficient of determination (R2), mean absolute error (MAE), mean squared error (MSE) and root mean squared error (RMSE) for every developed model. A set of dynamic shear rheometer (DSR) test data was used on a range of the bitumen-filler mastics with three filler types (limestone, cement and grit stone) and two filler concentrations (35 and 65% by mass). Based on the analysis performed, it was found that both models were able to predict the complex modulus and phase angle of bitumen-filler mastics with the average R2 value exceeding 0.98. A comparison between the two types of models showed that the radial basis function network model has a higher accuracy than multilayer feed-forward neural network model with a higher value of R2 and lower value of MAE, MSE and RMSE. It can be concluded that the ANN model can be used as an alternative method to predict the rheological properties of bitumen-filler mastic.

Original languageEnglish
Pages (from-to)71-78
Number of pages8
JournalJurnal Teknologi
Volume80
Issue number1
DOIs
Publication statusPublished - 1 Jan 2018

Fingerprint

Fillers
Neural networks
Radial basis function networks
Feedforward neural networks
Multilayer neural networks
Rheometers
Limestone
Cements

Keywords

  • Artificial neural network
  • Complex modulus (G*) and phase angle (δ)
  • Multilayer feed-forward neural network
  • Radial basis function network

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Predicting the rheological properties of bitumen-filler mastic using artificial neural network methods. / Khamis, Nursyahirah; Mat Yazid, Muhamad Razuhanafi; Hamim, Asmah; Rosyidi, Sri Atmaja P.; Md Yusoff, Nur Izzi; Borhan, Muhamad Nazri.

In: Jurnal Teknologi, Vol. 80, No. 1, 01.01.2018, p. 71-78.

Research output: Contribution to journalArticle

Khamis, Nursyahirah ; Mat Yazid, Muhamad Razuhanafi ; Hamim, Asmah ; Rosyidi, Sri Atmaja P. ; Md Yusoff, Nur Izzi ; Borhan, Muhamad Nazri. / Predicting the rheological properties of bitumen-filler mastic using artificial neural network methods. In: Jurnal Teknologi. 2018 ; Vol. 80, No. 1. pp. 71-78.
@article{8caac2c6fcec4b5d9eb04d89515049b5,
title = "Predicting the rheological properties of bitumen-filler mastic using artificial neural network methods",
abstract = "This study was conducted to develop two types of artificial neural network (ANN) model to predict the rheological properties of bitumen-filler mastic in terms of the complex modulus and phase angle. Two types of ANN models were developed namely; (i) a multilayer feed-forward neural network model and (ii) a radial basis function network model. This study was also conducted to evaluate the accuracy of both types of models in predicting the rheological properties of bitumen-filler mastics by means of statistical parameters such as the coefficient of determination (R2), mean absolute error (MAE), mean squared error (MSE) and root mean squared error (RMSE) for every developed model. A set of dynamic shear rheometer (DSR) test data was used on a range of the bitumen-filler mastics with three filler types (limestone, cement and grit stone) and two filler concentrations (35 and 65{\%} by mass). Based on the analysis performed, it was found that both models were able to predict the complex modulus and phase angle of bitumen-filler mastics with the average R2 value exceeding 0.98. A comparison between the two types of models showed that the radial basis function network model has a higher accuracy than multilayer feed-forward neural network model with a higher value of R2 and lower value of MAE, MSE and RMSE. It can be concluded that the ANN model can be used as an alternative method to predict the rheological properties of bitumen-filler mastic.",
keywords = "Artificial neural network, Complex modulus (G*) and phase angle (δ), Multilayer feed-forward neural network, Radial basis function network",
author = "Nursyahirah Khamis and {Mat Yazid}, {Muhamad Razuhanafi} and Asmah Hamim and Rosyidi, {Sri Atmaja P.} and {Md Yusoff}, {Nur Izzi} and Borhan, {Muhamad Nazri}",
year = "2018",
month = "1",
day = "1",
doi = "10.11113/jt.v80.11097",
language = "English",
volume = "80",
pages = "71--78",
journal = "Jurnal Teknologi",
issn = "0127-9696",
publisher = "Penerbit Universiti Teknologi Malaysia",
number = "1",

}

TY - JOUR

T1 - Predicting the rheological properties of bitumen-filler mastic using artificial neural network methods

AU - Khamis, Nursyahirah

AU - Mat Yazid, Muhamad Razuhanafi

AU - Hamim, Asmah

AU - Rosyidi, Sri Atmaja P.

AU - Md Yusoff, Nur Izzi

AU - Borhan, Muhamad Nazri

PY - 2018/1/1

Y1 - 2018/1/1

N2 - This study was conducted to develop two types of artificial neural network (ANN) model to predict the rheological properties of bitumen-filler mastic in terms of the complex modulus and phase angle. Two types of ANN models were developed namely; (i) a multilayer feed-forward neural network model and (ii) a radial basis function network model. This study was also conducted to evaluate the accuracy of both types of models in predicting the rheological properties of bitumen-filler mastics by means of statistical parameters such as the coefficient of determination (R2), mean absolute error (MAE), mean squared error (MSE) and root mean squared error (RMSE) for every developed model. A set of dynamic shear rheometer (DSR) test data was used on a range of the bitumen-filler mastics with three filler types (limestone, cement and grit stone) and two filler concentrations (35 and 65% by mass). Based on the analysis performed, it was found that both models were able to predict the complex modulus and phase angle of bitumen-filler mastics with the average R2 value exceeding 0.98. A comparison between the two types of models showed that the radial basis function network model has a higher accuracy than multilayer feed-forward neural network model with a higher value of R2 and lower value of MAE, MSE and RMSE. It can be concluded that the ANN model can be used as an alternative method to predict the rheological properties of bitumen-filler mastic.

AB - This study was conducted to develop two types of artificial neural network (ANN) model to predict the rheological properties of bitumen-filler mastic in terms of the complex modulus and phase angle. Two types of ANN models were developed namely; (i) a multilayer feed-forward neural network model and (ii) a radial basis function network model. This study was also conducted to evaluate the accuracy of both types of models in predicting the rheological properties of bitumen-filler mastics by means of statistical parameters such as the coefficient of determination (R2), mean absolute error (MAE), mean squared error (MSE) and root mean squared error (RMSE) for every developed model. A set of dynamic shear rheometer (DSR) test data was used on a range of the bitumen-filler mastics with three filler types (limestone, cement and grit stone) and two filler concentrations (35 and 65% by mass). Based on the analysis performed, it was found that both models were able to predict the complex modulus and phase angle of bitumen-filler mastics with the average R2 value exceeding 0.98. A comparison between the two types of models showed that the radial basis function network model has a higher accuracy than multilayer feed-forward neural network model with a higher value of R2 and lower value of MAE, MSE and RMSE. It can be concluded that the ANN model can be used as an alternative method to predict the rheological properties of bitumen-filler mastic.

KW - Artificial neural network

KW - Complex modulus (G) and phase angle (δ)

KW - Multilayer feed-forward neural network

KW - Radial basis function network

UR - http://www.scopus.com/inward/record.url?scp=85038090894&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85038090894&partnerID=8YFLogxK

U2 - 10.11113/jt.v80.11097

DO - 10.11113/jt.v80.11097

M3 - Article

VL - 80

SP - 71

EP - 78

JO - Jurnal Teknologi

JF - Jurnal Teknologi

SN - 0127-9696

IS - 1

ER -