Risk Identification Method for Ship Navigation in the Complex Waterways via Consideration of Ship Domain

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper proposes a risk identification method in the context of a complex waterway, aligning with the scope of the journal. The following points need to be addressed:

 

- The novelty of the research should be derived through differentiation from prior studies. While previous research is mentioned, please specify its limitations and emphasize the uniqueness of your approach.

- Please state the clear purpose of the paper in the introduction.

- Provide information about the data collection period for AIS data used in developing the ship domain model.

- Specify what the X and Y axes represent and include the units for all graphs.

- Please explain what Equation B and L represent.

- When selecting density data, the author applied thresholds such as 40% and 15%. Please provide the rationale for these choices.

- The discussion section is lacking. It predominantly describes experimental results. Use the discussion to explain the significance of the results obtained.

- There is no validation of the proposed method.

- Please provide the rationale for dividing levels in the proposed method.

 

Author Response

Dear Reviewer:

Thank you very much for your comments concerning our manuscript entitled “Risk Identification Method for Ship Navigation in the Complex Waterways via Consideration of Ship Domain”. Those comments are all valuable and very helpful for revising and improving our paper, and have an important guiding significance to our researches as well. We have checked the manuscript and revised it according to the comments. The main corrections in the paper and the replies to the comments are as follows.

1. The novelty of the research should be derived through differentiation from prior studies. While previous research is mentioned, please specify its limitations and emphasize the uniqueness of your approach.

Reply：According to this comment, the uniqueness of our approach compared to other previous research is supplemented in Section 1, paragraph 5. Our approach mainly focuses on the characteristics of navigation separation in Yangtze River waterways for ship navigation risk identification research. As follows:

“In summary, on the one hand, these studies mainly focus on open water areas, and there is a lack of empirical research on ship navigation risks based on complex inland waterways. On the other hand, research methods based on statistical analysis use actual traffic flow data. Early research was mostly based on radar data, and in recent years, it has shifted towards AIS data. However, for research on the risk identification with ship domain in the complex waterways, qualitative research is often used, and data containing ship navigation information is rarely used for quantitative research. Thus, while taking the risk identification computations in complex waterways, a ship domain model considering the characteristics of navigation separation based on AIS data needs to be developed.”

2. Please state the clear purpose of the paper in the introduction.

Reply：According to this comment, we have stated the clear purpose of the paper in Section 1, paragraph 6.

“By this way, we can evaluate the collision risk of ship navigation based on AIS data, and assist ship navigation by providing collision avoidance decision-making basis for navigators.”

3. Provide information about the data collection period for AIS data used in developing the ship domain model.

Reply： We agree with it. We have provided information about the data collection period for AIS data in Section 3.

“The selected data is sourced from AIS onboard equipment on the ship "Channel 1", which includes over 2.7 million AIS and GPS data records of the ships from March 2023 to July 2023 in the Wuhan reach of the Yangtze River.”

4. Specify what the X and Y axes represent and include the units for all graphs.

Reply：According to this comment, we have specified what the X and Y axes represent and included the units for all graphs in our manuscript.

5. Please explain what Equation B and L represent.

Reply：Equation B is the abbreviation of equation for the safety distance boundary in the ship domain, which is formulas (3) in Section 3.5. And Equation L is the abbreviation of equation for the collision risk detection line, which is formulas (11) in Section 4.1.

6. When selecting density data, the author applied thresholds such as 40% and 15%. Please provide the rationale for these choices.

Reply：According to research conducted with experienced navigators, the safe distance between the own ship and the target ship ahead in the same direction is about 200m in the Yangtze River. In Figure 10 (a), the Normalized ship distribution density corresponding to a distance of 200 from the target ship ahead is about 40%. Therefore, we chose 40% in Figure 10 and Figure 11. But in Figure 10 (b), choosing 40% is clearly inappropriate, so a moderate 15% was chosen. The main purpose here is to determine the eccentric position of the ellipse. In order to ensure the availability of the final ship field, adjustments will be made in the subsequent formulas(3) through k_a and k_b.

.

(a) the same direction                        (b) the opposite direction.

Figure 10. Density curve of the ship under x₀=0 section for ship navigation.

7. The discussion section is lacking. It predominantly describes experimental results. Use the discussion to explain the significance of the results obtained.

Reply： We agree with it. And we have added discussion.

“In this paper, a new method for identifying ship navigation risks by combining the ship domain with AIS data is proposed, in which the collision time with the target ship is calculated based on the collision hazard detection line and safety distance boundary, and formed a method for dividing the danger level of the ship navigation situation.

The research focuses on the Wuhan reach of the Yangtze River, where ships navigate in accordance with the Regulations for Preventing Collisions in Inland Rivers of the People's Republic of China. In the complex waterway of the Yangtze River, the simulation results of our method are consistent with the actual navigation behavior and superior to the traditional TCPA method. This also indicates that our method can accurately reflect the consolidation risk of the target ship relative to the own ship, and the warning data is stable and has good real-time performance, which can effectively assist the navigator in identifying consolidation risks during navigation.

