When AHP is used for calculation of weights only and SAW is used for ranking the alternatives , how to carry out the sensitivity analysis ?
Dear Vidya
In SAW, you can multiply each value by the corresponding criterion weight.
Sensitivity analysis is performed at present varying the value of the weights of criteria. If the ranking holds, then it is assumed that it is correct.
This process is clearly incorrect, because the following reasons:
1. Only one criterion is selected to vary, and thus, assumming that all remaining criteria are constant, and this is incorret, because normally, in MCDM, all elements are related, therefore, if you vary one, you will also alter others.
2. It is normally selected a criterion with the maximum weight. This is wrong, for this proceeure is not based on any mathematical principle, it is only intuition.
3. Performing sensitivity analysis with subjective weights is wrong, because weights, which in reality are trade-offs, can't be used to evaluate alternatives
Thank you sir for your response,
Sensitivity analysis is performed at present varying the value of the weights of criteria. If the ranking holds, then it is assumed that it is correct.
I have observed a similar method in research papers, assigning maximum weight to anyone criteria and checking the variation in the ranking of alternatives. Is there any other way of doing this?
Dear Vidya
Unfortunately, what you observed is very common, and again, false.
I am sure that nobody can justify this procedure.
Instead, you can use objective weights fron entropy, or statistics, or still bether, use the marginal values for each criterion. Both are exact, and based on values extracted from the data.
And most important, both can be used to evaluate alternatives. They are not trade-offs.
If more information is needed, please don't hesitate and contact me. I will be happy to help
- Badges
- Science topic