Fuzzy AHP-Based Design Performance Index for Evaluation of Ferries

Round 1

Reviewer 1 Report

Dear Authors,

The paper estimate ferries by using AHP in fuzzy environment, it looks interesting, but there are some concerns to improve the paper:

 

 

- The author should better highlight the objective of their paper and to what extent it contributes to closing a gap in the existing literature and/or practice. What is the innovative value of the contribution proposed by the authors? This is an essential part of the Introduction section.

 

 

- Explain why you decided to use AHP and Fuzzy sets, since, the founder of AHP Saaty stated that AHP is already fuzzified. Please, justify why you selected Fuzzy AHP.

 

 Why and did not use some other methods such as BWM, FUCOM. The FUCOM and BWM methods have a similar basis as AHP, it was logical to include the FUCOM method in the research. Why didn't you use some area that considers uncertainties (fuzzy logic, rough sets, etc).

 

-  Add advantages of the proposed methodology and this study.

 

 

Please, remove all undiscussed cited works.

Literature analysis needs to be added. In the entire, there is only two paper from 2020. Add another 15-20 papers of more recent date (period 2020-2022), such as:

 

• Evaluating public transport service quality using picture fuzzy analytic hierarchy process and linear assignment model. Applied Soft Computing, 100, 106920.
• Hospital performance assessment using interval-valued spherical fuzzy analytic hierarchy process. In Decision Making with Spherical Fuzzy Sets (pp. 349-373). Springer, Cham.
• Interval-valued spherical fuzzy analytic hierarchy process method to evaluate public transportation development. Informatica, 32(4), 661-686.
• Estimating Driver Behavior Measures Related to Traffic Safety by Investigating 2-Dimensional Uncertain Linguistic Data—A Pythagorean Fuzzy Analytic Hierarchy Process Approach. Sustainability, 14(3), 1881.

 

 

Be sure that the Literature review part should be significant, not only about listing references. Someone said this, however, others said that. This has no value. You have to confront the methods, reveal their merits or pitfalls. Also, we should amend the Literature review by specific references on the selected case and model.

 

- Discussion part should be separated from conclusion part in the paper. 

 

Finally, some of the afore mentioned content related remarks could be (also) addressed in the conclusion section where this section also suggest issues for further research.

 

 

-The reference list should be extended and you have to discuss more recent works related to MCDM.

- The structure of the paper should be based on below sections:

1- Introduction
1.1. motivation and contribution of the study
2. Literature review
3. Research method
4. Results
5. discussion
6. Conclusion
References

 

Author Response

Dear Reviewer,

Thank you very much for sharing your thoughts.

The paper introduces a framework that could support public transport providers in comparative evaluation of ferries enhancing decisions while doing (a) ferry procurements and (b) ferry refurbishments. Correspondingly, the performance drivers represent the central motive of the paper and the chosen MCDM methodology is one way to achieve this, it is not necessarily the optimal choice but a reasonable one.

We will try to address your concerns in the following paragraphs. Our thoughts are marked in red.

Dear Authors,

The paper estimate ferries by using AHP in fuzzy environment, it looks interesting, but there are some concerns to improve the paper:

 - The author should better highlight the objective of their paper and to what extent it contributes to closing a gap in the existing literature and/or practice. What is the innovative value of the contribution proposed by the authors? This is an essential part of the Introduction section.

We will try to make these three points clearer in the introduction section.

Briefly, these are expressed as,

The objectives of the paper are – (a) to create a structure of ferry operational requirements that considers economic, social, and environmental factors. (b) create a framework where the ferry’s design can be evaluated using these requirements.

The knowledge gap is that prevalent operational requirements structures do not consider requirements holistically, particularly not social (1). The extent of social performance factors is limited to travel time and travel fares (2).

In the waterborne public transport segment, a structure for setting and following up requirements, is novel. The paper suggests a possible structure which together with the proposed modular design approach are tools for the public transport providers to tailor ferries and actively participate in the procurement process. There lies the innovative value.

 - Explain why you decided to use AHP and Fuzzy sets, since, the founder of AHP Saaty stated that AHP is already fuzzified. Please, justify why you selected Fuzzy AHP.

