Water Distribution Network Partitioning Based on Complex Network Theory: The Udine Case Study

Round 1

Reviewer 1 Report

Dear Authors,

After reading, I found your paper interesting. The paper presents the application of an optimization approach, based on the theory of complex networks, combined with an optimization technique based on genetic algorithms. In my opinion, such research is not novel and has little international relevance, but the results can be beneficial for the design and management of the local water distribution system.

The work is well organized and contains all the expected elements. The methodology is effective in achieving the purpose of the work. In the introduction, the authors provide a brief research context, discussing the current state of water system design and optimization. However, the Results section lacks a discussion of the results and a comparison with studies by other authors. This should be added.

The text requires minor editorial revision.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

This paper deals with water distribution Network Partitioning based on the complex network theory. The subject of the manuscript fits the Water Journal. Figures and tables are clear with clear captions. References are adequate in quantity and in quality, but they can be improved to better deploy the gap in the literature and novelty; adjustments may seem necessary for representativeness. The approach, organization, and results of the manuscript need some minor adjustments. The English of the document is proper but adjustments in grammar, style, and fixing typos is necessary.

Major points to revise:

- Novelty needs to be highlighted;

- A full framework or flowchart for the study is recommended and aims at developing the study performed and allowing a full understanding of the method and the proposed global validation, which are not clear yet;

- Further discussion of the results with a comparison with a proper literature review is necessary. Proper geographical distribution of the references will broaden the study and allow it to be extended to other places;

- The validation method for the study performed needs clarification. The discussion part could reference the hypotheses and assumptions made and how this can be generalized and how the validation was achieved, not only the model validation;

- A clearer and more detailed explanation of how the proposed method and the importance of the findings would be relevant to better understand the purpose, outcomes, and implications of the study. How these findings can be used to prevent these situations and how the method could be applied elsewhere;

- A discussion regarding the operation of this method, its usefulness and adaptability for practitioners in the future, and its implications is essential for the validation of the proposed methodology.

Author Response

The authors thank the reviewer for the useful observations.

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

The paper is part of a fairly active research line on algorithms and procedures for the division of a WDN into DMAs. There are many works published in this line, with different objectives, goals to be pursued, methodologies, targets, software used, etc. So the potentially interested audience is quite large, and this is good for the authors and for the journal.

In particular, the core of this research is the case study on a WDN of a small-medium sized city in North-Western Italy called Udine. In this type of work, the most difficult part is to have enough data on the network that one intends to study.

 

The manuscript must undergo a work of enhancement as much as possible to aspire to achieve a good editorial position (as in MDPI Water). First of all, I recommend better explaining the innovations that the paper promotes, emphasizing above all the generic applicability. Which are the original things, the innovative ideas? In addition to the technical report on the Udine WDN, what can you emphasize? Why Is your research important? How is it useful in more general contexts?

- Section 2 ends with a Python flowchart. The foregoing are references to used indices. You could do better. Furthermore, the most innovative contributions of recent years around the material on your page 6 are given by doi 10.1016/j.amc.2020.125726, which demonstrates the need for a more correct approach to use many of the commonly used indices (e.g., Todini's index [57], with about 1000 citations and widely used also in Asia, presents intrinsic problems: this is a very surprising thing, which unfortunately has big consequences everywhere... Similar problems seem to emerge for many other indices.) This approach seems applied in doi 10.3390/w12082210 and others, but the introductory work to understand the new setting is perhaps doi 10.1080/23311916.2019.1643057. This could be very useful to Section 2 and to the readers.

 

- Many similar case studies have been done recently in Asia and comparisons would be useful in Section 3 and 4. For example, doi 10.1007/s13201-022-01791-4, comparing methods and results on 3 different networks (the reviewer is not an author of the suggested papers, obviously). Do you have another network, even an experimental one, on which to compare Sect. 3?

Many data are missing in that section and on the chosen network. For example on tanks, elevations, pressures, pipes, characteristics, etc.

Language and editing are good. Just some minor observations. Line 101: use math mode to write P1,...Pk. Line 102: correct, explain symbols ⊏ P_і ⊏ V. Do you mean the proper subset relation? (It is usually denoted differently.)

Eq. (4): "i≠j,". Explain "D^{-1/2}" in Eq. (2). Some indices are used for too many different things (for example, i).

 

Author Response

The authors thank the reviewer for the useful observations.

Please see the attachment

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

The responses presented by the authors did not improve the paper but were mere justifications. My previous comments were not properly taken in perpective e.g. the framework is for the whole method depicted in Figure 1 is part of the method, not the method itself and it should include in a broder and comprehensive framework.

Author Response

The authors thank again the reviewer for the useful observations.

We added a complete description of the method followed by the flow chart of the code (line 283-295). Moreover, we extended the references adding recent works (line 79-91) and more infos about resilience indices (line 251-259).

Reviewer 3 Report

You answered: "it is reported that Todini’s index does not fit for the Udine WDN so MRI has been used."

 

That's exactly the point. Not only modified and derived resilience indices, but many other energetic indices could have invariance problems. However the authors on other indexes are not very clear. In any case such things are very surprising and I think it will open many discussions in the future.

 

Author Response

The authors thank again the reviewer for the useful observations.

In line 267 we added the fact that invariance affects resilience indices. We did not use energetic indices in this work, so they are not included. Moreover, we have extended the references about the resilience indices (line 251-259) and cited more recent works (line 79-91)