However, our method did not consider the hydrodynamic interaction between the ships. The external hydrodynamic interference, such as waves and currents, is one of the important influencing factors for ship navigation decisions. Our method has limitations by determining a reasonable size of the ship domain and avoiding the impact of hydrodynamic interference through navigators. Due to the lack of actual flow field data, poor ship informatization conditions, and the lack of equipment such as recorders and meteorological instruments, our research cannot consider external disturbances. In addition, hydrodynamic simulation requires a certain amount of computational power support, which may improve the accuracy of early warning but reduce the speed of early warning.

In the future, we will conduct research on rapid simulation and early warning methods that consider hydrodynamic interaction based on our proposed method.”

8. There is no validation of the proposed method.

Reply：We agree with it.  It is very important to our proposed method. The validation of our manuscript has been listed in Section 5, and we have added a comparison with traditional methods (DCPA and TCPA) to validate our proposed method.

“

 

Figure 18. Relationship between Tca and heading change rate

The change rate of the own ship's heading objectively reflects the behavior of the ship's driving operation. In other words, when the navigator discovers that there is a collision risk of the target ship, he will take corresponding avoidance actions, resulting in a change in the ship's heading. Therefore, comparing the Tca change curve of the fused data with the change rate curve of the own ship's heading can prove whether the collision risk judgment result is consistent with the navigator's risk judgment result, which can also verify the reliability of our risk identification method.

Figure 18 shows that at 15:50:00, our method began to alert collision risks from the target ship, and almost at the same time, the rate of the own ship’s heading change gradually increased, indicating that the navigator also judged that there was a risk of collision from the target ship and turned the rudder. Until 15:50:32, our method ends the collision risk alarm for the target ship and the rate of the own ship's heading change is gradually decreasing, indicating that at this time, the navigator also determines that the collision risk of the target ship is relieved, and turns the rudder in reverse to return to the original heading for navigation. Overall, our method's collision risk alarm judgment is consistent with the navigator's judgment, so our method is effective in identifying the collision risk during navigation and in line with the navigator's cognitive habits in other word, our proposed method has been validated.”

Figure 19. Comparison between our method and TCPA (the blue line is T_ca and The yellow line is the TCPA)

According to traditional methods, DCPA less than 200m and TCPA less than 80s are used as alarm judgment conditions, where DCPA is less than 100 meters throughout the entire process, so only the changes in TCPA are considered[].

Figure 19 shows that our proposed method is more accurate than the traditional DCPA and TCPA judgment methods for early warning. Firstly, the traditional DCPA and TCPA judgment methods lag by 8 seconds before starting to alarm. Secondly, after the navigation risk alarm, the traditional methods always maintain the alarm state for a long time, and only stop the alarm after the ship has passed and cleared, which is inconsistent with the driver's judgment of navigation risk and cannot accurately reflect the navigation situation.

9. Please provide the rationale for dividing levels in the proposed method.

Reply：We have classified the risk levels in our method based on the research conducted by Li Lina's team at Jimei University.

The evaluation system consists of two parts: the threshold of evaluation indicators and the classification of evaluation levels. The evaluation index threshold system consists of collision risk assessment threshold and danger degree assessment threshold, and the evaluation level division consists of two parts: collision risk level division and safety level division for predicting the effectiveness of avoidance schemes. By correlating quantified rule clauses with avoidance behavior, the risk of collision is divided into four levels, and the safety level for predicting the effectiveness of avoidance schemes is divided into four levels.

[30] Guo, J.; Li L.N.; Gao J.J.; Li G.D. Early warning of collision and avoidance assistance for river ferry. Navigation of China 2020, 43(3): 9-14.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Overall impression & major drawbacks:

Despite the paper deals with ship domain on restricted water areas including narrow waterways which is relatively novel and very interesting topic it has serious drawbacks in the following fields:

1.      State of the art in field of ships domains is not studied deeply enough. Authors did not mention several older and recent publications. Especially there was some papers deals with domain on narrow waterway treating it in 1 dimension only. It was the Chinese scientist idea as far as I remember. Several applications and publications were made in this field especially the one deals with waterway capacity optimization (Chinese, Vietnamese, Turkish, etc.).

2.      Conclusions are not acceptable and discussion part is missing. The reason of this comes partly from point 1 which is lacking the in-depth study of state of the art in the field of maritime risk and ship domains. Authors need to discuss the shape and usefulness of created domain.

