Fuzzy clustering validity for contractor performance evaluation: Application to UAE contractors

Several statistical algorithms are used to categorize contractors. The number of categories depends on the clustering algorithm used. This paper presents a framework for classifying contractors using five of the most common clustering algorithms and assesses their performance with appropriate validi...

Full description

Saved in:
Bibliographic Details
Published in:Automation in construction Vol. 31; pp. 158 - 168
Main Authors: Nassar, Khaled, Hosny, Ossama
Format: Journal Article
Language:English
Published: Kidlington Elsevier B.V 01.05.2013
Elsevier
Subjects:
Algorithms
Applied sciences
Building construction
Buildings. Public works
Categories
Categorization
Clustering
Contractor performance
Contractors
Exact sciences and technology
Fuzzy clustering
Fuzzy logic
Mathematical analysis
Miscellaneous
United Arab Emirates
Asia
United Emirates
Contractor performance
Building construction
Fuzzy clustering
Categorization
Cluster analysis
Performance evaluation
Building industry
Building contractor
Construction industry
Fuzzy logic
Qualitative analysis
Application
Deterministic approach
Quantitative analysis
ISSN:0926-5805
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Several statistical algorithms are used to categorize contractors. The number of categories depends on the clustering algorithm used. This paper presents a framework for classifying contractors using five of the most common clustering algorithms and assesses their performance with appropriate validity measures. The framework was implemented on actual data for 14 contractors working in UAE using a database of 294 projects. Quantitative measures were suggested and calculated for the contractors in the database. Qualitative measures were determined using AHP. The quality of contractor's staff and equipment was deemed to be the most important measure. The results show that contractors are grouped into four categories based on the quantitative and qualitative measures identified. The Fuzzy-C means algorithm had the highest validity measures when applied to the studied data set. The results show that the proposed framework can be used to categorize contractors into different performance groups in a rational and unbiased way. ► Framework for categorizing contractors using five clustering algorithms is introduced. ► Qualitative measures for assessing contractors' performance are identified. ► Different methods for categorizing contractors based on performance are introduced.
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:0926-5805
DOI:10.1016/j.autcon.2012.11.013