A Spatiotemporal Statistical Method of Ship Domain in the Inland Waters Driven by Trajectory Data

Round 1

Reviewer 1 Report

The article is devoted to current and important issues paying attention to the determination of ship domain in the inland waters driven by trajectory data. The article has an adequate theoretical basis, relevant information and analysis, good partial (in the article) and final (in the conclusion) conclusions. The article uses the author's original research, which enriches its content and can add value to the area under study. The analyses, on the other hand, are well grounded in theory and supported by the results of other theoretical and empirical studies. The paper presents a statistical method for conducting a spatial-temporal analysis of ship domain in the inland waters driven by using AIS data. The authors have undertaken an analysis of an area of research with high utility value to the economy, with the correct selection and application of statistical methods. The article is written in good language and is based on the analysis of current and well-chosen literature, although it could be further enriched with other items. Thus, the article should be treated rather as an interesting introduction to a very important issue and treated as a scientific article. It is an original concept that inspires deeper research.
In the introduction, there is no clearly defined aim and research thesis, which should be completed.  

Author Response

Point1: The article is devoted to current and important issues paying attention to the determination of ship domain in the inland waters driven by trajectory data. The article has an adequate theoretical basis, relevant information and analysis, good partial (in the article) and final (in the conclusion) conclusions. The article uses the author's original research, which enriches its content and can add value to the area under study. The analyses, on the other hand, are well grounded in theory and supported by the results of other theoretical and empirical studies. The paper presents a statistical method for conducting a spatial-temporal analysis of ship domain in the inland waters driven by using AIS data. The authors have undertaken an analysis of an area of research with high utility value to the economy, with the correct selection and application of statistical methods. The article is written in good language and is based on the analysis of current and well-chosen literature, although it could be further enriched with other items. Thus, the article should be treated rather as an interesting introduction to a very important issue and treated as a scientific article. It is an original concept that inspires deeper research.
In the introduction, there is no clearly defined aim and research thesis, which should be completed.

Response1: Thanks for your suggestions. A more complete introduction has been added in the revised manuscript, including the research motivations, the research objectives, and the research thesis of this study.

Reviewer 2 Report

The manuscript entitled “A spatiotemporal statistical method of ship domain in the inland waters driven by trajectory data” raises an interesting issue of statistical ship domain applicable on inland waters. The manuscript is generally well-structured, there are however many issues impacting the scientific foundations of the study and reliability of the results. Thus, this paper cannot be considered for publication. The major issues are as follows:

 

1. The objectives of the study and especially the novelty are not clearly stated in the introduction.

 

2. In the reviewer’s opinion, the authors are mixing two types (as per their application) of ship domains. Firstly, this concept was introduced to assess the waterway capacity, see (Fujii and Tanaka, 1971). Then, this idea was adopted to maritime collision-avoidance but only as a parallel research-line. It is not clear from the Abstract and Introduction what kind of ship domain is elaborated by the authors and for what purpose.

 

3. Why the authors assumed that: “The target ship and the encountered ships are the research objects, and the minimum safety distance between which is boundary of the ship domain”? Maybe the ships are not passing each other at a ”safe distance”. This assumption impacts directly the entire study.

 

4. Did the authors use the antenna position which is provided via AIS? I do not see information about that within the text. This will significantly increase the accuracy of the results.

 

5. The insights and analysis of the results are overshaded by many figures (see Fig. 16-19; Fig. 20-22, etc.). Please decrease their number and provide more description about the outcome of the results and their analysis.

 

6. In the Discussion section mainly the limitations are elaborated – I would like to see more discussion about the results, future work, etc.

 

7. There are many limitations impacting this study, which make the results unreliable in the reviewer’s opinion.

 

Minor issues:

1. “There are some factors should be considered when ships sailing on the inland waters, i.e. the navigation rules, meteorological conditions, hydrologic conditions.” Only in inland waters? The same should be taken into account in restricted sea waters.

 

2. “Józef et al., 2000;” should be „Lisowski J. et al., 2000;”

 

3. There are also concepts called ”critical areas” which are a kind of ship domain for collision-avoidance purposes. There are many works analyzing the influence of the ship maneuvering or environmental conditions on the shape and size of the critical area. See, for instance, (Gil et al., 2020a, 2020b; Montewka et al., 2020). There were also many studies utilizing ship domains in risk assessment/near-miss detection analyses based on AIS data, which should be mentioned in the literature see e.g. (Zhang et al., 2020, 2016).

