Tendency toward Mega Containerships and the Constraints of Container Terminals
Round 1
Reviewer 1 Report
1. Regression models. At first, please label the axis of Fig. 1, 2 and 3 with units. Please discuss why power curve model was used in the study. For example the second order polynomial model could perform also well.
2. Please discuss also the visible effect of clustering in the basic ships data L, B, and T. For example on Fig. 1 it can be seen the length of ships groups near 366m and 400m, which is the results of several factors, related to the infrastructure like: accessible quay length, turning places, new Panama Canal of Suez Canal parameters etc. How does this factors influence the model accuracy and utilitarian character? For example, Fig.1 maximal limit length 400m seems to be the current limitation of the ships length. The same situation is with the breadth of container ships the limit is set to 59m and seems not be bigger for the same period of time until some infrastructure will not be changed. Ships shall fit to the existing infrastructure otherwise they cannot be operated at all. It influence the model and make it useless.
3. The containerised cargo enforce the ships dimensions discreet grow (stairs - like) especially taking breadth into account (due to designed number of rows) – power curve does not describe such phenomenon. Finalizing those serious remarks, the predictions made in Tab. 6 are mathematical and academic only. They cannot be seriously taken into consideration as the future model of container ships parameters.
4. The predicted ships breadth by KMI model is B=864.1m, please discuss it. The presented by Authors models called “This paper” of LBT in function of TEU give following numbers: for 20.000TEU the results are L=430.5m, B=63.4m, T=16.7m and for 30.000TEU the results are L=453m, B=67.7m, T=17.1m. The numbers presented in Tab. 7 and 8 are different sometimes significantly different in presented Tab. . Please explain this.
1. 3. Authors use the alternative names Power curve and S-curve the proper name shall be power curve in my opinion. Please note that the power curve Y=a*X^b could be presented in linear form as log(Y)=b*log(x)+log(a) and it could be easily presented on linear graph. The linear form Y=bX+a then would be much easier to work with in statistical analysis then non-linear form.
5. What are the units of time in Fig 4 and how they was estimated?
6. Conclusions are not acceptable in this kind of papers. Does not reflect he results. Especially second part of the study – most interesting – where infrastructure related factors are described.
7. Very limited number of literature (16 only). It shall be increased.
8. Please check carefully the language and avoid colloquial statements like for example „ship put into ports” in introduction
Author Response
1. Regression models. At first, please label the axis of Fig. 1, 2 and 3 with units. Please discuss why power curve model was used in the study. For example the second order polynomial model could perform also well.=> The label of Fig. 1,2 and 3 with units is fixed in section 3.3. The reason why we choose the power model is that the standard error(RMSE) is the least one among several models.
2. Please discuss also the visible effect of clustering in the basic ships data L, B, and T. For example on Fig. 1 it can be seen the length of ships groups near 366m and 400m, which is the results of several factors, related to the infrastructure like: accessible quay length, turning places, new Panama Canal of Suez Canal parameters etc. How does this factors influence the model accuracy and utilitarian character? => We discussed the LOA cluster near 366m and 400m, showing new Panama, Suez Cannel dimensions and other factors in section 3.3.
For example, Fig.1 maximal limit length 400m seems to be the current limitation of the ships length. The same situation is with the breadth of container ships the limit is set to 59m and seems not be bigger for the same period of time until some infrastructure will not be changed. Ships shall fit to the existing infrastructure otherwise they cannot be operated at all. It influence the model and make it useless.=> We discussed about the breadth near 59 m which you pointed in section 3.3.
3. The containerised cargo enforce the ships dimensions discreet grow (stairs - like) especially taking breadth into account (due to designed number of rows) – power curve does not describe such phenomenon. Finalizing those serious remarks, the predictions made in Tab. 6 are mathematical and academic only. They cannot be seriously taken into consideration as the future model of container ships parameters. => We made decision to apply linear model to estimate the breadth as you pointed in section 3.3. while the RMSE of the model is not the least. On the breadth estimation, the Power curve are replaced to Linear Model, i.e Y=bX+a reflecting a type of stair trend, even though the RMSE shows more than that of power curve. However, S-curve model cannot be replaced because that actual time series is not following stair type and also the RMSE of Linear model is not the least.
4. The predicted ships breadth by KMI model is B=864.1m, please discuss it. The presented by Authors models called “This paper” of LBT in function of TEU give following numbers: for 20.000TEU the results are L=430.5m, B=63.4m, T=16.7m and for 30.000TEU the results are L=453m, B=67.7m, T=17.1m. The numbers presented in Tab. 7 and 8 are different sometimes significantly different in presented Tab. . Please explain this.=> We fixed it using the other citation in the section 3.3.