The uncertainties arising from inadequate mapping of decision maker’s preferences due to (a) reluctance, (b) incomplete information, (c) unquantifiable and (d) non obtainable information can lead to a wrong assignment of weights. Even if preferences are scored correctly, there is a risk of the aggregation being misunderstood with reference to the decision maker’s intentions (3). This can lead to the selection of an inefficient ferry.

By using linguistic metrics that represent an interval, the decision makers can feel more confident in their judgements (4). By using fuzzy sets in pair wise comparisons in evaluating weights, one can address the shortcomings highlighted above (5).

Why and did not use some other methods such as BWM, FUCOM. The FUCOM and BWM methods have a similar basis as AHP, it was logical to include the FUCOM method in the research. Why didn't you use some area that considers uncertainties (fuzzy logic, rough sets, etc).

Our primary focus was the development of framework for holistic evaluation of ferries. In treating this as a MCDM problem, we performed a literature survey and identified the methods - ELECTRE, PROMETHEE, TOPSIS, and AHP were widely used. Among these AHP was chosen for its simplicity, popularity and the wide range of industrial application examples.

We understand that BWM and FUCOM eliminate the need for large number of pair wise comparisons (6). But in our case, with 4 sub-criteria per criteria, the number of pair wise comparisons are limited to 6. In comparison, BWM would have resulted in 5 comparisons and FUCOM would have resulted in 3 comparisons. Since the challenge of performing 6 comparisons is not as significant, using AHP is justified in our case.

However, we acknowledge the advantages of BWM and FUCOM and we will incorporate these in the discussion section. We would like to maintain the primary focus of the paper to be the development of an operational requirements framework for evaluation of ferries.

 -  Add advantages of the proposed methodology and this study.

We will add advantages in the discussion section.

Please, remove all undiscussed cited works.                                                                                    

Literature analysis needs to be added. In the entire, there is only two paper from 2020. Add another 15-20 papers of more recent date (period 2020-2022), such as:

 Evaluating public transport service quality using picture fuzzy analytic hierarchy process and linear assignment model. Applied Soft Computing, 100, 106920.

• Hospital performance assessment using interval-valued spherical fuzzy analytic hierarchy process. In Decision Making with Spherical Fuzzy Sets (pp. 349-373). Springer, Cham.
• Interval-valued spherical fuzzy analytic hierarchy process method to evaluate public transportation development. Informatica, 32(4), 661-686.
• Estimating Driver Behavior Measures Related to Traffic Safety by Investigating 2-Dimensional Uncertain Linguistic Data—A Pythagorean Fuzzy Analytic Hierarchy Process Approach. Sustainability, 14(3), 1881.

Be sure that the Literature review part should be significant, not only about listing references. Someone said this, however, others said that. This has no value. You have to confront the methods, reveal their merits or pitfalls. Also, we should amend the Literature review by specific references on the selected case and model.

Thank you for the advice. The papers you shared are relevant and informative. We will try our best to tailor the sections accordingly. But considering our primary focus, we would like to keep MCDM methods as a non-primary part of the paper and instead address these in the discussions section.

 - Discussion part should be separated from conclusion part in the paper. 

 Finally, some of the afore mentioned content related remarks could be (also) addressed in the conclusion section where this section also suggest issues for further research.

 -The reference list should be extended and you have to discuss more recent works related to MCDM.

- The structure of the paper should be based on below sections:

1- Introduction
1.1. motivation and contribution of the study
2. Literature review
3. Research method
4. Results
5. discussion
6. Conclusion
References

We understand your point on paper structure and will separate discussions from conclusions.

 

Thank you very much for taking the time to review our paper.

We hope to submit a revised manuscript soon. Please feel free to discuss with us on our comments.

Kind regards,

Harsha Cheemakurthy and Karl Garme

 

References

1. Tanko M, Cheemakurthy H, Kihl SH, Garme K. Water transit passenger perceptions and planning factors: A Swedish perspective. Travel Behaviour and Society. 2019;16:23-30.
2. de Dios Ortúzar J, Willumsen LG. Modelling transport: John wiley & sons; 2011.
3. Duleba S, Kutlu Gündoğdu F, Moslem S. Interval-valued spherical fuzzy analytic hierarchy process method to evaluate public transportation development. Informatica. 2021;32(4):661-86.
4. Javanbarg MB, Scawthorn C, Kiyono J, Shahbodaghkhan B. Fuzzy AHP-based multicriteria decision making systems using particle swarm optimization. Expert systems with applications. 2012;39(1):960-6.
5. Kulak O, Durmuşoğlu MB, Kahraman C. Fuzzy multi-attribute equipment selection based on information axiom. Journal of materials processing technology. 2005;169(3):337-45.
6. Rezaei J. Best-worst multi-criteria decision-making method. Omega. 2015;53:49-57.