 

4. “Ships are divided into three categories based on the ship length, 60 to 80m, 80 to 100m, 100 to 120m.” (page 6). So to which group authors will classify a ship that is exactly 80 or 100 m in length? Later on, another classification is provided: “The ships were divided into three groups based on the ship length, 60 to 79m, 80 to 99m, and 100 to 120m, respectively.” (page 8). Please unify the breakdowns. Figure 8 also gives another classification…

 

5. “Fig. 7 shows the study area from 30.533 to 30.706°E and 114.296 to 114.508°N.” (4.1, p. 8) The hemispheres are miss-written. The latitude reaches 90deg in maximum, while the longitude 180deg.

 

6. In Figure 9 the units are missing (traffic volume).

 

7. Proofreading with the assistance of a native speaker should be made as there are still many grammar mistakes, typos and editorial issues.

 

References:

Fujii, Y., Tanaka, K., 1971. Traffic Capacity. Journal of Navigation 10.

Gil, M., Montewka, J., Krata, P., Hinz, T., Hirdaris, S., 2020a. Determination of the dynamic critical maneuvering area in an encounter between two vessels: Operation with negligible environmental disruption. Ocean Engineering 213, 107709. https://doi.org/10.1016/j.oceaneng.2020.107709

Gil, M., Montewka, J., Krata, P., Hinz, T., Hirdaris, S., 2020b. Semi-dynamic ship domain in the encounter situation of two vessels, in: Soares, C.G. (Ed.), Developments in the Collision and Grounding of Ships and Offshore Structures. Taylor and Francis Group, London, pp. 301–307.

Montewka, J., Gil, M., Wróbel, K., 2020. Discussion on the article by Zhang & Meng entitled “Probabilistic ship domain with applications to ship collision risk assessmentˮ [Ocean Eng. 186 (2019) 106130]. Ocean Engineering 209, 107527. https://doi.org/10.1016/j.oceaneng.2020.107527

Zhang, W., Feng, X., Goerlandt, F., Liu, Q., 2020. Towards a Convolutional Neural Network model for classifying regional ship collision risk levels for waterway risk analysis. Reliability Engineering & System Safety 204, 107127. https://doi.org/10.1016/j.ress.2020.107127

Zhang, W., Goerlandt, F., Kujala, P., Wang, Y., 2016. An advanced method for detecting possible near miss ship collisions from AIS data. Ocean Engineering 124, 141–156. https://doi.org/10.1016/j.oceaneng.2016.07.059

Author Response

The manuscript entitled “A spatiotemporal statistical method of ship domain in the inland waters driven by trajectory data” raises an interesting issue of statistical ship domain applicable on inland waters. The manuscript is generally well-structured, there are however many issues impacting the scientific foundations of the study and reliability of the results. Thus, this paper cannot be considered for publication. The major issues are as follows:

Point1: The objectives of the study and especially the novelty are not clearly stated in the introduction.

Response1: Thank you for your suggestion. The research motivations and the objectives of this work have been fixed in the introduction.

Point2: In the reviewer’s opinion, the authors are mixing two types (as per their application) of ship domains. Firstly, this concept was introduced to assess the waterway capacity, see (Fujii and Tanaka, 1971). Then, this idea was adopted to maritime collision-avoidance but only as a parallel research-line. It is not clear from the Abstract and Introduction what kind of ship domain is elaborated by the authors and for what purpose.

Response2: Thank you for your comments. The idea of ship domain is adopted to ship collision avoidance (Fujii and Tanaka, 1971), the explanations of which have been added in abstract and introduction.

Point3: Why the authors assumed that: “The target ship and the encountered ships are the research objects, and the minimum safety distance between which is boundary of the ship domain”? Maybe the ships are not passing each other at a ”safe distance”. This assumption impacts directly the entire study.

Response3: Thank you for your comments. The phenomena really exist that two ships are not keeping a safe distance when they are passing each other. Two ships would be maneuvered back to a safe distance as soon as possible when they are not at a safe distance, therefore, it rarely happens ships are not keeping a safe distance throughout the voyage. In this work, the whole navigational trajectories of object ships in the research area are utilized to dynamic statistic the grid density of the nearest other ships and then establishing ship domain. Therefore, this assumption has little influence on the entire study, and it will be considered to improve the reliability of the inland ship domains in our future works. Based on your comments, we have modified Section 3 in the revised manuscript.

