Improved Hydraulic Simulation of Valve Layout Effects on Post-Earthquake Restoration of a Water Distribution Network

Round 1

Reviewer 1 Report

Overall I enjoyed reading the paper.  I would recommend the authors consider more practical aspects/ limitations. 

 

99: open source?

99: Does EPANET 3.0 available? To the best of my knowledge, only EPANET 2.2 is available (still the beta version though). Also, the link in the references does not work.

114: How about TIME TO dispatch/ repair?

121: clarify on repair vs. replace?

128: step 1 through 8: when does these recommended steps can be applied?
For example, if the magnitude of 8.1 hits, then most of the infrastructure will be broken. Do the steps (outlined) still apply? People have no electricity/ water so bottled water will be the most efficient one. Any thoughts/ limitation of model simulations in terms of applicability/ practical aspects?

152: step 2 - 3 needs more explanations; step 2 mentions the PGA vs. Probability, but Step 3 all of a sudden mentions failure/ normal based on a random number. How the actual seismic simulation result is considered in Step 3?

155-161: If the pump/ tank is damaged, likely damage electricity as well. Any thoughts on the correlative/ conjunctive/ collateral damages on electricity? This likely is the limit of focusing only on water systems because water-electricity are very closely related.

186: the only repair is considered? or does repair includes replacement?

202: only water quantity/ pressure is considered? Even though leaks can still convey water, it will have significant impacts on water quality. Also, the authors are considering only the municipal level water systems. After the seismic incidents, maybe many of building water systems can be damaged, which means customers will not be able to get water. any thoughts?

Figure 5: legend for yellow, black dots, etc?

Table 2: what're the objectives of this? How does Table 2 is used in the manuscript/ overall objectives?

222: OK to use Serviceability using equation (5); but I see significant limitations considering only Flow rate. From a citizen point of view, will they really get good water, which is the actual level of service? water quality is critical & infrastructure within the building itself is also critical. The authors should acknowledge the limitations.

406-431: Figure 10 is very informative. However, again, without connecting to other critical infrastructure, if the authors focus on only water infrastructure, this is very limited/ may not be applicable in real life.

 

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

In abstract: Delete "In this study, we improved the seismic damage restoration simulation model of a 12 WDN developed in our previous study."

Page 4, line 167: Please use SI measures (not feet)

Also, equations are not very clear (measures and units are not clearly specified).

References: DOI links are not provided

I do not see significant improvements in respect to the already existing approachs.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Dear Editor

The paper proposes an interesting procedure to study the restoration management of a water distribution network after an earthquake, studying the impact of valve layout on the system performance by adopting a pressure-driven-analysis. The paper is well-written; however, major revisions are strongly recommended for improving the manuscript. In particular, I recommend to rewrite the State of Art, to clarify some crucial aspects of the methodology and to better describe the results. Following are my comments that I would like to see addressed before the acceptance:

The Introduction and the State of the Art have to be improved listing papers dealing with other strategies adopted for the analysis and management of water distribution networks under disasters. Attached some works I suggest to refer to:

• Zhang, Q., Zheng, F., Chen, Q., Kapelan, Z., Diao, K., Zhang, K., & Huang, Y. (2020). Improving the Resilience of Postdisaster Water Distribution Systems Using Dynamic Optimization Framework. Journal of Water Resources Planning and Management, 146(2), 04019075.
• Bałut, A., Brodziak, R., Bylka, J., & Zakrzewski, P. (2019). Ranking Approach to Scheduling Repairs of a Water Distribution System for the Post-Disaster Response and Restoration Service. Water, 11(8), 1591.
• Paez, D., Fillion, Y., & Hulley, M. Battle of post-disaster response and restauration (BPDRR): problem description and rules. In 1st International Water Distribution System Analysis / Computing and Control in the Water Industry Joint Conference, Kingston, Canada, July 23-25, 2018.
• Bristow, E. C., & Brumbelow, K. (2013). Simulation to aid disaster planning and mitigation: Tools and techniques for water distribution managers and emergency planners. Journal of Water Resources Planning and Management, 139(4), 376-386.
• Tabucchi, T., Davidson, R., & Brink, S. (2010). Simulation of post-earthquake water supply system restoration. Civil Engineering and Environmental Systems, 27(4), 263-279.

