Title: Dealing with inconsistent judgments in multiple criteria sorting models
Authors: Mousseau, Vincent 
Dias, Luís 
Figueira, José 
Issue Date: 2006
Citation: 4OR: A Quarterly Journal of Operations Research. 4:2 (2006) 145-158
Abstract: Abstract. Sorting models consist in assigning alternatives evaluated on several criteria to ordered categories. To implement such models it is necessary to set the values of the preference parameters used in the model. Rather than fixing the values of these parameters directly, a usual approach is to infer these values from assignment examples provided by the decision maker (DM), i.e., alternatives for which (s)he specifies a required category. However, assignment examples provided by DMs can be inconsistent, i.e., may not match the sorting model. In such situations, it is necessary to support the DMs in the resolution of this inconsistency. In this paper, we extend algorithms from mous5ejor03 that calculate different ways to remove assignment examples so that the information can be represented in the sorting model. The extension concerns the possibility to relax (rather than to delete) assignment examples. These algorithms incorporate information about the confidence attached to each assignment example, hence providing inconsistency resolutions that the DMs are most likely to accept.
URI: http://hdl.handle.net/10316/7931
Rights: openAccess
Appears in Collections:FEUC- Artigos em Revistas Internacionais

Files in This Item:
File Description SizeFormat 
obra.pdf102.53 kBAdobe PDFView/Open
Show full item record
Google ScholarTM
Check
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.