Evaluation of Criteria for the Implementation of High-Performance Computing (HPC) in Danube Region Countries Using Fuzzy PIPRECIA Method

Round 1

Reviewer 1 Report

-The authors could present in more detail how data from chapter 3 were obtained, to convince readers that them are accurate, original and complete.

-I advise a more detailed description of the fuzzy mechanism by which the results from chapter 4 are obtained starting from the data in the chapter 3. The theory stated in chap. 2 is not specifically exemplified by the studied example. So, the research design will be more appropriate and the method will be adequately described. In this way the authors will assure a logical flow and structure.

-I strongly recommend the authors to use citation appropriately and references current, as the following materials as a bibliography:

1. Volosencu, C., Properties of Fuzzy Systems, WSEAS Transactions on Systems, Issue 2, Vol. 8, Feb. 2009, pp. 210-228.
2. Volosencu, C. (Ed.), Fuzzy Logic, IntechOpen Ltd., London, UK, 2020.

So, the introduction will provide more background and will include more relevant references.

-I suggest the authors to present more detailed the results related to the fuzzy mechanism used in study.

Author Response

Reviewer 1:

Thank you very much for the useful suggestions. We accepted all of the suggestions and we are sure that this will improve the quality and contribute to a better understanding of the paper.

Comment 1: The authors could present in more detail how data from chapter 3 were obtained, to convince readers that them are accurate, original and complete.

Response to comment 1: Data in the third section represents input parameters in the model. Figure 1 shows the total number of respondents (decision-makers) per each of 14 Danube region countries. The number of respondents is various and depending on the number of experts in each country (marked with green bar). Some of the decision-makers not properly fulfilled surveys or just gave equal assessments for all criteria. Such surveys are not useful for computation in our methodology, so we have excluded them. The number of correctly completed questionnaires is presented in Figure 1 and marked with red line. For example, in Austria 3 respondents were included in the research, but one of them hasn't fulfilled the questionnaire in proper way, so 2 questionnaires entry in further model. Figure 2 shows these data in percentages. Figure 2 shows that the performance of correctly completed questionnaires is 100% in the following countries: B&H, Bulgaria, Croatia, Czech Republic, Germany and Ukraine. Moldova, Romania, Slovenia and Serbia have a slightly lower percentage, 83%, 82%, 77% and 75% respectively. Countries with 50% or more correctly completed questionnaires are Slovakia (50%), Hungary and Montenegro (55%) and Austria (67%).

We really have explained all data in the third section, but if you think that can be better explained, please give us more concrete instructions and we will follow them.

Comment 2: I advise a more detailed description of the fuzzy mechanism by which the results from chapter 4 are obtained starting from the data in the chapter 3. The theory stated in chap. 2 is not specifically exemplified by the studied example. So, the research design will be more appropriate and the method will be adequately described. In this way the authors will assure a logical flow and structure.

Response to comment 2: Thank you for suggestion. We have added more computations. We shown example of calculation for each step of used methodology (Fuzzy PIPRECIA and Fuzzy Inverse PIPRECIA). Please see the fourth section. Data from the third section is related to results shown in the fourth section in the following way. Respodents from third section are made assessment of criteria (example presented in Table 4). After that we have applied proposed methodology.

Comment 3: I strongly recommend the authors to use citation appropriately and references current, as the following materials as a bibliography:

1. Volosencu, C., Properties of Fuzzy Systems, WSEAS Transactions on Systems, Issue 2, Vol. 8, Feb. 2009, pp. 210-228.
2. Volosencu, C. (Ed.), Fuzzy Logic, IntechOpen Ltd., London, UK, 2020.

So, the introduction will provide more background and will include more relevant references.

Response to comment 3: Thank you for your suggestion. We have included both references in the paper. Additionaly we have included 17 more references.

Comment 4: I suggest the authors to present more detailed the results related to the fuzzy mechanism used in study.

Response to comment 4: This comment is related to the second comment. We have added more detail about fuzzy computation.

Reviewer 2 Report

Dear Authors 

The article seems powerful from the methodology point of view. My main concern is about the literature review. Neither the scope nor the methodology. The article seems so solid in methodology. You need to add more explanation about sustainability. Also, please present some points about that as a discussion in the conclusion section. Please check all the references one more time. In table 3 it is better to use criteria instead of criterion. As a keyword, you have sustainability which is not related to the title at all. You may need to change the title or the keyword. 

Author Response

Reviewer 2:

Thank you very much for the useful suggestions. We accepted all of the suggestions and we are sure that this will improve the quality and contribute to a better understanding of the paper.