5. 3. Authors use the alternative names Power curve and S-curve the proper name shall be power curve in my opinion. Please note that the power curve Y=a*X^b could be presented in linear form as log(Y)=b*log(x)+log(a) and it could be easily presented on linear graph. The linear form Y=bX+a then would be much easier to work with in statistical analysis then non-linear form.=> Statistics package automatically calculates the proper model with the least RMSE and produces the formula(Y=a*X^b) in section 3.3.
6. What are the units of time in Fig 4 and how they was estimated?=> We fixed it in section 4.2.
7. Conclusions are not acceptable in this kind of papers. Does not reflect the results. Especially second part of the study – most interesting – where infrastructure related factors are described.
=>We fixed conclusive issues about berth length, QC outreach, sea depth, ship waiting time, yard capacity and equipment, mega ship calling at the terminal.
8. Very limited number of literature (16 only). It shall be increased.=> We increased them to 21 literature references.
9. Please check carefully the language and avoid colloquial statements like for example „ship put into ports” in introduction=> American professor has checked the statements of paper carefully.
Submission Date
25 February 2019
Date of this review
08 Mar 2019 10:35:49
Author Response File: 
Author Response.docx
Reviewer 2 Report
This paper presents a realistic study to analyse the constraints of ports to handle megaships. The paper first estimates the correlation between TEU capacity and ship dimensions. Then it analyzes the constraints of ports. It is relevant and sound. I am generally positive about the contributions achieved.
But, the paper requires a major revision to improve. All of my following notes should be well addressed in the paper. I think you should make changes for all bullets.
General Comments.
- In abstract, I suggest authors to add one more paragraph about the general findings of this study.
- In the title, "Constrains" should be replaced with "Constraints"
- In the paper, authors use 300 vessels that visited port of Busan and other ports. It is not clear what are other ports. It should be clarified. Authors should also indicate some basic statistics of 300 ships and the interval of visiting dates. I also suggest authors to increase the number of instances by considering the megaships that are currently under construction for liner shipping companies.
- In table 1, LOA is 399.999 and 400 is kind of same. Why do not call all of them 400?
- In table 8, ship breadth is written as 864.1. I think it is wrong. It requires correction.
- I suggest authors to comment about yard capacity as well. As more ships are calling the port, the available stacking area and the number of handling equipment can be limited too. I think you should note it in the paper.
Comments for section 4.1.
- In section 4.1, I suggest authors to conduct further analysis on the effect of tides. I am aware that Rotterdam is tide-dependent port. There are many more tide dependent ports. You should detail the analysis considering tides (available berthing time etc).
https://www.tide-forecast.com/locations/Rotterdam-Netherlands/tides/latest
Comments for section 4.2.
- Is Figure 4 for waiting time before berthing in hours?. Or is it total time spend at the port basin and at berth (total time spent at port)? It is missing in the paper, you should clarify it.
- If authors claim that large ships (more than 10.000TEU) wait 33.4 hours before berthing on average, it means ships wait almost 1.5 days before handling begins. this is too much. Please clarify in the paper how did you obtain these numbers.
- What is the current berth length in PNC? Please specify in section 4.2
Comments for section 2.
- In section 2, I suggest authors to add two paragraphs about optimization studies. Please add following text and three citations in the paper:
Iris et al. and Li et al studied integrated berth allocation and quay crane assignment problem. They try to solve the problem to optimality by considering quay crane coverage ranges (ship breadth). Iris et al. showed that they can efficiently solve the problem for 2 km berth length where 60 ships call the port per week. (Iris et al. 2015, Li et al 2015)
Add references
1- Iris, C., Pacino, D., Ropke, S., Larsen, A., "Integrated Berth Allocation and Quay Crane Assignment Problem: set partitioning models and computational results", Transportation Research Part E: Logistics and Transportation Review 81, 75-97, 2015.
2- Li, F., Sheu, JB, Gao, ZY., "Solving the continuous berth allocation and specific quay crane assignment problems with quay crane coverage range", Transportation Science 49 (4), 968-989, 2015.
- In section 2, add following paragraph and citations about yard management and mega vessels.
