A Fuzzy-Interval Dynamic Optimization Model for Regional Water Resources Allocation under Uncertainty
Round 1
Reviewer 1 Report
There are formatting issues (such as references needing to be in square brackets) and the English grammar needs to be improved but overall a very good paper and very well laid out. I like the comparison done between the proposed model and FILP and the succinctness of the work and organization.
I would suggest editing Figure 1 to exclude the map of the world on the top left as everyone knows where China is located. I would leave the three that you have on the top right, and bottom but each one must be accompanied by a scale bar. I would select a slightly lighter shade of blue and make the writing in each region black (more like Figure 3).
That being said can you remove the background colour in the figures 3 onward (like the blue background in Figure 8, the grey background in z-axis in figures 4-7, etc)? I would make the text bolder or darker for contrast and remove this background shading as it detracts from the crispness (no pun intended) of the contents.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The manuscript proposes a fuzzy-interval dynamic programming (FIDP) model for regional water management under uncertainty that combines fuzzy-interval linear programming (FILP) with dynamic programming (DP).
The text is hard to follow due to several reasons. Within the text, several numbers appear which look like references but cannot be followed through due to bad formating.
The theoretical background of the modeling techniques is described extremely succinctly, with few explanations on the justifications and goals for such approaches. The description of some of the symbols included in the equations is missing. For example, I am assuming that X in equation 1 is the decision variable vector, but what are the indexes g, h, and i.
The case study is poorly explained and the reader struggles to understand the objectives, constraints, and uncertainties of the water management problem and the equations that represent them. For instance, what represents the indexes t, i, j, and k? How does F represent respect for the environment? The way the fairness among users is represented should be discussed in more detail, as it looks as the authors assume that fairness is obtained if each user receives the same percentage of their original demand G.
The advantage of FIDP over FILP should be better explained and justified. For example, if the model formulation is linear, what is the advantage of the dynamic programming approach? The inferior results obtained by FILP are due to what specific issue?
The authors state that the FIDP model takes into account the changing factors of the water resources system in a more in-depth and detailed way but do not explain why. What prevented the use of a more detailed formulation in FILP.
Another topic of additional discussion is how the proposed formulation deals with the stated multiple objectives, as LP and DP are originally single objective optimization techniques.
Finally, there is a need for a thorough English language review. Some examples of confusing sentences are presented below:
p. 31 - "unreasonable water allocation will lead to contradictions and conflicts among departments." - what do the authors mean by "unreasonable" or by "contradictions"?
p. 36 "Water resources system is a complex system followed with many uncertain factors" - what do the authors mean by "followed"?
p.47 - "superior models" - what do the authors mean by "superior"?
It is, therefore, my opinion that the authors should revise their manuscript to improve their readability, explaining in greater detail the formulation of their proposed models. In contrast, too much detail is put into discussing the specific results of the case study.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
The authors have answered most of my comments, but I believe the manuscript readability could be greatly improved, without significantly expanding its length.
A short but clear explanation of the decisions variables, objective functions and constraints included in the application of FIDP and FILP models to the case study is missing, as this would help the users understand the comparison of the two models results. In addition, it would help the reader to have the formulation of the FILP model when applied to the case study.
Some sentences are quite confusing (probably because the above mentioned explanation is missing). A few examples:
line 202 - "Because the FIDP model takes into account the dynamic factors in different stage of water resources system, the solution results of this model cannot only conform to the actual situation, but also achieve global optimization under the optimal conditions of each stage" - what are the dynamic factors? what do the authors mean by "actual situation"?
line 213 - "Moreover, the stage changes of water demand from each user would be reflected by dynamic programming" - what are stage changes of water demand?
The symbol list of Annex A should be distributed throughout the text to avoid flipping the pages when reading the model formulation.
Overall, I find the manuscript quite interesting and of good quality. But quite hard to read in detail, even for experts on optimization techniques.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
