Dear Marcos dos Santos , Igor Pinheiro de Araújo Costa, Carlos Francisco Simões Gomes

I have read your paper

Multi Criteria Decision-Making in the selection of warships: A new approach to the AHP method

Comments:

1- In page 1 you refer to cognitive effort.

Merriam-Webster Dictionary explain ‘cognitive’ as:

“Relating to, being, or involving conscious intellectual activity (such as thinking, reasoning, or remembering)”

As far as I know AHP is devoid of these characteristics since it is based on intuition

The Dictionary explains that intuition is “a thing that one knows or considers likely from instinctive feeling rather than conscious reasoning”

It appears that one definition contradicts the other.

2- Page 8 “Decision-making may involve simple everyday situations or complex issues that require the use of quantitative and qualitative parameter”

This is correct and applicable to AHP for simple everyday situations, but not for complex ones. Why?

Because complex projects cannot be addressed by the lineal AHP hierarchy, but using all vertical and transversal relationships as those materialized in a network, like ANP. Possibly, this is the reason by which Saaty developed it. AHP was good within the military because it copied its top to bottom hierarchy, but complex projects do not follow that scheme since de 60s or 70s. Any Industrial Organization book will tell you that other kinds of structures appeared and that the lineal hierarchy no longer applied, and this was more than 50 years ago.

3- “The importance of the criteria is defined by the DM in an interactive process with other technical-political actors”

I agree 100% in this statement. Unfortunatelly, many publications assert that criteria are defined only by the DM, therefore what you say is correct.

4- “In multiple criteria decisions, problem alternatives are compared pairwise, and the results express the preferences of the DM with the use of comparative notions”

True in most MCDM methods, but not in all of them. These methods don’t make co parisons between alternatives.

5- Page 9 “The Multicriteria Decision Analysis (MCDA) methods have been widely used because they are scientific and subjective”

Yes, some methods are scientific, but do you consider AHP, ANP, B&W and many other methods scientific, when they are based on opinions, comparing criteria and assigning importance arbitrarily, deciding in a set of criteria which is the best and which the worse, assuming that a trade-off can be used as a weight?

6- Page 13 . You mention 9 criteria. In my opinion, C1, C2 and C3 are related since “Action radios” (C1) depends on the speed. Consequently, you cannot use AHP that clearly specifies that criteria must be independent.

7- Page 14. What normalization method did you use?

8- “The nine criteria were compared two by two, through an interview with the specialists”

Where they talked about the AHP restriction regarding independency? Apparently not, and they don’t have the obligation to be aware of it.

9- In page 18 “It was concluded that the best alternative would be the construction of a new vessel, and this option was endorsed through the sensitivity analysis that followed.”

The paper does not show where the sensitivity analysis (SA) was performed, or what procedure was followed. As a matter of fact, and considering SA, the second model offers the strongest evidence of best alternative stability. Why?

Because it is the model that gives the largest variation of two criteria without altering the ranking. The other criteria don’t allow even the minimum variation. If you are interested, I can show you from where I get this information.

The function of a SA is related with the best alternative but not with its selection. The SA aims at finding how strong the best alternative is regarding the variation of its binding criteria.

By computing the entropy of each criterion, it can be seen that the most important is C6 and in second place C4. Since both are related with the weapons on a warship, it appears to be correct. An old definition of a warship was that it is a platform to carry guns.

The fact that for you the most important criterion is C9 (construction time), is irrelevant since you did not specify the maximum number of years allowed, or fines, or urgency.

In general, and in my opinion, you cannot address a problem of this nature with only 9 criteria. As a matter of fact, I think that you should consider as criteria the 41 characteristics that define each vessel.

I solved your exercise using SIMUS (Linear Programming), and the ranking is A2 = A3 > A1.

Entropy shows that the least important criteria are C8 and C9, followed by C1

Hope my comments may help

Nolberto Munier