Graham, D., Smith, Simon D. and Crapper, Martin (2006) Refining the case-based reasoning model of construction processes. In: Proceedings 22nd Annual ARCOM Conference. ARCOM, pp. 135-142. ISBN 0 9552390 0 1
Full text not available from this repository. (Request a copy)Abstract
Case-based reasoning has been shown in previous research to be capable of modelling construction processes with a good degree of accuracy. However, for the technology to become accepted amongst construction practitioners, the accuracy, reliability and especially the efficiency of the modelling methodology must be improved. To this end, a new method of performing one of the most critical functions of a case-based reasoning model, case retrieval, is proposed. This method, known as 'cluster-based retrieval' involves removing the cases which are likely to be poor solutions to a problem at an early stage, allowing a focus to be placed on finding a solution from a greatly reduced group of cases. A model based upon real construction data and utilising the 'cluster-based retrieval' method have been developed, validated and compared with the model developed in previous research. This comparison aimed to measure any differences in the accuracy, reliability and efficiency as a result of the introduction of 'cluster-based retrieval'. The model utilising the 'cluster-based retrieval' method produced results more efficiently and with more accuracy and reliability than the original model. These results indicate that the use of a 'cluster-based retrieval' method in a case-based reasoning model is a 'step in the right direction' to industry acceptance of the technology as a method of modelling construction processes.
Item Type: | Book Section |
---|---|
Uncontrolled Keywords: | case-based reasoning; construction planning; estimation |
Subjects: | H200 Civil Engineering |
Department: | Faculties > Engineering and Environment > Mechanical and Construction Engineering |
Depositing User: | Becky Skoyles |
Date Deposited: | 08 Jun 2016 08:37 |
Last Modified: | 12 Oct 2019 22:52 |
URI: | http://nrl.northumbria.ac.uk/id/eprint/27051 |
Downloads
Downloads per month over past year