Using Decision Trees for Determining Attribute Weights in a Case-Based Model of Early Cost Prediction

Doğan, Sevgi Zeynep; Arditi, David; Günaydın, Hakan

doi:10.1061/(asce)0733-9364(2008)134:2(146)

Cited by 75 publications

(31 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Researchers have made various attempts to develop a prediction model for construction cost. An analysis of these previous studies showed the following three important issues (Al-Harbi, Johnston, & Fayadh, 1994;Arditi & Tokdemir, 1999;Attalla & Hegazy, 2003;Christian & Pandeya, 1997;Dogan, Arditi, & Günaydin, 2008;Hegazy & Ayed, 1998;MOCT, 2008).…”

Section: Introductionmentioning

confidence: 99%

CBR-based cost prediction model-II of the design phase for multi-family housing projects

Hong

Hyun

Moon

2011

Expert Systems with Applications

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

CBR-based cost prediction model-II of the design phase for multi-family housing projects

Hong

Hyun

Moon

2011

Expert Systems with Applications

View full text Add to dashboard Cite

“…The prediction method that uses a neural network is a combination of a statistical method and a syntactic approach. Doğan (2008) proposed another data mining technique that is called the decision tree model. Ji and Kim (2011) proposed a cost prediction algorithm that combines CBR and GA. Park et al (2011) proposed a quantity prediction method for the design development stages that uses the standard database for quantities of reinforcing bars.…”

Section: Literature Reviewmentioning

confidence: 99%

Cost Estimating Methods for RC Structures by Quantity Takeoff and Quantity Prediction in the Design Development Stage

Cho

Jin

2015

Journal of Asian Architecture and Building Engineering

View full text Add to dashboard Cite

The cost estimating methods for reinforced concrete structures based 3D design can be categorized into quantity takeoff (estimation) and quantity prediction in the design development stage. The main estimating elements are concrete and formwork that can be considered in a quantity takeoff, whereas the quantity prediction element is the reinforcing bars in the design development stage.This study proposes a cost estimation system that applies automatic quantity takeoff and quantity prediction with parameters. Firstly, the logic of a quantity takeoff calculates the quantity of formwork with the object information. The Standard Method of Measurement (SMM) automatically estimates the quantity of formwork upon receiving physical shape information from the 3D-object. Secondly, the logic of quantity prediction estimates the quantity of reinforcing bars with data mining tools such as the Decision Tree Model (DTM) and Case Based Reasoning (CBR). To predict the quantity of reinforcing bars, Information Gains (IG) of design attributes are evaluated, and then the valid attributes are selected as the parameters. The Parameters can statistically classify the quantity of reinforcing bars by generating decision tree nodes.

show abstract

“…Emsley et al (2002) studied a method to roughly estimate costs with the use of a nonlinear retrogression method based on historical data. As a method of utilizing a media parameter, Lowe et al (2006) suggested an early stage estimation technique and Dogan et al (2006) proposed a method which can estimate construction cost at an early stage by using reasoning skill to estimate the construction cost of a similar project, based on the construction cost of a previously estimated project, by using the Case-Based Reasoning (CBR) technique. Shaheen et al (2007) and Chou et al (2009) proposed a simulation method that can estimate construction cost at an early stage of planning by determining the range of estimated construction cost with the use of Monte Carlo Simulation.…”

Section: Related Studiesmentioning

confidence: 99%

Framework of the approximate cost estimating model for river dredging construction

Lee

Woo

et al. 2010

KSCE J Civ Eng

View full text Add to dashboard Cite

Enhancing the accuracy of an approximate cost estimate is a critical issue since the estimate ordinarily predicts the total cost of a project with insufficient project information. In order to facilitate the approximate cost estimate process, this paper proposes a computer-based integrated system that enables the flexibility to varying amount and accuracy of available information, higher accuracy by using several different adjustment factors at the same time, and decision supporting feature by fast manipulation of the profits and contingencies in determining the final bid price. Proposed methodology has been implemented for a steel mill construction company in Korea, and its viability was validated. Findings and practical implications are outlined. Keywords: approximate estimate, historical database, adjustment factors, information system 1. INTORDUCTION Preliminary estimating, especially for the design-build contracts, is a critical process predicting the total cost of a project with insufficient project information. In order to enhance the accuracy of a preliminary or an approximate estimate, several methodologies have been developed and utilized. Approximate cost estimating methods ordinarily utilize parameters, factors, and indices. In addition, the advanced applications including probability and statistics, neural networks (NN), and relational database management systems (RDBMS) have been employed for the purpose of improving the accuracy. Findings of previous studies [1][2][3][4] indicate that the accuracy of the estimate using these advanced applications can be increased up to 20%. The approxi mate cost estimating method based on historical cost records is a frequently used one as well. When using this method, the quality of the historical cost data from the previously performed projects is a crucial factor determining the quality of the estimate result. The purpose of this research was to develop an estimating methodology and a computerized system for an engineering and construction firm in Korea, that has abundant experience in planning, designing, procurement, construction, inspection, and test operation of the Pohang and Kwangyang steelworks of the world's second largest steel company. The estimating system developed as a result of this research is integrated with the project management information system (PMIS) of the company in order to obtain quality information in a timely manner. 2. SYSTEM DEVELOPEMENT The system developed in this research, named CEDA (Capital cost Estimate Data Analysis), uses relational database management systems (RDBMS) that manipulate all relevant estimate data and the historical database as well. CEDA consists of two subsystems; one for the estimate of construction cost and the other for the estimate of equipment cost. In estimating the cost of construction, the system mainly utilizes the shared historical construction performance database focused on actual labor productivity. For the cost estimate of major equipment to be installed in the steel mill, this system provides a...

show abstract

Using Decision Trees for Determining Attribute Weights in a Case-Based Model of Early Cost Prediction

Cited by 75 publications

References 11 publications

CBR-based cost prediction model-II of the design phase for multi-family housing projects

CBR-based cost prediction model-II of the design phase for multi-family housing projects

Cost Estimating Methods for RC Structures by Quantity Takeoff and Quantity Prediction in the Design Development Stage

Framework of the approximate cost estimating model for river dredging construction

Contact Info

Product

Resources

About