Neural network model incorporating a genetic algorithm in estimating construction costs

Gwang H. Kim, Jie E. Yoon, Sung Hoon An, Hun Hee Cho, Kyung In Kang

Research output: Contribution to journalArticle

80 Citations (Scopus)

Abstract

This paper applies the back-propagation network (BPN) model incorporating genetic algorithms (GAs) to cost estimation. GAs were adopted in the BPN to determine the BPN's parameters and to improve the accuracy of construction cost estimation. Previously, there have been no appropriate rules to determine these parameters. The construction cost data for 530 residential buildings constructed in Korea between 1997 and 2000 were used for training and evaluating the performance of the model. This study showed that a BPN model incorporating a GA was more effective and accurate in estimating construction costs than the BPN model using trial and error.

Original languageEnglish
Pages (from-to)1333-1340
Number of pages8
JournalBuilding and Environment
Volume39
Issue number11
DOIs
Publication statusPublished - 2004 Nov 1

Fingerprint

back propagation
Backpropagation
genetic algorithm
neural network
Genetic algorithms
Neural networks
costs
cost
Costs
residential building
Korea
performance
parameter

Keywords

  • Construction cost estimating
  • Genetic algorithms
  • Neural networks

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Environmental Engineering
  • Geography, Planning and Development

Cite this

Neural network model incorporating a genetic algorithm in estimating construction costs. / Kim, Gwang H.; Yoon, Jie E.; An, Sung Hoon; Cho, Hun Hee; Kang, Kyung In.

In: Building and Environment, Vol. 39, No. 11, 01.11.2004, p. 1333-1340.

Research output: Contribution to journalArticle

@article{0b81f35541c5400aac3a8025ec2b4688,
title = "Neural network model incorporating a genetic algorithm in estimating construction costs",
abstract = "This paper applies the back-propagation network (BPN) model incorporating genetic algorithms (GAs) to cost estimation. GAs were adopted in the BPN to determine the BPN's parameters and to improve the accuracy of construction cost estimation. Previously, there have been no appropriate rules to determine these parameters. The construction cost data for 530 residential buildings constructed in Korea between 1997 and 2000 were used for training and evaluating the performance of the model. This study showed that a BPN model incorporating a GA was more effective and accurate in estimating construction costs than the BPN model using trial and error.",
keywords = "Construction cost estimating, Genetic algorithms, Neural networks",
author = "Kim, {Gwang H.} and Yoon, {Jie E.} and An, {Sung Hoon} and Cho, {Hun Hee} and Kang, {Kyung In}",
year = "2004",
month = "11",
day = "1",
doi = "10.1016/j.buildenv.2004.03.009",
language = "English",
volume = "39",
pages = "1333--1340",
journal = "Building and Environment",
issn = "0360-1323",
publisher = "Elsevier BV",
number = "11",

}

TY - JOUR

T1 - Neural network model incorporating a genetic algorithm in estimating construction costs

AU - Kim, Gwang H.

AU - Yoon, Jie E.

AU - An, Sung Hoon

AU - Cho, Hun Hee

AU - Kang, Kyung In

PY - 2004/11/1

Y1 - 2004/11/1

N2 - This paper applies the back-propagation network (BPN) model incorporating genetic algorithms (GAs) to cost estimation. GAs were adopted in the BPN to determine the BPN's parameters and to improve the accuracy of construction cost estimation. Previously, there have been no appropriate rules to determine these parameters. The construction cost data for 530 residential buildings constructed in Korea between 1997 and 2000 were used for training and evaluating the performance of the model. This study showed that a BPN model incorporating a GA was more effective and accurate in estimating construction costs than the BPN model using trial and error.

AB - This paper applies the back-propagation network (BPN) model incorporating genetic algorithms (GAs) to cost estimation. GAs were adopted in the BPN to determine the BPN's parameters and to improve the accuracy of construction cost estimation. Previously, there have been no appropriate rules to determine these parameters. The construction cost data for 530 residential buildings constructed in Korea between 1997 and 2000 were used for training and evaluating the performance of the model. This study showed that a BPN model incorporating a GA was more effective and accurate in estimating construction costs than the BPN model using trial and error.

KW - Construction cost estimating

KW - Genetic algorithms

KW - Neural networks

UR - http://www.scopus.com/inward/record.url?scp=3042646757&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=3042646757&partnerID=8YFLogxK

U2 - 10.1016/j.buildenv.2004.03.009

DO - 10.1016/j.buildenv.2004.03.009

M3 - Article

VL - 39

SP - 1333

EP - 1340

JO - Building and Environment

JF - Building and Environment

SN - 0360-1323

IS - 11

ER -