In this paper, we proposed the novel integration method for AHP based on multidimensional scaling analysis which enable us to know the multidimensional stretch of the data. Conventional method considers consistency of them in one-dimention. Multidimensionally suitable weights for the data integration are calculated by the similarity matrix derived from the pairwise comparison matrix. We presents several experimental results to compare the characteristic features of the method and research confirming the efficacy. © 2013 The Authors.
Notsu, A., Kawakami, H., Tezuka, Y., & Honda, K. (2013). Intergration of information based on the similarity in AHP. In Procedia Computer Science (Vol. 22, pp. 1011–1020). Elsevier B.V. https://doi.org/10.1016/j.procs.2013.09.186