Meng, Weng and Suyic simulated the impact of mega vessels on the container port performance. They showed that for scenarios with different container fleet mix and throughputs, there will be a bottleneck at the container terminal if the larger ships continue calling the port. (Meng 2017). Iris et al. optimized loading operations and yard management for mega vessels. For mega vessels, the loading operations can be coordinated with quay cranes, yard trucks and yard equipment (Iris et al., 2018).
Add references
1- Meng, Q., Weng, J., Suyic, L., "Impact Analysis of Mega Vessels on Container Terminal Operations", Transportation Research Procedia, Volume 25, 2017, Pages 187-204.
2- Iris, C., Christensen, J., Pacino, D., Ropke, S., "Flexible ship loading problem with transfer vehicle assignment and scheduling", Transportation Research Part B: Methodological 111, 113-134, 2018.
Language
Abstract, paragraph 2: put --> berths
Abstract, paragraph 2: needs to increase --> needs to be increased
Section 2 title: Precedent --> Relevant
In table 1, write LOA in open form
page 6: as same as --> same as
page 6: recently opened. --> is recently opened.
Author Response
General Comments.
- In abstract, I suggest authors to add one more paragraph about the general findings of this study.
=>Authors fixed it in section 4.2 in the abstract.
- In the title, "Constrains" should be replaced with "Constraints" =>Authors fixed it in section 4.2 in the title.
- In the paper, authors use 300 vessels that visited port of Busan and other ports. It is not clear what are other ports. It should be clarified. =>Authors fixed it in section 4.2 in section 3.3
Authors should also indicate some basic statistics of 300 ships and the interval of visiting dates=>Authors fixed it in section 4.2 in section 3.3
I also suggest authors to increase the number of instances by considering the megaships that are currently under construction for liner shipping companies.
=> Authors found the reference site of order book, but we cannot get the detailed specification of mega ship. Refer the following table.
Delivery Date | Operator | No of Order | Max TEU | |
2019 | MSC | 11 | 22,000 | MSC latest carrier to order 22,000 TEU vessels”. WWW.JOC.COM |
2020 | HMM | 12 | 23,000 | HMM orders mega-ships amid shareholder frustration, https://www.joc.com/ |
2019 | CMA CGM | 9 | 22,000 | SC latest carrier to order 22,000 TEU vessels”. WWW.JOC.COM |
2018-2019 | COSCO | 6 | 2 21,237 | “Ccosco raises $1.94bn to fund 20 boxships” Splash 247, October 2017. |
2018 | CMA CGM | 3 | 20,656 | “ SC latest carrier to order 22,000 TEU vessels”. WWW.JOC.COM |
2018-2019 | Evergreen | 11 | 20,000 | “Japan’s Imabari Shipbuilding to expand: IHS Fairplay Fairplay.ihs.com.”
|
- In table 1, LOA is 399.999 and 400 is kind of same. Why do not call all of them 400? => Authors try to keep formal figures following shipping company’s public announcement
- In table 8, ship breadth is written as 864.1. I think it is wrong. It requires correction.: =>Authors fixed it.
- I suggest authors to comment about yard capacity as well. As more ships are calling the port, the available stacking area and the number of handling equipment can be limited too. I think you should note it in the paper.=> Authors fixed it in section 3.4.
Comments for section 4.1.
- In section 4.1, I suggest authors to conduct further analysis on the effect of tides. I am aware that Rotterdam is tide-dependent port. There are many more tide dependent ports. You should detail the analysis considering tides (available berthing time etc).=> Authors fixed it in section 4.1
https://www.tide-forecast.com/locations/Rotterdam-Netherlands/tides/latest
Comments for section 4.2.
- Is Figure 4 for waiting time before berthing in hours?. Or is it total time spend at the port basin and at berth (total time spent at port)? It is missing in the paper, you should clarify it.
- If authors claim that large ships (more than 10.000TEU) wait 33.4 hours before berthing on average, it means ships wait almost 1.5 days before handling begins. this is too much. Please clarify in the paper how did you obtain these numbers =>Authors fixed it in section 4.2
- What is the current berth length in PNC? Please specify in section 4.
Comments for section 2. =>Authors fixed it in section 4.2
- In section 2, I suggest authors to add two paragraphs about optimization studies. Please add following text and three citations in the paper:
Iris et al. and Li et al studied integrated berth allocation and quay crane assignment problem. They try to solve the problem to optimality by considering quay crane coverage ranges (ship breadth). Iris et al. showed that they can efficiently solve the problem for 2 km berth length where 60 ships call the port per week. (Iris et al. 2015, Li et al 2015)
Add references =>Authors fixed it in references
1- Iris, C., Pacino, D., Ropke, S., Larsen, A., "Integrated Berth Allocation and Quay Crane Assignment Problem: set partitioning models and computational results", Transportation Research Part E: Logistics and Transportation Review 81, 75-97, 2015.
