Basic Ship-Planning Support System Using Big Data in Maritime Logistics for Simulating Demand Generation

Round 1

Reviewer 1 Report

I am thankful for the opportunity to read this insightful paper.

It opens room for new original and useful experiments in the optimal ships allocation at high scale.

Suggestions for the authors:

1. To take into account S-AIS, besides AIS. Ref., e.g.: https://www.researchgate.net/publication/335160557_A_Review_of_NAVDAT_and_VDES_as_Upgrades_of_Maritime_Communication_Systems
2. To explain in some more detail used deep learning algorithm being applied.
3. To consider weather routing in further investigation and to at least mention this important parameter in this research.

I recommend the paper for publishing. 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The paper suggests a stage-wise analytics approach for tramp shipping vessel management. Although there is a merit in the work, I think paper needs an extensive revision on several components. All of following comments are important and should be addressed:

1) This paper seems to split from real practice. The existing tramp shipping market does not work as authors' claim. In tramp shipping, there is no fixed routes. Therefore, there can be no consistent network as ships are fixed to contracts in short notice in many cases. Ship brokers finds the most convenient dry bulk ship for each cargo demand. This study should have a clear discussion about how tramp shipping market works and make your paper realistic with respect to actual practice.

2) In equation 1, even if the ship sails ballast, ship has a draught value. Therefore, there is a risk that equation 1 can obtain a positive cargo amount even when she sails ballast. How do you ensure this does not happen? 

3) Equations 1 and 2 are not clear. Authors can show an example calculation for equation 2 using data in table 2. 

4) The details for proposed deep learning are very limited. You should explain figure 3 a lot more. 

draught, speed estimation and port stay times depend heavily on the contract. But, in figure 3, there is no clear input for contract details for specific output.

How do you train your DL?

5) The introduction section should be improved. The motivation for solving this problem is not clear. You should add one more paragraph to explain the motivation, research questions and contributions in the introduction.

6) The literature review is missing a subsection about optimisation studies in ship deployment and contracts. The current literature review is only about big data in shipping. However, there are several optimisation and mathematical modelling papers on ship planning. You should cite and discuss optimisation studies in shipping. Additionally Zhang et al and Venturini et al use speed vs emission relationship and they can be cited at line 263:

Zhang, X., Lam, J.S.L. and Iris, Ç., 2020. Cold chain shipping mode choice with environmental and financial perspectives. Transportation Research Part D: Transport and Environment, 87, p.102537.

Lin, D.Y. and Liu, H.Y., 2011. Combined ship allocation, routing and freight assignment in tramp shipping. Transportation Research Part E: Logistics and Transportation Review, 47(4), pp.414-431.

Venturini, G., Iris, Ç., Kontovas, C.A. and Larsen, A., 2017. The multi-port berth allocation problem with speed optimization and emission considerations. Transportation Research Part D: Transport and Environment, 54, pp.142-159.

Arslan, A.N. and Papageorgiou, D.J., 2017. Bulk ship fleet renewal and deployment under uncertainty: A multi-stage stochastic programming approach. Transportation Research Part E: Logistics and Transportation Review, 97, pp.69-96.

7) In section 3.1, data used is explained. However, this section misses explanation on data pre-processing elements such as data cleansing, missing data procedures, etc. You should add relevant explanation.

8) Details in section 5.2.2 are very limited and explanations should be improved.

9) Limitations of this work considering assumptions should be noted in conclusion section.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

The revision is very comprehensive and addresses the needs arising from the review report. In this round, minor modifications are needed before publication;

1. Does the study make any difference between time charter or voyage charter?. From description, it sounds like this is more applicable to time charter market. You may want to note this in the paper.
2. Following papers use AIS data and conducts data analytics applications in tramp shipping market as your study. I encourage you to carefully read the following works and cite and discuss them in your paper:

https://doi.org/10.1016/j.tre.2022.102617

https://doi.org/10.1080/01441647.2019.1649315

Bai, X.; Liangqi, C.; Iris, C. Data-driven financial and operational risk management: Empirical evidence from the global tramp shipping industry. Transportation Research Part E: Logistics and Transportation Review, 2022, Volume 158, 102617, doi.org/10.1016/j.tre.2022.102617.

Yang, D., Wu, L., Wang, S., Jia, H. and Li, K.X., 2019. How big data enriches maritime research–a critical review of Automatic Identification System (AIS) data applications. Transport Reviews, 39(6), pp.755-773, doi.org/10.1080/01441647.2019.1649315.

3. On line 179, "into voyage records". On line 177, "dates of arrival and departure".  On line 178 "draughts of arrival and departure". On line 305, "despite of...". On line 350, why did you delete figure 3 caption? 

Author Response

Please see the attachment.

Author Response File: Author Response.docx