%0 Conference Proceedings
%T An Empirical Examination of the Consistency Ratio in the Analytic Hierarchy Process (AHP)
%+ National Research University Higher School of Economics [St. Petersburg]
%+ Science of St. Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences (SPIIRAS )
%A Lukinskiy, Valery
%A Lukinskiy, Vladislav
%A Sokolov, Boris
%A Bazhina, Darya
%Z Part 10: Regular Session: Advanced Modelling Approaches
%< avec comité de lecture
%@ 978-3-030-85913-8
%( IFIP Advances in Information and Communication Technology
%B IFIP International Conference on Advances in Production Management Systems (APMS)
%C Nantes, France
%Y Alexandre Dolgui
%Y Alain Bernard
%Y David Lemoine
%Y Gregor von Cieminski
%Y David Romero
%I Springer International Publishing
%3 Advances in Production Management Systems. Artificial Intelligence for Sustainable and Resilient Production Systems
%V AICT-634
%N Part V
%P 477-485
%8 2021-09-05
%D 2021
%R 10.1007/978-3-030-85914-5_51
%K Expert judgements
%K Consistency ratio
%K Tangible objects
%Z Computer Science [cs]Conference papers
%X Papers related to solving practical problems in all areas of economic activity use the AHP methodology from the 1980s, proposed on the basis of studying a limited number of tangible objects. One possible option to overcome this situation could be a new approach based on probability theory and mathematical statistics. We used information obtained from processing 292 matrices collected between 2014 and 2020 and reflecting expert judgements. We were able to compare a particular expert’s judgement with an array of expert judgements from previous studies to refine the computational operations of the AHP. It was found that the probability estimates of the consistency indices for expert judgements differ significantly from the probability estimates for the generated values; e.g., for 6 $$\times$$× 6 matrices, they differ by a factor of 20. The results of our study do not confirm the need to revise expert judgements if the consistency ratio (C.R.) is greater than 0.10, and confirm the need for more example calculations for tangible objects and for new dependencies to estimate intangible objects based on artificial intelligence models and methods.
%G English
%Z TC 5
%Z WG 5,7
%2 https://inria.hal.science/hal-03897864/document
%2 https://inria.hal.science/hal-03897864/file/520762_1_En_51_Chapter.pdf
%L hal-03897864
%U https://inria.hal.science/hal-03897864
%~ IFIP
%~ IFIP-AICT
%~ IFIP-TC
%~ IFIP-TC5
%~ IFIP-WG
%~ IFIP-APMS
%~ IFIP-WG5-7
%~ IFIP-AICT-634