Reviewer 2 Report

Dear Authors,

I congratulate you on this valuable effort.

There are minor comments and it is better to be addressed for general readers.

Section	Line	Statement	Review/suggestion
Abstract	8	Well presented the paper content.
Introduction	39	Leading to poor competitiveness with alternate modes	Any examples with statistics would strengthen your justifications!
	39-40	difficulty in scheduling	Compare to which transport modes?
Methodology	89	Broadly, they need to communicate (a) the route characteristics and the expected number of passengers, (b) the vessel particulars (size, speed) and (c)vessel’s performance (costs, emissions, and commuter preferences)	It is better to add the vessels efficacy compared to other transport facilities (public or private)
Applications	329	Figure 3: (a) Performance comparison of three ferries operating on Line 80 in Stockholm. (b) Break-down of social performance.	lack of explanation to Figure 3 (b)
Discussion and conclusions	427-428 430-431	hull module H1 has an economic index of 0.45. (This could lead to targeted development of new modules, encouraging innovation)	This is a critical conclusion as you need to justify it with reference if there is.

 

Kindest regards,

 

Dr. Evan Nadhim

Comments for author File: Comments.odt

Author Response

Dear Dr. Nadhim

Thank you very much for sharing your thoughts. We will try to address your concerns in the following table. Our thoughts are marked in red.

Section	Line	Statement	Review/suggestion	Response
Abstract	8	Well presented the paper content.		Thank you.
Introduction	39	Leading to poor competitiveness with alternate modes	Any examples with statistics would strengthen your justifications!	We will try to incorporate relevant statistics from Bignon and Pojani (1) who address this issue.
	39-40	difficulty in scheduling	Compare to which transport modes?	Scheduling difficulties of a mixed fleet ferry consisting of fast, slow, large and small ferries such that passenger flow is maintained with minimum operational costs and maintaining customer satisfaction (2).
Methodology	89	Broadly, they need to communicate (a) the route characteristics and the expected number of passengers, (b) the vessel particulars (size, speed) and (c)vessel’s performance (costs, emissions, and commuter preferences)	It is better to add the vessels efficacy compared to other transport facilities (public or private)	We can find efficacy statistics from Stenius, Garme (3).
Applications	329	Figure 3: (a) Performance comparison of three ferries operating on Line 80 in Stockholm. (b) Break-down of social performance.	lack of explanation to Figure 3 (b)	We will explain the chart better in the supporting text. The chart refers to performance of the three vessels against respective social performance criteria.
Discussion and conclusions	427-428 430-431	hull module H1 has an economic index of 0.45. (This could lead to targeted development of new modules, encouraging innovation)	This is a critical conclusion as you need to justify it with reference if there is.	References for the advantage of modularity can be found in the book engineering design: a systematic approach (4). We shall add it to the text.

Very kind regards,

Harsha Cheemakurthy, Karl Garme

 

References

1. Bignon E, Pojani D. River-based public transport: Why won’t Paris jump on board? Case studies on transport policy. 2018;6(2):200-5.
2. Wang Z, Lo HK, Lai M. Mixed-fleet ferry routing and scheduling. Computer-aided Systems in Public Transport: Springer; 2008. p. 181-94.
3. Stenius I, Garme K, Hall Kihl S, Burman M. WATERWAY 365: system analysis of challenges in increased urban mobility by utilization of the water ways. KTH Royal Institute of Technology; 2014.
4. Beitz W, Pahl G, Grote K. Engineering design: a systematic approach: Cambridge University Press; 1996.

Reviewer 3 Report

Dear Authors,

Thank you for sending your paper to the journal Sustainability. Your paper is in the journal's scope and presents the use of the Analytic Hierarchic Process (AHP) to convert 16 the public transport providers' (PTP) subjective preferences and ferry performance into a single dimensionless index.

