TY - JOUR
T1 - The development of a construction cost prediction model with improved prediction capacity using the advanced CBR approach
AU - Koo, Choongwan
AU - Hong, Taehoon
AU - Hyun, Changtaek
N1 - Copyright:
Copyright 2011 Elsevier B.V., All rights reserved.
PY - 2011/7
Y1 - 2011/7
N2 - Decision-making in the early stages of a construction project will have a significant impact on the project. Limited and uncertain information, however, makes it difficult to accurately predict constriction costs. To solve this problem, this study developed the advanced case-based reasoning (CBR) model with 101 cases of multi-family housing projects. The advanced CBR model was developed to integrate the advantages of prediction methodologies such as CBR, multiple regression analysis (MRA), and artificial neural networks (ANN), and the optimization process using a genetic algorithm. This study defined four optimization parameters, as follows: (i) the minimum criterion for scoring the attribute similarity, (ii) the range of attribute weight, (iii) the range of case selection and (iv) the tolerance range of cross range between MRA and ANN. Since the system was developed using the Microsoft-Excel-based Visual Basic Application (VBA) for ease of use, it is expected that the model supports the stakeholders in charge of predicting and managing a construction cost in the early stages of a construction project to get more accurate result from historical cases as a reference.
AB - Decision-making in the early stages of a construction project will have a significant impact on the project. Limited and uncertain information, however, makes it difficult to accurately predict constriction costs. To solve this problem, this study developed the advanced case-based reasoning (CBR) model with 101 cases of multi-family housing projects. The advanced CBR model was developed to integrate the advantages of prediction methodologies such as CBR, multiple regression analysis (MRA), and artificial neural networks (ANN), and the optimization process using a genetic algorithm. This study defined four optimization parameters, as follows: (i) the minimum criterion for scoring the attribute similarity, (ii) the range of attribute weight, (iii) the range of case selection and (iv) the tolerance range of cross range between MRA and ANN. Since the system was developed using the Microsoft-Excel-based Visual Basic Application (VBA) for ease of use, it is expected that the model supports the stakeholders in charge of predicting and managing a construction cost in the early stages of a construction project to get more accurate result from historical cases as a reference.
UR - http://www.scopus.com/inward/record.url?scp=79952443537&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=79952443537&partnerID=8YFLogxK
U2 - 10.1016/j.eswa.2011.01.063
DO - 10.1016/j.eswa.2011.01.063
M3 - Article
AN - SCOPUS:79952443537
VL - 38
SP - 8597
EP - 8606
JO - Expert Systems with Applications
JF - Expert Systems with Applications
SN - 0957-4174
IS - 7
ER -