2- Li, F., Sheu, JB, Gao, ZY., "Solving the continuous berth allocation and specific quay crane assignment problems with quay crane coverage range", Transportation Science 49 (4), 968-989, 2015.
- In section 2, add following paragraph and citations about yard management and mega vessels.
Meng, Weng and Suyic simulated the impact of mega vessels on the container port performance. They showed that for scenarios with different container fleet mix and throughputs, there will be a bottleneck at the container terminal if the larger ships continue calling the port. (Meng 2017). Iris et al. optimized loading operations and yard management for mega vessels. For mega vessels, the loading operations can be coordinated with quay cranes, yard trucks and yard equipment (Iris et al., 2018).
Add references
1- Meng, Q., Weng, J., Suyic, L., "Impact Analysis of Mega Vessels on Container Terminal Operations", Transportation Research Procedia, Volume 25, 2017, Pages 187-204.
2- Iris, C., Christensen, J., Pacino, D., Ropke, S., "Flexible ship loading problem with transfer vehicle assignment and scheduling", Transportation Research Part B: Methodological 111, 113-134, 2018.
Language=> We fixed it following the advice of American Professor Dr. Ross
Abstract, paragraph 2: put --> berths
Abstract, paragraph 2: needs to increase --> needs to be increased
Section 2 title: Precedent --> Relevant
In table 1, write LOA in open form
page 6: as same as --> same as
page 6: recently opened. --> is recently opened.
Submission Date
25 February 2019
Date of this review
05 Mar 2019 08:49:31
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
Thank you for corrections
Author Response
A American professor, named Dr. Corey Ross reviewed and modified English language and style and spell according to reviewer request. Specially section 4.2 is modified significaltly as the request of reviewer 2.
Thanks a lor for your comment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This study has improved significantly after the revision. Authors have made important and valuable changes. But it requires one more revision.
- In section 4.1., the analysis of tides should be mathematically calculated and detailed. Authors know the tide intervals during the day from different websites. The time window that each port can handle large ships should be mentioned by considering tides.
- In section 2, authors should add following paragraphs in the paper. Please add following text:
Iris et al. and Li et al studied integrated berth allocation and quay crane assignment problem. They try to solve the problem to optimality by considering quay crane coverage ranges (ship breadth). Iris et al. showed that they can efficiently solve the problem for 2 km berth length where 60 ships call the port per week. [18,23]
- In section 2, add following paragraph and citations about yard management and mega vessels.
Meng, Weng and Suyic simulated the impact of mega vessels on the container port performance. They showed that for scenarios with different container fleet mix and throughputs, there will be a bottleneck at the container terminal if the larger ships continue calling the port [20]. Iris et al. optimized loading operations and yard management for mega vessels. For mega vessels, the loading operations can be coordinated with quay cranes, yard trucks and yard equipment [21].
Author Response
This study has improved significantly after the revision. Authors have made important and valuable changes. But it requires one more revision.
- In section 4.1., the analysis of tides should be mathematically calculated and detailed. Authors know the tide intervals during the day from different websites. The time window that each port can handle large ships should be mentioned by considering tides. +> Authors fixed it in section 4.1 significaltly.
- In section 2, authors should add following paragraphs in the paper. Please add following text:
Iris et al. and Li et al studied integrated berth allocation and quay crane assignment problem. They try to solve the problem to optimality by considering quay crane coverage ranges (ship breadth). Iris et al. showed that they can efficiently solve the problem for 2 km berth length where 60 ships call the port per week. [18,23] +> Authors fixed it in section 2.
- In section 2, add following paragraph and citations about yard management and mega vessels.
Meng, Weng and Suyic simulated the impact of mega vessels on the container port performance. They showed that for scenarios with different container fleet mix and throughputs, there will be a bottleneck at the container terminal if the larger ships continue calling the port [20]. Iris et al. optimized loading operations and yard management for mega vessels. For mega vessels, the loading operations can be coordinated with quay cranes, yard trucks and yard equipment [21]. Authors added them in section 2.
Authors revised English language and syntax with the help of America professor Dr. Corely Ross.
Author Response File: Author Response.pdf
Round 3
Reviewer 2 Report
I am significantly content with the revision. I suggest editors to accept the paper for publication. Congratulations.