3.      Narrow waterway is special area according to its vertical restriction. First authors need to start discussion if the 2D domain exist there. In my opinion not in full extend. The ships have limited space to make decisions and Y-axis of domains results from hydrodynamic interaction between the ships, not the free will of navigators of keeping this area clear of other vessels or anticollision COLREG behavior. Sure, the navigators try the best to keep ships away from them but influence of restrictions force them to come very close to each other. This results in the impossibility of using the domain concept in restricted waters, especially as in the case of the authors to use it as a decision-making tool in anti-collision aspects.

4.      Authors presented the application of proposed method as a tool for detection of risky situation. However, the situation probably does not reflect the data on which the domain was created. This is more like crossing situation and domain was created mostly on overtaking, following, and passing situations. Therefore, in my opinion this domain cannot be use for anticollision early warning.

 

Major corrections:

The creation of domain shall be explained more deeply. Please write mathematical formulas for normalized density in figures 10 and 11.

Maritime English of this paper needs to extensive correction for example Authors use: “ship drive”, “Electronic channel chart”

Line 39-40: “However, ship collisions still exist and have not decreased with the advances in navigation technology.[3]” Couldn’t find this kind of comments/extract in the article of [3]. May be there any error in the citation numbers? If so, could you please specify that part in that article as a response?

3.1. Preprocessing of AIS data: This section should be more detailed. This section directly effects the dimensions of the ship domain.

3.1. Preprocessing of AIS data: Are there any ship type classification during this process? Did you used AIS data of all types of ships in the area?

3.1. Preprocessing of AIS data: In line 147, it is mentioned as “certain time”. AIS data came from the vessel on different times. So, own ship and target ship positions should calculate in the exact time. Did you do any interpolation during the calculation of vessel distribution? If yes, how many seconds or minutes interval did you used to create the new position?

Line 149: “in order to ensure that the own ship’s position is the coordinate origin” but in Figure 3(b), own ship position is not in the center of the coordinate origin. Some explanations in the Line 175 but not clear.

Line 170: “shape and size of the own ship domain can be obtained by analyzing density map data.” Is the shape and size obtained from density map data? How did you used the equation (1) Please describe how you did that?

Line 175: “The center of the ellipse in the ship field is located to the left rear of the center of the ship.” Why did you deviate the ellipse from the center? According to the Figure 3a, vessels are close on the starboard side rather than port side. Does this not show you the vessels tend more distance on port side?

Line 220-223: Is there any explanation to use +-20 degree to calculate overtaking or head situation? For example, why you did not consider the +-30 or 40 data of the relative bearing when calculating the ship domain?

Figure 5: Which vessel did you used to create this scatter distribution map? For example, if you used all vessels in the traffic, some vessels would overtake the other vessels from port side and scatter distribution for those vessels will be more on starboard side. But at the same time, other vessels will be overtaken from port side and scatter distribution for those vessels will be more on port side. When we look at all data, we must see more equal distribution on both sides.

Line 232-235: Why you mirrored the data in Figure 6? This may lead false inference than the actual situation.

Figure 7: Why there are some data on the starboard side in the opposite direction scatter distribution map? Is it because some vessels are not following the inland rules? Or because of the error in the AIS data?

Line 245-246: Why you mirrored the data in Figure 8? This may lead false inference than the actual situation.

Line 274-291: It is not clear why you choose %40, %15 etc. I.e., how did you decide that %40 is appropriate threshold value?

Line 444-446: “Overall, our method's collision risk alarm judgment is consistent with the navigator's judgment, so our method is effective in identifying the collision risk during navigation and in line with the navigator's cognitive habits.” Can we say this with one example of one navigator? How did you define “TAlarm is 60s.”?

 

Minor corrections:

Dots should be after that citation brackets. Ie. “[1].”

AIS data coverage of the study area can be shown in the article to understand the waterway characteristics.

Line 24-25: For the sentence “correct collision avoidance decisions can be made with the Regulations for Preventing Collisions in Inland Rivers of the People's Republic of China”, what kind of differences in these regulations rather than universal standard like COLREG?

Figure 1: There is only head-on and overtaking situation in the flowchart. Crossing situation?

Line 138: Why did you used 0.5m/s to determine the vessels on stopped position. For example, vessel with 2 knots speed also might be on drifting or doing anchoring maneuver.

3.2. Scatter distribution map of the target ships: Which data did you used to calculate orientation, heading etc.

Line 202: “Crossing” is better than “crosswalk”.

Line 240: “60” will probably be “160”.

Line 256: “the front is long and the back long front and short back” these words can be better explained.

Line 299: “Equation B” or “Equation 2”?

Line 338, 345: There is no Equation L or Equation B

Comments for author File: Comments.pdf

Comments on the Quality of English Language

Maritime professional language SHALL be proofread. Ship is not driving.

