Hybrid models of neural networks and genetic algorithms for predicting preliminary cost estimates

G. H. Kim, D. S. Seo, Kyung In Kang

Research output: Contribution to journalArticle

65 Citations (Scopus)

Abstract

This technical note applies hybrid models of neural networks (NN) and genetic algorithms (GA) to cost estimation of residential buildings to predict preliminary cost estimates. Data used in the study are for residential buildings constructed from 1997 to 2000 in Seoul, Korea. These are used in training each model and evaluating its performance. The models applied were Model I, which determines each parameter of a back-propagation network by a trial-and-error process; Model II, which determines each parameter of a back-propagation network by GAs; and Model III, which trains weights of NNs using genetic algorithms. The research revealed that optimizing each parameter of back-propagation networks using GAs is most effective in estimating the preliminary costs of residential buildings. Therefore, GAs may help estimators overcome the problem of the lack of adequate rules for determining the parameters of NNs.

Original languageEnglish
Pages (from-to)208-211
Number of pages4
JournalJournal of Computing in Civil Engineering
Volume19
Issue number2
DOIs
Publication statusPublished - 2005 Apr 1

Fingerprint

Genetic algorithms
Neural networks
Backpropagation
Costs

Keywords

  • Algorithms
  • Construction industry
  • Cost estimates
  • Hybrid methods
  • Neural networks

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Computer Science Applications
  • Computational Theory and Mathematics

Cite this

Hybrid models of neural networks and genetic algorithms for predicting preliminary cost estimates. / Kim, G. H.; Seo, D. S.; Kang, Kyung In.

In: Journal of Computing in Civil Engineering, Vol. 19, No. 2, 01.04.2005, p. 208-211.

Research output: Contribution to journalArticle

@article{796d21b193e24306852db84f3db15dca,
title = "Hybrid models of neural networks and genetic algorithms for predicting preliminary cost estimates",
abstract = "This technical note applies hybrid models of neural networks (NN) and genetic algorithms (GA) to cost estimation of residential buildings to predict preliminary cost estimates. Data used in the study are for residential buildings constructed from 1997 to 2000 in Seoul, Korea. These are used in training each model and evaluating its performance. The models applied were Model I, which determines each parameter of a back-propagation network by a trial-and-error process; Model II, which determines each parameter of a back-propagation network by GAs; and Model III, which trains weights of NNs using genetic algorithms. The research revealed that optimizing each parameter of back-propagation networks using GAs is most effective in estimating the preliminary costs of residential buildings. Therefore, GAs may help estimators overcome the problem of the lack of adequate rules for determining the parameters of NNs.",
keywords = "Algorithms, Construction industry, Cost estimates, Hybrid methods, Neural networks",
author = "Kim, {G. H.} and Seo, {D. S.} and Kang, {Kyung In}",
year = "2005",
month = "4",
day = "1",
doi = "10.1061/(ASCE)0887-3801(2005)19:2(208)",
language = "English",
volume = "19",
pages = "208--211",
journal = "Journal of Computing in Civil Engineering",
issn = "0887-3801",
publisher = "American Society of Civil Engineers (ASCE)",
number = "2",

}

TY - JOUR

T1 - Hybrid models of neural networks and genetic algorithms for predicting preliminary cost estimates

AU - Kim, G. H.

AU - Seo, D. S.

AU - Kang, Kyung In

PY - 2005/4/1

Y1 - 2005/4/1

N2 - This technical note applies hybrid models of neural networks (NN) and genetic algorithms (GA) to cost estimation of residential buildings to predict preliminary cost estimates. Data used in the study are for residential buildings constructed from 1997 to 2000 in Seoul, Korea. These are used in training each model and evaluating its performance. The models applied were Model I, which determines each parameter of a back-propagation network by a trial-and-error process; Model II, which determines each parameter of a back-propagation network by GAs; and Model III, which trains weights of NNs using genetic algorithms. The research revealed that optimizing each parameter of back-propagation networks using GAs is most effective in estimating the preliminary costs of residential buildings. Therefore, GAs may help estimators overcome the problem of the lack of adequate rules for determining the parameters of NNs.

AB - This technical note applies hybrid models of neural networks (NN) and genetic algorithms (GA) to cost estimation of residential buildings to predict preliminary cost estimates. Data used in the study are for residential buildings constructed from 1997 to 2000 in Seoul, Korea. These are used in training each model and evaluating its performance. The models applied were Model I, which determines each parameter of a back-propagation network by a trial-and-error process; Model II, which determines each parameter of a back-propagation network by GAs; and Model III, which trains weights of NNs using genetic algorithms. The research revealed that optimizing each parameter of back-propagation networks using GAs is most effective in estimating the preliminary costs of residential buildings. Therefore, GAs may help estimators overcome the problem of the lack of adequate rules for determining the parameters of NNs.

KW - Algorithms

KW - Construction industry

KW - Cost estimates

KW - Hybrid methods

KW - Neural networks

UR - http://www.scopus.com/inward/record.url?scp=18144382336&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=18144382336&partnerID=8YFLogxK

U2 - 10.1061/(ASCE)0887-3801(2005)19:2(208)

DO - 10.1061/(ASCE)0887-3801(2005)19:2(208)

M3 - Article

AN - SCOPUS:18144382336

VL - 19

SP - 208

EP - 211

JO - Journal of Computing in Civil Engineering

JF - Journal of Computing in Civil Engineering

SN - 0887-3801

IS - 2

ER -