Aull-Hyde, R., Erdogan, S., & Duke, J. M. (2006). An experiment on the consistency of aggregated comparison matrices in AHP. European Journal of Operational Research, 171(1), 290–295.
Added by: Klaus-admin (05 Jun 2019 23:08:11 Asia/Singapore) Last edited by: Klaus D. Goepel (06 Jun 2019 01:27:31 Asia/Singapore)
|Resource type: Journal Article
BibTeX citation key: AullHyde2006
Email resource to friend
View all bibliographic details
Keywords: Analytic Hierarchy Process (AHP), consistency, group decision, Simulation
Creators: Aull-Hyde, Duke, Erdogan
Collection: European Journal of Operational Research
Views index: %
Popularity index: 34.5%
The analytic hierarchy process can be used for group decision making by aggregating individual judgments or individual priorities. The most commonly used aggregation methods are the geometric mean method and the weighted arithmetic mean method. While it is known that the weighted geometric mean comparison matrix is of acceptable consistency if all individual comparison matrices are of acceptable consistency, this paper addresses the following question: Under what conditions would an aggregated geometric mean comparison matrix be of acceptable consistency if some (or all) of the individual comparison matrices are not of acceptable consistency? Using Monte Carlo simulation, results indicate that given a sufficiently large group size, consistency of the aggregate comparison matrix is guaranteed, regardless of the consistency measures of the individual comparison matrices, if the geometric mean is used to aggregate. This result implies that consistency at the aggregate level is a non-issue in group decision making when group size exceeds a threshold value and the geometric mean is used to aggregate individual judgments. This paper determines threshold values for various dimensions of the aggregated comparison matrix.