Point4: Did the authors use the antenna position which is provided via AIS? I do not see information about that within the text. This will significantly increase the accuracy of the results.

Response4: Thank you for your comments. In this study, the antenna of AIS equipment is regarded as to be installed on the center of the ship, AIS data provides the ship’s position is the center position of the ship. Based on the ship’s position obtained from AIS data, ship width, and ship length, the location range of the ship can be calculated. The detailed explanations have been added in Session 3.2, please check it.

Point5: The insights and analysis of the results are overshaded by many figures (see Fig. 16-19; Fig. 20-22, etc.). Please decrease their number and provide more description about the outcome of the results and their analysis.

Response5: Thank you for your suggestions. Fig. 16-19 were removed to Appendix IV in the revised manuscript. More explanation and description about results have been added in Session 4.2.2, please check it.

Point6: In the Discussion section mainly the limitations are elaborated – I would like to see more discussion about the results, future work, etc.

Response6: Thank you for your suggestions. A more detailed discussion has been added to the manuscript, please check it.

Point7: There are many limitations impacting this study, which make the results unreliable in the reviewer’s opinion.

Response7: Thank you for your comments. In this work, there are three factors that would influence the reliability of results. The AIS equipment is assumed to be installed on the center of the ship because of the lack of the installed position of AIS equipment. The error of the foundation data would influence the results. It is impossible to obtain the whole activity data of the regional ships, therefore, the reliability of the established ship domains is also influenced by the quality of the foundation data. In future works, we will attempt to improve the reliability of the establishment method of ship domain in inland waters. This is discussed in Section 5 in the revised manuscript.

Minor issues:

Point8: “There are some factors should be considered when ships sailing on the inland waters, i.e. the navigation rules, meteorological conditions, hydrologic conditions.” Only in inland waters? The same should be taken into account in restricted sea waters.

Response8: Thank you for your comments. In the Yangtze River, there are obviously differences between the water levels in the summer and in the winter because of the influences of the meteorological conditions and the hydrologic conditions. Due to the influences of the navigation rules, the shape and the size of ship domains in inland waters have significant differences between and the restricted sea waters, it has been analyzed in Session 4.2.2.

Point9: “Józef et al., 2000;” should be „Lisowski J. et al., 2000;”

Response9: Thank you for pointing it out. It has been fixed.

Point10: There are also concepts called ”critical areas” which are a kind of ship domain for collision-avoidance purposes. There are many works analyzing the influence of the ship maneuvering or environmental conditions on the shape and size of the critical area. See, for instance, (Gil et al., 2020a, 2020b; Montewka et al., 2020). There were also many studies utilizing ship domains in risk assessment/near-miss detection analyses based on AIS data, which should be mentioned in the literature see e.g. (Zhang et al., 2020, 2016).

Response10: Thank you for your suggestions. It has been fixed in the Literature Review.

Point11: “Ships are divided into three categories based on the ship length, 60 to 80m, 80 to 100m, 100 to 120m.” (page 6). So to which group authors will classify a ship that is exactly 80 or 100 m in length? Later on, another classification is provided: “The ships were divided into three groups based on the ship length, 60 to 79m, 80 to 99m, and 100 to 120m, respectively.” (page 8). Please unify the breakdowns. Figure 8 also gives another classification…

Response11: Thank you for your point out. In the revised manuscript, ships are divided into three categories based on ship length, 60 to 79m, 80 to 99m, and 100 to 120m. This classification was utilized in Fig.8 in the revised manuscript, please check it.

Point12: “Fig. 7 shows the study area from 30.533 to 30.706°E and 114.296 to 114.508°N.” (4.1, p. 8) The hemispheres are miss-written. The latitude reaches 90deg in maximum, while the longitude 180deg.

Response12: Thank you for your point out. It has been fixed.

Point13: In Figure 9 the units are missing (traffic volume).

Response13: Thank you for your pointing it out. It has been fixed.

Point14: Proofreading with the assistance of a native speaker should be made as there are still many grammar mistakes, typos and editorial issues.

Response14: Thank you for your suggestion. The academic writing of this paper has been improved.

References:

Fujii, Y., Tanaka, K., 1971. Traffic Capacity. Journal of Navigation 10.

