i am developing a website that gives the best recommendation of school for all student using Fuzzy AHP(analytic hierarchy process). and i have a problem on how to integrate student data into my AHP calculation.

To my knowledge AHP gives the ranking of alternatives based on a Pairwise Comparison between each criteria, sub-criteria and alternatives from a scale of 1-9 and it's reciprocal.

let's say i have a criteria of Money(A) and Performance(B). Sub-Criteria of Income is : Income(A1). Sub-Criteria of Performance is : Test Score(B1) and Absent(B2).

and i have Alternative of School X, School Y and School Z

The end result of AHP would be something like

School X : 25% | School Y : 57% | School Z : 18%

that means the best recommended school is Y.

but that solution is purely based on the Pairwise Comparison Matrix. what i want is to have the student input each sub criteria for example :

  1. Andy : Income = $500, Test Score = A, Absent = 0.
  2. Susan : Income = $1200, Test Score = C, Absent = 5.

based on that input the system calculate the best alternative for each of them. maybe it'll output Andy is best suited for School X and Susan is best suited for School Z.

is it possible with Fuzzy AHP or maybe i should use other MCDA Method?

i have tried to give ordinal scale to the user input Income < 500 -> 1
500 < Income <1000 -> 2
1000 < Income -> 3

A <= Test Score <= B -> 3
C <= Test Score <=D -> 2

And so on for all criteria. by then i would have something like :

  1. Andy : [1, 3, 2].
  2. Susan : [3, 2, 2].

i thought of multiplying it with the Sub-criteria Priority Vector, but it doesn't turn out the way i thought it would, and now i'm stuck. most reference would have this kind of input for the alternative.

Your help would be much appreciated.

0

There are 0 best solutions below