#217

Dear Jarosław Watróbsk, Aleksandra Baczkiewicz , Ewa Ziemba, Wojciech Sałabun

I read your paper:

Temporal VIKOR- A New MCDA Method Supporting Sustainability Assessment

My comments

1. In the abstract you say “A framework with a complete set of indicators requires a tool including all objectives simultaneously, such as Multi-Criteria Decision Analysis (MCDA) methods”

It is not clear what you mean by stating “tool including all objectives simultaneously” because if it is true that the alternatives are subject to a complete set of criteria (that are also partial objectives), no MCDM method considers them simultaneously for solving, not even Linear Programming, if you refer to a set of objectives, then I suggest to use MOO or Multi-Objective Optimization, pertaining to Linear and no Linear Programming, being the SIMUS method an example of the first.

2. In the abstract “Therefore, this paper proposes a framework with a newly developed method called the Temporal VIKOR and measurement of data variability that allows aggregation of alternatives’ efficiency over investigated time”

It is good news since there are few MCDM methods, if any, that consider time dependency. However, I am not sure that data variability allows aggregation of alternatives. An alternative A1 at time ‘n’ is a vector, assume for a certain a21 = 0.2. If at time ‘n+1’ data changes to 0.3, you may be having a different alternative A’1, and even a change it the vector’s direction. What you have is a different vector. I do not see how this change can produce a gain in the efficiency; it can certainly increase the gain on the value of the benefit, when the criterion calls for maximization, but in efficiency?

Perhaps it will produce an economic gain, but it also will consume more of the same resource in another activity that can increase the social impact in a larger degree, for instance, consuming more water, that is also necessary for people cooking and cleaning. Everything in life is possibly directly or indirectly related, and the positive variation in one activity may strongly impact another/s, and this is what is missing in all MCDM methods, not only in VIKOR.

You can have a large increase in income by paying visitors in a unique national park, however, this may also produce a deterioration; and this happening in real life when in some caves with ancient petroglyphs, the access has been restricted due to humidity provoked by visitors breathing.

If the DM does not link both effects, what is the value of his result?

3 .Page 2“However, the presented method requires assigning significant values, namely weights, for particular periods of time, therefore, the authors suggest setting the highest significance to the most recent period”

Does it mean that certain periods are more important than others? If so, why? On what grounds are you based to assign the greatest significance to the recent period? Isn’t that arbitrary?

In addition, it seems erroneous, since it means that there is negative improvement in subsequent periods. It appears more logical to take the latest period as a base or value to be improved.

4. “The method proposed by the authors requires conducting pairwise comparisons between criteria for each period examined and aggregating the results received. In the next step, obtained rank relations are transformed into preference relations?????for each pair of alternatives and each period”

The change of preferences makes sense, but in my opinion, what is very debatable is using pair-wise comparisons that are completely artificial and no reliable for several reasons. One of them, is that in the next period there could be a different DM that thinks different from the former. In addition, aggregating results does not solve the problem since you should use intersections, NOT summation. as commented above.

This is the base of Systems Theory. that you are violating by summing

This is very well established in sustainability, where what you look is a common space that contains the coincidences in Economics, Social and Environment. Remember that not always a result is equal to the sum of its parts'. I suggest to look at the very-well known Venn diagram for sustainability

You do not define it, but Sustainable Development Goal 11 (SDG 11) focuses on making cities and human settlements inclusive, safe, resilient, and sustainable. It consists of 10 targets measured by 15 indicators. The targets include:

This is a list of the different factors considered in SDG 11, obtained from AI or from United Nations.

