Improved similarity measure in case-based reasoning: a case study of construction cost estimation

Purpose Applied a hybrid approach using genetic algorithms (GAs) for a case-based retrieval process in order to increase the overall improved cost accuracy for a case-based library. The paper aims to discuss this issue. Design/methodology/approach A weight optimization approach using case-based reas...

Full description

Saved in:
Bibliographic Details
Published in:Engineering, construction, and architectural management construction, and architectural management, 2020-03, Vol.27 (2), p.561-578
Main Authors: Hyung, Won-Gil, Kim, Sangyong, Jo, Jung-Kyu
Format: Article
Language:English
Subjects:
Accuracy
Case studies
Construction costs
Construction industry
Decision making
Design optimization
Financial planning
Genetic algorithms
Reasoning
Similarity measures
Weight
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Purpose Applied a hybrid approach using genetic algorithms (GAs) for a case-based retrieval process in order to increase the overall improved cost accuracy for a case-based library. The paper aims to discuss this issue. Design/methodology/approach A weight optimization approach using case-based reasoning (CBR) with proposed GAs for developing the CBR model. GAs are used to investigate optimized weight generation with an application to real project cases. Findings The proposed CBR model can reduce errors consistently, and be potentially useful in the early financial planning stage. The authors suggest the developed CBR model can provide decision-makers with accurate cost information for assessing and comparing multiple alternatives in order to obtain the optimal solution while controlling cost. Originality/value The system can operate with more accuracy or less cost, and CBR can be used to better understand the effects of factor interaction and variation during the developed system’s process.
ISSN:0969-9988
1365-232X
DOI:10.1108/ECAM-01-2019-0035