What is it about?
AHP is a multicriteria decision making approach widely adopted in industry. AHP is based on additive utility function, which is basically weighted sum of the utility of each alternative from each criteria. In order to determine the weights as well as the utility of the alternative from the criteria, we assess the pairwise comparisons of the decision makers. This process is based on linguistic preferences such as "I prefer this alternative to that one VERY STRONGLY according to this criteria" etc. Next these linguistic preferences are converted to numbers by using the static 1-9 Saaty Scale. This paper argues that the linguistic terms don't have same meaning across different decision makers, and we need to individualize a scale to convert them to numbers. BAGINS provides a simple and analytically sound individualized scale.
Featured Image
Why is it important?
AHP is a multicriteria decision making approach widely adopted in industry. AHP is based on additive utility function, which is basically weighted sum of the utility of each alternative from each criteria. In order to determine the weights as well as the utility of the alternative from the criteria, we assess the pairwise comparisons of the decision makers. This process is based on linguistic preferences such as "I prefer this alternative to that one VERY STRONGLY according to this criteria" etc. Next these linguistic preferences are converted to numbers by using the static 1-9 Saaty Scale. This paper argues that the linguistic terms don't have same meaning across different decision makers, and we need to individualize a scale to convert them to numbers. BAGINS provides a simple and analytically sound individualized scale.
Perspectives
I believe that as we wıll have more intelligent agents that will communicate with us, they should have some idea what is the numerical score associated with the linguistic terms that we use in our conversations. That is to say, they need to have individualized scales. Thus, BAGINS is an attempt that would assist to such research and practices as well.
Prof. Kemal Kilic
Sabanci Universitesi
Read the Original
This page is a summary of: A basic algorithm for generating an individualized numerical scale, Expert Systems with Applications, December 2023, Elsevier,
DOI: 10.1016/j.eswa.2023.120915.
You can read the full text:
Contributors
The following have contributed to this page
