Artificial neural network modeling studies to predict the amount of carried weight by iran khodro transportation system

Mohd Nizam Ab Rahman, Saeid Jafarzadeh-Ghoushchi, Dzuraidah Abd. Wahab, Majid Jafarzadeh-Ghoushji

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

This paper investigates the use of three artificial neural network (ANNs) algorithms, namely, incremental back propagation algorithm (IBP), genetic algorithm (GA) and Levenberg-Marquardt algorithm (LM) for predicting Carried weight, with an automobile industry namely, Iran Khodro Company (IKCO) used as the study case. These algorithms belong to three classes: gradient descent backpropagation algorithm, genetic algorithm and Levenberg- Marquardt algorithm. The above algorithms were compared according to their prediction ability, prediction accuracy, as well as degree of generalization. The network structure was trained with the algorithms by using some numerical measures as the training set. Those algorithms were then compared according to their performances in training and prediction accuracy in testing based on root mean square error (RMSE) and correlation coefficient (R2). The results indicate that incremental back propagation performs better than the other algorithms in training and has higher prediction accuracy during the learning period.

Original languageEnglish
Article number25
Pages (from-to)146-154
Number of pages9
JournalLife Science Journal
Volume11
Issue numberSPEC.ISS.2
Publication statusPublished - 2014

Fingerprint

Iran
Neural networks
Weights and Measures
Backpropagation algorithms
Genetic algorithms
Backpropagation
Automotive industry
Mean square error
Automobiles
Aptitude
Industry
Testing
Learning

Keywords

  • Artificial neural network
  • Genetic algorithm
  • Incremental back propagation algorithm
  • Levenberg-marquardt algorithm
  • Prediction

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology(all)

Cite this

Artificial neural network modeling studies to predict the amount of carried weight by iran khodro transportation system. / Ab Rahman, Mohd Nizam; Jafarzadeh-Ghoushchi, Saeid; Abd. Wahab, Dzuraidah; Jafarzadeh-Ghoushji, Majid.

In: Life Science Journal, Vol. 11, No. SPEC.ISS.2, 25, 2014, p. 146-154.

Research output: Contribution to journalArticle

@article{64a8e418cd8642c2af40fdea6e8a6bd5,
title = "Artificial neural network modeling studies to predict the amount of carried weight by iran khodro transportation system",
abstract = "This paper investigates the use of three artificial neural network (ANNs) algorithms, namely, incremental back propagation algorithm (IBP), genetic algorithm (GA) and Levenberg-Marquardt algorithm (LM) for predicting Carried weight, with an automobile industry namely, Iran Khodro Company (IKCO) used as the study case. These algorithms belong to three classes: gradient descent backpropagation algorithm, genetic algorithm and Levenberg- Marquardt algorithm. The above algorithms were compared according to their prediction ability, prediction accuracy, as well as degree of generalization. The network structure was trained with the algorithms by using some numerical measures as the training set. Those algorithms were then compared according to their performances in training and prediction accuracy in testing based on root mean square error (RMSE) and correlation coefficient (R2). The results indicate that incremental back propagation performs better than the other algorithms in training and has higher prediction accuracy during the learning period.",
keywords = "Artificial neural network, Genetic algorithm, Incremental back propagation algorithm, Levenberg-marquardt algorithm, Prediction",
author = "{Ab Rahman}, {Mohd Nizam} and Saeid Jafarzadeh-Ghoushchi and {Abd. Wahab}, Dzuraidah and Majid Jafarzadeh-Ghoushji",
year = "2014",
language = "English",
volume = "11",
pages = "146--154",
journal = "Life Science Journal",
issn = "1097-8135",
publisher = "Zhengzhou University",
number = "SPEC.ISS.2",

}

TY - JOUR

T1 - Artificial neural network modeling studies to predict the amount of carried weight by iran khodro transportation system

AU - Ab Rahman, Mohd Nizam

AU - Jafarzadeh-Ghoushchi, Saeid

AU - Abd. Wahab, Dzuraidah

AU - Jafarzadeh-Ghoushji, Majid

PY - 2014

Y1 - 2014

N2 - This paper investigates the use of three artificial neural network (ANNs) algorithms, namely, incremental back propagation algorithm (IBP), genetic algorithm (GA) and Levenberg-Marquardt algorithm (LM) for predicting Carried weight, with an automobile industry namely, Iran Khodro Company (IKCO) used as the study case. These algorithms belong to three classes: gradient descent backpropagation algorithm, genetic algorithm and Levenberg- Marquardt algorithm. The above algorithms were compared according to their prediction ability, prediction accuracy, as well as degree of generalization. The network structure was trained with the algorithms by using some numerical measures as the training set. Those algorithms were then compared according to their performances in training and prediction accuracy in testing based on root mean square error (RMSE) and correlation coefficient (R2). The results indicate that incremental back propagation performs better than the other algorithms in training and has higher prediction accuracy during the learning period.

AB - This paper investigates the use of three artificial neural network (ANNs) algorithms, namely, incremental back propagation algorithm (IBP), genetic algorithm (GA) and Levenberg-Marquardt algorithm (LM) for predicting Carried weight, with an automobile industry namely, Iran Khodro Company (IKCO) used as the study case. These algorithms belong to three classes: gradient descent backpropagation algorithm, genetic algorithm and Levenberg- Marquardt algorithm. The above algorithms were compared according to their prediction ability, prediction accuracy, as well as degree of generalization. The network structure was trained with the algorithms by using some numerical measures as the training set. Those algorithms were then compared according to their performances in training and prediction accuracy in testing based on root mean square error (RMSE) and correlation coefficient (R2). The results indicate that incremental back propagation performs better than the other algorithms in training and has higher prediction accuracy during the learning period.

KW - Artificial neural network

KW - Genetic algorithm

KW - Incremental back propagation algorithm

KW - Levenberg-marquardt algorithm

KW - Prediction

UR - http://www.scopus.com/inward/record.url?scp=84894425970&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84894425970&partnerID=8YFLogxK

M3 - Article

VL - 11

SP - 146

EP - 154

JO - Life Science Journal

JF - Life Science Journal

SN - 1097-8135

IS - SPEC.ISS.2

M1 - 25

ER -