• Safe and affordable housing: Ensuring access to adequate, safe, and affordable housing and upgrading slums.
• Affordable and sustainable transport: Expanding public transport and improving road safety, especially for vulnerable groups.
• Inclusive and sustainable urbanization: Enhancing urban planning and management.
• Protecting cultural and natural heritage: Strengthening efforts to safeguard heritage sites.
• Reducing disaster impact: Minimizing deaths, economic losses, and disruptions caused by disasters.
• Lowering environmental impact: Addressing air pollution and waste management.
• Access to green and public spaces: Ensuring safe, inclusive, and accessible spaces for all.
• Strong national and regional planning: Strengthening urban-rural development planning.
• Policies for inclusion and resilience: Implementing strategies for resource efficiency and disaster risk reduction.
• Cities occupy only 3% of Earth's land but account for 60–80% of energy consumption and 75% of carbon emissions. By 2030, an estimated 5 billion people will live in cities, making sustainable urban development crucial.

You know of course, that a global scale, SDG 11 is linked with SDGs: 1, 2, 3, 7, 8, 9, 10, 12 ,15, 16 and 17, however, you consider GDS 11 in isolation. This is against the same sustainability principle, that considers all of them at the same time, and in so doing, reflecting the influence of a particular SDG on another/s.

5. Page 3 “The results of the VIKOR method are utility function values for each alternative representing proximity to the ideal solution”

Yes, distances from alternatives to the boundaries of a dimensional space, are taken as a measure of alternatives importance, but results of VIKOR are not true utility function values, albeit it is related with utility values. VIKOR balances global utility or best alternative performance, with worst-case scenario.

SD is effectively one of the best indicators; it measures variability, produced by new performance values in the matrix. And what is the gain in doing this?

6-Using entropy you say: “This method determines the criteria weights based on the measurement of information in the decision matrix with performance data [13].”

Using entropy derived values for each criterion is an excellent choice, but it is not a weight but a measure of quantity of information, based on values dispersion. Fundamentally, entropy measures the capacity of the values inside a criterion to evaluate alternative, while a weight measures the relative importance of each criterion, obtained subjectively, and it is the value of the envelope, not of its content, while entropy is based on data. This is the reason why criteria weights are not fit to evaluate alternatives.

7- In Table 2 you correctly used entropy to determine relative importance for evaluation in each one of the 10 criteria that compose SDG 11, but this is only applicable to each of the 10 isolated indicators.

8- In page 8 “There is a clear need for a temporal approach that considers the dynamics and direction of results variability over the years under investigation. This gap is addressed by the Temporal VIKOR approach proposed in this work”

I agree, and your case is a good and valuable example to analyze behaviour along time, and I am impressed by your consideration of so many countries, showing very clearly how each one addressed this thorny issue. However, that is history, very valuable indeed, but past. Nevertheless, it is important that you bring to the reader attention the necessity to consider time, subject that has been normally ignored.

Iin my understanding, the main contribution of your paper, is to stand out the importance of working scenarios dynamically instead of statically, because the first are not real, at least not in our times

For this reason, I believe that it could be more productive to work trying to reasonably forecast the future, than evaluating the past. A very actual issue nowadays, is energy transition to reach zero CO2 from energy generators by 2050, something so obvious, and that was not even mentioned 30 or 40 years ago. That is, how to progressively decrease CO2 from actual levels to zero in 25 years. There are some papers published about this scenario.

9- In page 10, you relate some decrease in some countries to compensate increase in others. From the theoretical compensation point of view this is true, but why the behaviour of a country must depend on the performance of another? It does not make too much sense.

10- In the conclusion you say “-A proper and complete assessment of sustainable development”:

Sorry, I do not think that you can say that, when considering independently each one of the 10 indicators, and forgetting their mutual relationships with the other 16 SDG.

11-Such a strategy provides an opportunity to assess the achievement of long-term goals”

May I remind you that SDG established goals or targets for each SDG, and that they play a paramount role. How can you speak about achieving goal if you did not establish any? By the way, for SDG 11, the goal, by UN, was 81.8 for 2030.

12-The multi-criteria method of temporal sustainability assessment proposed in this paper fulfills these requirements

Not in my opinion

I hope my comments may help you. If you don’t agree with some of them, I will be very glad to have your public rebuttal or criticism, so we can discuss the issue.

Nolberto Munier