Could you better explain the step 3 in line 152? What do you want to simulate with the generation of a random number?

Equation 1: could you better explain the coefficients C_i? Where is it possible to read these values? What these coefficient represent? how were these coefficients obtained?

Line 204: what does “water supply capacity” represent? How do you quantify it?

Line 232: “The range of this service suspension area varies based on the number and location of valves installed in the network, which greatly affects the water supply capacity during seismic damage restoration.”. According to the Authors, different topology for the WDN could affect this aspect? A more looped system is expected to have several possible paths for supplying the same area; on the other way around, a branched system is more affected by the closure of the valves for the restoration of the damaged pipes. I think this is another aspect that you should mention in your paper, and deeply investigate in the future; the importance of the topology and network layout;  

Line 262: this is the first part in which you mentioned the concept of district metered areas. After the analysis, as the Authors stated (line 332), the optimal strategy is to exploit the valves already installed for the partitioning of the network. Accordingly, I think that it is important to better describe what does partitioning represent for a water system. I suggest to add some details about this management strategy and refer to:

• Giudicianni, C., Herrera, M., di Nardo, A., & Adeyeye, K. (2020). Automatic multiscale approach for water networks partitioning into dynamic district metered areas. Water Resources Management, 1-14.

It is not clear which characteristic of the earthquake the Authors changed for the different scenarios. It is evident that, simulation results strongly depend on the simulated scenarios especially if you change the spatial position of the epicentre. I suggest to better stress this aspect since if the goal of the proposed procedure is to design the optimal valves layout, several earthquake scenarios, that start from different areas, should be analysed. In this regard, it would be interesting also to test different earthquakes generated from different spatial positions and compare the results for the same valve installation case and restoration rule;

In the section 2.2.1 the Authors described the procedure for simulating the damages for tanks and pumps; anyway, as far as I understood, the analysed case study does not have these elements. Accordingly, it would be better or to analyse another WDN or to point out this lack of consistency;  

Even if the adopted criterion for the valve placement, based on the pipe diameter, it is easy to adopt, it is clear that by using a heuristic optimisation approach the resulting solutions would have been better both in terms of performance and costs. This is an aspect that I suggest to mention in the paper and further investigate;

It would be interesting to analyse the case with only the valves of the DMAs in order to further validate the advantage of having a WDN already partitioned; a Case 9 without Case 5. In this regard, I also suggest to list the diameter of these boundary pipes. It would be interesting to discuss about this further task that water network partition addressed, the protection against earthquake damages in addition to all other benefits. I suggest to refer to:

• Giudicianni, C., Herrera, M., di Nardo, A., Carravetta, A., Ramos, H. M., & Adeyeye, K. (2020). Zero-net energy management for the monitoring and control of dynamically-partitioned smart water systems. Journal of Cleaner Production, 252, 119745.

and pointing out the fact that the valves in this case are already installed in the network for other purposes leading to a strong reduction in terms of costs and a multi-use of the same management strategy;

Line 307: the correct reference should be equation 5? Please check;

The Authors often refer to the economic impact of the different solutions; in this regard, it would be interesting to add also an economic analysis and provide a decision support system in which you compare costs and performance of the obtained solutions;

Figure 7: I suggest to show the curves for the 10 earthquake scenarios instead of the shaded area and better discuss the results, for example, adding some considerations about the time for the restoration for the different solutions;

From Figure 7 it is clear that for all the solutions the initial serviceability is around 0.6 for all the cases. The crucial aspect that affects the subsequent phases is the number of closures. Could you better explain and check how is possible that an initial larger number of closures causes a smaller initial drop of the performance? For example, Case 4, you close all the pipes but the performance reaches the value of 0.9 in less than 1 hour. Furthermore, did you consider that the valves are closed simultaneously? In this regard, some of these scenarios are realistic? How many repair crew do they need? Could you better stress these aspects?

Line 375: what does “time-average system serviceability” represent?