The article seems powerful from the methodology point of view. My main concern is about the literature review. Neither the scope nor the methodology. The article seems so solid in methodology. You need to add more explanation about sustainability. Also, please present some points about that as a discussion in the conclusion section. Please check all the references one more time. In table 3 it is better to use criteria instead of criterion. As a keyword, you have sustainability which is not related to the title at all. You may need to change the title or the keyword.  

Response to comment: Thank you for your positive and constructive comments.

We have integrated short state in art with the significance of proposed research into section Introduction. Please see newly added references at the end of the paper. We have tried to include more relevant references. We hope that these improvements will satisfy your expectations.

Conclusion has been extended from sustainability (cost aspect).

We have changed in Table 3 word „criterion“ into criteria.

We have changed the keywords (we have deleted sustainability from keywords).

Round 2

Reviewer 1 Report

After the first review the authors practically "flooded" the work with bibliographic references, the vast majority of them without any connection with the methods used in the paper. The paper is based on computation with fuzzy logic, so those references may remain, and not for example on neural networks, so references to other methods should be removed. It is necessary to eliminate the vast majority of cited references, such as: paragraph from the lines 47-50, references 1-10, 29, 31, 32, 33, 44. It is recommended to carefully select the references that are in the field of work.

Author Response

Reviewer 1: After the first review the authors practically "flooded" the work with bibliographic references, the vast majority of them without any connection with the methods used in the paper. The paper is based on computation with fuzzy logic, so those references may remain, and not for example on neural networks, so references to other methods should be removed. It is necessary to eliminate the vast majority of cited references, such as: paragraph from the lines 47-50, references 1-10, 29, 31, 32, 33, 44. It is recommended to carefully select the references that are in the field of work.

Response: We appreciate your opinion and experience, but please keep in mind that you are not the only reviewer for this paper. As first, please read another reviewer's report and comments by the second reviewer. He has asked for an extension of the literature review based on the use of supercomputers which is the main focus of our paper.

The focus of this study is the possibility of implementation supercomputers in Danube region countries. As can be seen from the paper research was performed in 14 countries and we have used Fuzzy PIPRECIA for determining the significance of eleven criteria in each country. We didn't develop a new methodology or novel approach. In that case, your observation would be justified, but not now. If authors proposing some new mathematical methodology or novel approach, they should be shown literature review as you said (for example for fuzzy logic), but not in this case. How you can ask that we remove references that explain the significance of HPC? With those references we have shown motivation for research, the usefulness of study and justification for performing such research. Some of these references were in the original version of the paper, some we have added in the revised version. It is a necessity for literature review in the field of work (as required the second reviewer), but as we aforementioned field of work is HPC, not fuzzy logic or MCDM methodology. We have integrated short state in art with significance of proposed research into section Introduction.

Fuzzy PIPRECIA method is developed in October 2018 year and so far has used in few studies. Taking into account all previously said, the justification for literature review with fuzzy logic doesn't exist. On the contrary, would be redundant.

We have added two references in methodology, which you are proposed in the first round:

1. Volosencu, C., Properties of Fuzzy Systems, WSEAS Transactions on Systems, Issue 2, Vol. 8, Feb. 2009, pp. 210-228.
2. Volosencu, C. (Ed.), Fuzzy Logic, IntechOpen Ltd., London, UK, 2020.

Now, in this second round we have added text shown below that represents short review of using Fuzzy PIPRECIA method.

The fuzzy PIPRECIA method [39] consists of 11 steps that are shown below and so far has been used in a few studies. Stanković et al [38] have used Fuzzy PIPRECIA in integration with a new developed Fuzzy Measurement Alternatives and Ranking according to the COmpromise Solution (fuzzy MARCOS) for road traffic risk analysis. The significance of the six criteria for evaluation road sections was evaluated using Fuzzy PIPRECIA. Results show that the most important criterion in their study is the number of access points on each section i.e. the second criterion. Original study [39] in which Fuzzy PIPRECIA has been developed dealt with the assessment of conditions for implementation information technology in the warehouse. This method is integrated into SWOT analysis to obtain values of each SWOT dimension. Marković et al. [40] shown that Fuzzy PIPRECIA as a subjective MCDM method can be very successfully applied in integration with other objective methods such is Criteria Importance Through Intercriteria Correlation (CRITIC). They have used an integrated model for the ranking banks in order to achieve business excellence and sustainability. Fuzzy PIPRECIA in the study [41] is used for determining criteria weights for evaluation of green suppliers. This method is successfully applied in integration with Rough Simple Additive Weighting (Rough SAW) method.