I like to propose some improvements to the current version of the paper:

1. All abbreviations must be explained first and then used. Recheck the abstarct and rest of the paper.
2. Between chapter and subchapter need to be at least one paragraph. For Recheckexample: chapter 2 and subchapter 2.1
3. Please check formulas 7 and 8. And, of course, related text.
4. Please check formula 11. And, of course, related text.
5. For sure, it will be better to split chapter Discussion and conclusions into two. Chapter Discussion explain your results in more detail with advantages and disadvantages. The conclusion is an overall short presentation of all your research steps with concrete results and very important next steps for research.
6. Recheck that you fill in all necceresy technical data in your paper. For example: affiliation
7. Recheck the reference list according to the journal rules.

Regards,

Author Response

Dear Reviewer,

Thank you for sharing your thoughts. We will try our best to address your suggestions. Please find our replies marked in red.

• All abbreviations must be explained first and then used. Recheck the abstract and rest of the paper.
  • We will make the necessary changes.
• Between chapter and subchapter need to be at least one paragraph. For Recheck example: chapter 2 and subchapter 2.1
  • We added a few lines between chapter and subchapters.
• Please check formulas 7 and 8. And, of course, related text.
  • We slightly changed the formatting of the equation.
  • The ‘1-‘ portion of the equation was added so that higher costs reflect a lower rating.
• Please check formula 11. And, of course, related text.
  • We added another equation to support the older equation.
• For sure, it will be better to split chapter Discussion and conclusions into two. Chapter Discussion explain your results in more detail with advantages and disadvantages. The conclusion is an overall short presentation of all your research steps with concrete results and very important next steps for research.
  • Thank you for the tip. We have separated the two sections and discussed the method, results, advantages and disadvantages.
  • We added future work as part of the conclusions section.
• Recheck that you fill in all necessary technical data in your paper. For example: affiliation
  • We have now updated it with our department and university details.
• Recheck the reference list according to the journal rules.
  • We have added journal abbreviations and italicised them.

Thank you for taking the time to review our paper. We hope to submit a revised manuscript soon. Please feel free to discuss with us on our comments.

Kind Regards

Harsha Cheemakurthy and Karl Garme

Round 2

Reviewer 1 Report

Thank you for improving the paper,

However, it needs more efforts!

Table 6, after converting the crisp numbers to fuzzy numbers and constructing the PCM, show in details the computing process of weights, for both FAHP-EA and FAHP-PSO. The reader has to a clear idea about calculation process.

Line 155, you to mention Table 2 before presenting it.

Figure 2, what is the difference between sub-criteria wise performance scores and sub-criteria wise performance ratings?

Line 189, cite a work for consistency threshold.

Line 282, use a mathematical method for comparison between the obtained results from the different approaches.

The methodology part has to be separate from the case study, it is confusing and not acceptable. Even some results are presented there, why?

Present the adopted methodology in details step by step.

Line 419, why not in this work?

I want you to compare the results with BWM approach.

Add in the appendix the applied survey.

 

Author Response

Dear Reviewer,

Thank you for your comments. We will try our best to answer your queries and adopt your suggestions where relevant. Our thoughts are marked in blue.

Table 6, after converting the crisp numbers to fuzzy numbers and constructing the PCM, show in details the computing process of weights, for both FAHP-EA and FAHP-PSO. The reader has to a clear idea about calculation process.

We have added the process to appendix B so that interested readers may refer to them.  By keeping it in the appendix, we can keep the paper’s focus on the ferry evaluation method.

Line 155, you to mention Table 2 before presenting it.

We have updated this now.

Figure 2, what is the difference between sub-criteria wise performance scores and sub-criteria wise performance ratings?

Performance scores are data that we get directly from the manufacturers or surveys. Since they all have different units and ranges, we normalise them on a scale of 0 to 1 and call them ratings.

We will add an explanation to this in the caption as well as in the text in Step 4: Performance evaluation of alternatives (page 7).

Line 189, cite a work for consistency threshold.

We cited the original article by Saaty [1] here. However, we are aware of the shortcomings on this approach as identified in Zeshui and Cuiping [2]. We overcome this by using the consistency index defined in the fuzzy – PSO approach [3].