Did you consider that also the valves could be damaged during the earthquakes and affecting the response of the system in terms of suspension areas?

Line 456: the Author stated “As part of future research, the model will be further developed to consider not only the current emergent restoration (i.e., broken pipe repair) but also leakage detection and leakage restoration.”. could you better explain this point? It means that you did not consider the leakages in your case study even if in the methodology you describe how to take into account them? If yes, you should clarify in the paper this lack of consistency and how it is possible to reach the value of 1 for the system serviceability index;

It should be clarified that implicitly you assume that all the pipes are equally good candidates for valve installation (considered equally desirable from a cost and accessibility standpoint) but in real case this is not true; this aspect further validates the choice of exploiting the valves already installed in the network in case of partitioned layout through DMAs;

Finally, the Authors should admit that such an approach is strongly addicted from a deep knowledge about the system, but often this is not available. As a consequence, the proposed approach is a valid tool but requires a large amount of data and become more applicable during the design than during the management.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

The authors presented a very interesting but important topic in this paper - Improved hydraulic simulation of valve layout effects on post-earthquake restoration of a water distribution network. This model is modified and built based on the authors' previous work. Literature review is comprehensive. The case study is clear and the results are well interpreted. Overall speaking, it is good work. However, I do have some editorial and technical comments for the authors to address:

1. Page 2. Line 74. "Jun, and Loganathan [19] ... ". Delete ",". They are two authors for that reference.
2. Page 2. Line 83 to Line 85. "This study aims to improve ... considering the layout of shut-off valves." This is a long sentence kind of hard to follow. Please consider to break it into small sentences.
3. Page 3. Step 4. This paper applies only recovery personnel. In fact, equipment and recovery materials are also very important in reality, which may take additional time be transported to repair location. Therefore, the actual time used to recover will be longer than the numbers in Figure 7 and Figure 8 (x axis). The author should point this out. 
4. Page 4 Figure 2. The authors used the same fragility curve for tanks and pumps. However, in reality, different tanks with different characteristics (e.g. different diameter) have different fragility curves. It will be interesting to use different fragility curve for tanks with different diameters and types. The same for pumps. This can be included as one of future works.
5. Page 5. Equation 3. According to the Equation, the repair time only depends on pipe diameter. However, I would argue that pipes with same diameter but different damage level would require different repair time in reality. Let's say two pipes have same diameter D. Both of them are determined as damaged. One is badly damaged while the other is just damaged. I would envision the badly damaged one will require more repair time than the other. But this can not be reflected in Equation 3. Please consider.
6. Page 7. Line 250. Please provide full name of "CMD" the first time it appears so that readers will better understand.
7. Small error. Page 9 Line 307. "Equation (1)" should be "Equation (5)". Equation 5 is serviceability, not Equation 1.
8. How did the authors determine the number of MCS? In other words, why did the author use 10 MCS in the case study? Please explain. 
9. Page 12 Figure 8. The authors presented the restoration curves for different rules. Based on the discussion, the authors claim that Rule 2 is the best strategy. Well we propose combination strategies instead of single strategy in reality. For example, according to Figure 8. A combination strategy of Rule 3 and Rule 2 would be the most effective and efficient. At the early stage, use Rule 3 - focus on segments requires shorter repair time. Then at certain point. Apply Rule 2 - focus on segments with higher water suspension volume. The key would be identifying the transfer point. This can be another future study area. The author need to mention that in reality, a combination of Rules can be applied.
10. Page 16. Conclusion 2). The authors said "One-end valve installation is regarded to be more cost-effective than installation at both ends of a pipe". Well, this statement might be true most of the time. However, whether it is more cost-effective or not, we need to quantify the additional total economy loss caused by the two layouts, and the cost of installation. Consider the situation where extreme economic loss will occur for some specific region, perhaps both ends installation is more cost-effective than one-end. This statement is not rigorous and please consider to revise.

Hope my comments can help you further improve the paper.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

I do not realize how even as simple formula as Eq. (1) cannot be expressed in SI units (units like ft is absolutely obsolete)

Reviewer 3 Report

The Authors addressed all the comments and suggestions improving the manuscript.