Dear Adis Puška, Ilija Stojanovic, Anđelka Štilić

I read your paper

The Influence of Objective Weight Determination Methods on Electric Vehicle Selection in Urban Logistics

My comments are:

1- Page 4, Figure 1. You put weights, but where are criteria defined? Even if they are, why are they before the alternatives? To define criteria, you need to know to which alternatives they will evaluate. There are not universal criteria applicable to anything.

2) Page 5, formula 3 is incorrect. You must multiply summation of partial entropies (S) by -1/ln(n). This is not shown in the formula. You forgot the minus symbol.

Observe that quantity of information D= 1-S, consequently it never can have a value greater than 1, as you show in Table 4. I suggest to revise your work and conclusions. Compare with weights from the MEREC method in Table 6, all of them lower than 1, as it should be.

3) you say in page 11 “Accordingly, the ranking order is determined for all weights obtained from the various methods”

I am afraid that this is incorrect. Ranking is formed considering all possible interactions between alternatives and criteria, no by criteria

“The Entropy method and TOPSIS method are used by Dwivedi and Sharma [21] in their assessment of different e-cars against various criteria to identify optimal performance”

Optimization does not exist in multi criteria.

4) page 12 “Assigning weights to these criteria enables more effective decision-making, with subjective methods relying on the expertise of decision-makers while objective methods use 126 alternative values corresponding to individual criteria. The focus of the current research was to investigate how these objective weighting methods influence the final decision”

Where the 126 alternative values come from?

Assigning weights to criteria only determines their relative importance. These weights have nothing to do with evaluation of criteria, as is the case with entropy, and SD, because these depend on the criteria values for each alternative. They quantify criteria regrading to their contribution in evaluating alternatives.

Just think, if criteria, according to you and many MCDM methods, are independent of the alternatives they must evaluate, how can you think that they have influence on the alternatives?

“In the study, a total of nine small vans were evaluated against 12 criteria representing their technical characteristics”

Now, you say the opposite, criteria are related with alternatives, as it should be.

“This was because the MEREC method gave more weight to the criteria where this vehicle performed best”

I don’t think this is correct. MEREC weights are obtained by deleting one criterion at a time, running the software, and see the influence of this deletion. Observe that in so doing you are deleting the whole series of performance values.

5-Page 5 “The greater the dispersion of the data, the higher the Entropy Value and, by extension, the criterion weight”

Sorry, you are mistaken. The higher entropy means the opposite, it measures uncertainty and it has the lower dispersion, and by extension, the lower criterion weight. WHY?

Because it means that for a criterion all its performance values are very similar, and consequently, all of them have a very similar probability, like in rolling a die. All six faces have the same probability (1/6), consequently entropy is 1 and quantity of information is 0.

6- “The nature of certain criteria dictated the normalization technique employed, dependent on whether a criterion was of benefit or cost type”

Not necessarily

7- What you have in Table 4 is the entropy (Ei) of a criterion (summation of individual entropies ej), which of course is negative, but you must work with its average value, by multiplying Ei by constant

K = - 1/ln(n). In so doing Ei becomes positive and less than 1.

These are my comments

I hope they help

Nolberto Munier