Line 282, use a mathematical method for comparison between the obtained results from the different approaches.

We added an approach based on median absolute deviation which is a measure of statistical dispersion to compare the three methods.

The methodology part has to be separate from the case study, it is confusing and not acceptable. Even some results are presented there, why?

Thank you for the tip. We moved the fuzzy weight results into the results section.

We feel that the set of equations relating to the modular ferry DPI calculation are more apt after the introduction of the modular ferry concept in the case study.

Present the adopted methodology in details step by step.

We have described these in the form of steps in section 2.2. Ferry evaluation method. The methodology is also depicted in figure 2 (Page 5). The steps for the fuzzy approaches have been outlined in the appendix.

Line 419, why not in this work?

The purpose of the paper is to introduce metrics for ferry evaluation and outline a method for achieving this. For this we used a fuzzy AHP approach and compared two fuzzy methods for calculating weights.

I want you to compare the results with BWM approach.

In our view, the purpose of the paper is not to compare MCDM methods. We realize that other methods might be more optimal and we have mentioned this in the discussion section so that readers do not get the impression that we claim the used method outperforms other options.

In future work, we can focus on refining the ferry evaluation method. At that time, comparison with other MCDM methods would be more relevant to pursue.  

Add in the appendix the applied survey .

We have added it to the appendix.

 

Thank you for taking the time in carefully reviewing our paper. It has contributed in improving its quality. Please feel free to discuss with us on our replies.

Wish you a pleasant weekend.

Kind Regards,

Harsha Cheemakurthy and Karl Garme

Reviewer 3 Report

Dear Authors,

Thank you for sending the updated version of your paper.

You have a positive reaction to all my questions raised.

Regards,

Author Response

Dear Reviewer,

Thank you for sharing your thoughts in helping us improve the paper.

Wish you a nice day.

Kind Regards,

Harsha Cheemakurthy and Karl Garme

Round 3

Reviewer 1 Report

 

Regarding the survey, I was meaning presenteing the Fuzzy AHP survey, and appendix B does not illustrate that,  moreover, you have to present the translated version in English of the survey.

 

The main question was : presenting in details the computing process of weights, for both FAHP-EA and FAHP-PSO.

 

Your answer was: “We have added the process to appendix B so that interested readers may refer to them. By keeping it in the appendix, we can keep the paper’s focus on the ferry evaluation method.”

 

But, I did not find the calculation process, you just added the un-translated survey.

 

Present in a numerical example the computing process of FAHP-EA and FAHP-PSO.

This point must be covered.

 

 

Author Response

Dear Reviewer,

Thank you for sharing your thoughts.

We think there might have been a slight manuscript version mix-up. There is a difference in the version we uploaded and the one you received. We are not sure what happened, but we apologise for this confusion. We have incorporated your suggestions here and uploaded the latest version. We have named the file ‘manuscript_v4’. Please find our replies marked in blue.

Regarding the survey, I was meaning presenteing the Fuzzy AHP survey, and appendix B does not illustrate that, moreover, you have to present the translated version in English of the survey.

The fuzzy AHP survey is performed with the help of a graphical user interface (GUI) that we developed. The interface has now been added in the appendix. The existing survey in Appendix B is translated to English.

The main question was : presenting in details the computing process of weights, for both FAHP-EA and FAHP-PSO.

Your answer was: “We have added the process to appendix B so that interested readers may refer to them. By keeping it in the appendix, we can keep the paper’s focus on the ferry evaluation method.”

 But, I did not find the calculation process, you just added the un-translated survey.

The calculation process of the two fuzzy methods is outlined in Appendix C instead of B. Using the procedure here, the reader may calculate the fuzzy weights. 

Present in a numerical example the computing process of FAHP-EA and FAHP-PSO. This point must be covered.

The methodology outlined in Appendix C is implemented in MATLAB to calculate the weights.  In section 3.1, we report fuzzy input on input intensities of importance (Eq. 30) and report their fuzzification and final calculated weights. 

 

We hope that we were able to answer all your queries.

Wish you a nice week ahead.

Kind Regards,

Harsha Cheemakurthy and Karl Garme