Dear Shanmugasundar , Gaurav Sapkota , Robert Cep and Kanak Kalita
I have read your paper:
Application of MEREC in Multi-Criteria Selection of Optimal Spray-Painting Robot
And my comments are:
1- In page 4, where formula (2) comes from?
2- You provide an explanation of the MEREC method which in reality is very attractive.
As I understand it, you compute the initial performance of each alternative using formula (2). Further, you remove a criterion, say C1, and applies the formula to compute alternative’s performance; further, you find the difference between the two found values for each alternative.
However, every time you apply formula (2) you are missing or not considering the corresponding values of the criterion removed.
What happens for instance, if this criterion is very favourable to alternative A3?
By ignoring it, A3 will be assigned a value lower than real, and this is repeated in each criterion.
By removing one criterion at a time, the whole set of alternatives is never evaluated in its integrity regarding criteria. In fact, if you have, say seven criteria, there are seven different problems.
According to my reasoning, consider that each criterion is an equation that can be represented in the hyperplane. The removal of a criterion may eliminate an existent intersection with another criterion, and thus, changing a previois ranking, and producing Rank Reversal.
Please, correct me if I am mistaken.
3- In page 8 you say “Therefore, it is imperative that, while selecting an appropriate robotic system for spray[1]painting applications, due consideration is given to the specifications and capabilities of the robotic systems”
This is absolutely true. Unfortunatelly, many MCDM problems are addressed without considering the characteristics of the alternatives.
4- In page 14 “The weights of the criteria were calculated by using an objective weight-determination technique called MEREC”
There is no doubt that determining criteria weights using MEREC is free of subjectivity, but I do not think it is objective either, because not all criteria are computed simultaneously, like in using entropy. If in the latter, you eliminate a criterion, the weights will be different.
5- Page 14 “Therefore, it is imperative that, while selecting an appropriate robotic system for spray[1]painting applications, due consideration is given to the specifications and capabilities of the robotic systems”” This indicates that the methods are likely to give the same best solution”
I am afraid that I disagree with your last paragraph, since high correlation between methods does not mean that the solution found is the best.
I hope this helps
Nolberto Munier
- Badges
- Science topic
- Similar topics
- Vectorization