1985
DOI: 10.1016/0305-0483(85)90052-0
|View full text |Cite
|
Sign up to set email alerts
|

The legitimacy of rank reversal—A comment

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
51
0
4

Year Published

1990
1990
2018
2018

Publication Types

Select...
8
1

Relationship

0
9

Authors

Journals

citations
Cited by 162 publications
(58 citation statements)
references
References 2 publications
0
51
0
4
Order By: Relevance
“…Belton and Gear [6] and Dyer [9] pointed out that this method can suer from rank reversal (an alternative chosen as the best over a set of X, is not chosen when some alternative, perhaps an unimportant one, is excluded from X). Belton and Gear [7] and Dyer and Wendel [10] attacked the AHP on the grounds that it lacks a ®rm theoretical basis. Harker and Vargas [13] and Perez [19] discussed these major criticisms and proved with a theoretical work and examples that they are not valid.…”
Section: The Analytical Hierarchy Process (Ahp)mentioning
confidence: 99%
“…Belton and Gear [6] and Dyer [9] pointed out that this method can suer from rank reversal (an alternative chosen as the best over a set of X, is not chosen when some alternative, perhaps an unimportant one, is excluded from X). Belton and Gear [7] and Dyer and Wendel [10] attacked the AHP on the grounds that it lacks a ®rm theoretical basis. Harker and Vargas [13] and Perez [19] discussed these major criticisms and proved with a theoretical work and examples that they are not valid.…”
Section: The Analytical Hierarchy Process (Ahp)mentioning
confidence: 99%
“…Some decision-making algorithms have been criticized for the possible rank reversal phenomenon caused by the addition or deletion of an alternative [21][22][23][24].…”
Section: Rank Reversal In the Comparison Score Algorithmmentioning
confidence: 99%
“…Essa normalização é feita para cada critério, obtendo-se assim uma matriz, onde todos os valores estão entre zero e um, denominada matriz de desejabilidades parciais W = [W nm ], sendo que n indica o número de alternativas e m o número de critérios, conforme se mostra na Tabela 1. Uma vez que a escala onde se lê a medida do desempenho estimado de cada alternativa com respeito a cada critério é determinada a partir do próprio conjunto de alternativas, por meio da normalização supracitada, pode-se minimizar a eventual ocorrência de reversão de ordem pelos dois caminhos seguintes: (i) adição de uma alternativa nova que contribua para expandir o intervalo de variação dos valores normalizados; ou, alternativamente, (ii) ponderação de cada alternativa com relação a um critério com o valor de uma unidade da escala em que o critério é medido (Belton & Stewart, 2002, p. 159; Belton & Gear, 1983, 1985Tallarico, 1990 …”
Section: O Método Todimunclassified