Gil, M., Montewka, J., Krata, P., Hinz, T., Hirdaris, S., 2020a. Determination of the dynamic critical maneuvering area in an encounter between two vessels: Operation with negligible environmental disruption. Ocean Engineering 213, 107709. https://doi.org/10.1016/j.oceaneng.2020.107709

Gil, M., Montewka, J., Krata, P., Hinz, T., Hirdaris, S., 2020b. Semi-dynamic ship domain in the encounter situation of two vessels, in: Soares, C.G. (Ed.), Developments in the Collision and Grounding of Ships and Offshore Structures. Taylor and Francis Group, London, pp. 301–307.

Montewka, J., Gil, M., Wróbel, K., 2020. Discussion on the article by Zhang & Meng entitled “Probabilistic ship domain with applications to ship collision risk assessmentˮ [Ocean Eng. 186 (2019) 106130]. Ocean Engineering 209, 107527. https://doi.org/10.1016/j.oceaneng.2020.107527

Zhang, W., Feng, X., Goerlandt, F., Liu, Q., 2020. Towards a Convolutional Neural Network model for classifying regional ship collision risk levels for waterway risk analysis. Reliability Engineering & System Safety 204, 107127. https://doi.org/10.1016/j.ress.2020.107127

Zhang, W., Goerlandt, F., Kujala, P., Wang, Y., 2016. An advanced method for detecting possible near miss ship collisions from AIS data. Ocean Engineering 124, 141–156. https://doi.org/10.1016/j.oceaneng.2016.07.059

Author Response File: Author Response.docx

Reviewer 3 Report

The proposed method and results are valuable. This discrete approach has some advantages for the graphical presentation of ships' domains. Unfortunately, I see some missing issues.. Some modifications and extensions could help to improve your paper.

1. Is there some law regulations for Yangtze River inland ship traffic? If yes, the authors should present them in the Introduction or Literature Review.
2. I did not find the information for what you do your research about ships' domains. Please, give this information, for example, in the Conclusion section. Is it about the safety of traffic flow on the Yangtze River? Maybe, your goal is different. But it is important to give this information to readers. The paper will be more readable.
3. Please, give more precise information for formulae (1) - (4) and graphs in figures 16 - 19. I do not see or maybe not understand your concept of results' presentation.

I think the paper presents valuable results. But it should be improved. I suggest accepting the paper after minor revision. 

Author Response

The proposed method and results are valuable. This discrete approach has some advantages for the graphical presentation of ships' domains. Unfortunately, I see some missing issues. Some modifications and extensions could help to improve your paper.

Point1: Are there some law regulations for Yangtze River inland ship traffic? If yes, the authors should present them in the Introduction or Literature Review.

Response1: Thank you for your suggestions. There are some law regulations for the Yangtze River inland ship traffic, the introduction of which has been added in the introduction. It has been added in the Introduction.

 

Point2: I did not find the information for what you do your research about ships' domains. Please, give this information, for example, in the Conclusion section. Is it about the safety of traffic flow on the Yangtze River? Maybe, your goal is different. But it is important to give this information to readers. The paper will be more readable.

Response2: Thank you for your suggestions. It has been fixed in the Conclusion section.

 

Point3: Please, give more precise information for formulae (1) - (4) and graphs in figures 16 - 19. I do not see or maybe not understand your concept of results' presentation.

Response3: Thank you for your suggestions. Section 4.2.2 has been modified, and a more detailed analysis of results has also been added.

I think the paper presents valuable results. But it should be improved. I suggest accepting the paper after minor revision.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

I appreciate the authors’ effort made to the revised version of the JMSE-1134025 manuscript. In the last attempt, most of my crucial comments have been properly addressed, while the manuscript has been significantly improved. Therefore, after solving one, minor issue I can recommend the manuscript for publication.

 

1. The authors stated: “(…) area from 30.533°E to 30.706°E, and from 114.296°N to 114.508°N.” (p. 9). Please correct the hemispheres here (latitude ranges up to 90°, thus it cannot equals 114°. Also, double-check all figures and indexes used in notation of geographical positions within the entire manuscript.

Author Response

Point1: The authors stated: “(…) area from 30.533°E to 30.706°E, and from 114.296°N to 114.508°N.” (p. 9). Please correct the hemispheres here (latitude ranges up to 90°, thus it cannot equals 114°. Also, double-check all figures and indexes used in notation of geographical positions within the entire manuscript.

Response1: Thank you for pointing it out. It has been revised at line 284.