Author Response

Dear Reviewer:

Thank you very much for your comments concerning our manuscript entitled “Risk Identification Method for Ship Navigation in the Complex Waterways via Consideration of Ship Domain”. Those comments are all valuable and very helpful for revising and improving our paper, and have an important guiding significance to our researches as well. We have checked the manuscript and revised it according to the comments. The main corrections in the paper and the replies to the comments are as follows.

 

Despite the paper deals with ship domain on restricted water areas including narrow waterways which is relatively novel and very interesting topic it has serious drawbacks in the following fields:

1. State of the art in field of ships domains is not studied deeply enough. Authors did not mention several older and recent publications. Especially there was some papers deals with domain on narrow waterway treating it in 1 dimension only. It was the Chinese scientist idea as far as I remember. Several applications and publications were made in this field especially the one deals with waterway capacity optimization (Chinese, Vietnamese, Turkish, etc.).

Reply: We agree with it. We have supplemented the research content in the Section “Introduction”.

2. Conclusions are not acceptable and discussion part is missing. The reason of this comes partly from point 1 which is lacking the in-depth study of state of the art in the field of maritime risk and ship domains. Authors need to discuss the shape and usefulness of created domain.

Reply: We agree with it. We have conducted a more in-depth discussion on the manuscript and compared it with traditional methods (DCPA and TCPA).

3. Narrow waterway is special area according to its vertical restriction. First authors need to start discussion if the 2D domain exist there. In my opinion not in full extend. The ships have limited space to make decisions and Y-axis of domains results from hydrodynamic interaction between the ships, not the free will of navigators of keeping this area clear of other vessels or anticollision COLREG behavior. Sure, the navigators try the best to keep ships away from them but influence of restrictions force them to come very close to each other. This results in the impossibility of using the domain concept in restricted waters, especially as in the case of the authors to use it as a decision-making tool in anti-collision aspects.

Reply：We agree with the importance of hydrodynamic interaction. Based on the experience of a navigator, it is controllable to navigate 200 meters forward and backward, and 50 meters left and right. The main concept of using the ship domain is to achieve early warning. When the target ship invades the own ship domain, it is a very dangerous situation due to hydrodynamic interaction. Therefore, we have made a risk classification and designated it as the most collision avoidance emergency situation. This method is essentially to prevent the situation of invasion in the field of ships in advance.

4. Authors presented the application of proposed method as a tool for detection of risky situation. However, the situation probably does not reflect the data on which the domain was created. This is more like crossing situation and domain was created mostly on overtaking, following, and passing situations. Therefore, in my opinion this domain cannot be use for anticollision early warning.

Reply：We agree that our proposed method is not applicable to all situations. But our method is to navigate the ship in the Wuhan reach of the Yangtze River, in which ships mainly encounter two situations: head-on and overtaking. Our method can be used for anticollision early warning in this situation. We have cited the Regulations for Preventing Collisions in Inland Rivers of the People's Republic of China in the references and provided a PDF file of the rules in the attachment.

Major corrections:

The creation of domain shall be explained more deeply. Please write mathematical formulas for normalized density in figures 10 and 11.

Reply：We agree with it. And we have added mathematical formulas for normalized density.

“                                              (2)

Where, ρ(x, y) is the ship density at the left point of (x, y), Select a unit distance of 5m and calculate the density value of each coordinate point by dividing the number of ship points within the unit distance by the area.”

 

Maritime English of this paper needs to extensive correction for example Authors use: “ship drive”, “Electronic channel chart”

Reply：We agree with it. And we have revised it in manuscript.

Line 39-40: “However, ship collisions still exist and have not decreased with the advances in navigation technology.[3]” Couldn’t find this kind of comments/extract in the article of [3]. May be there any error in the citation numbers? If so, could you please specify that part in that article as a response?

Reply：We agree with it. And we have identified the reference [3].

“Abebe, M.:, Noh, Y.; Seo, C.; Kim, D.; Lee, I. Developing a ship collision risk index estimation model based on Demp-ster-Shafer theory. Applied Ocean Research 2021, 113, 102735.”

3.1. Preprocessing of AIS data: This section should be more detailed. This section directly effects the dimensions of the ship domain.

Reply：We agree with it. We have added the process steps for AIS preprocessing. Due to the fact that AIS preprocessing is a mature technology, which is not the focus of this article's research, there has been no extensive description.

3.1. Preprocessing of AIS data: Are there any ship type classification during this process? Did you used AIS data of all types of ships in the area?

Reply：During the AIS data preprocessing process, we did not consider the issue of ship classification. We chose ships within 1 kilometer of the own ship with a speed greater than 0.5 knots as the target ship for our research.

“Preprocessing of AIS data mainly includes data cleaning and data repairing. Data repairing mainly involves linear interpolation of the data, with an interpolation interval of 2 seconds.”

3.1. Preprocessing of AIS data: In line 147, it is mentioned as “certain time”. AIS data came from the vessel on different times. So, own ship and target ship positions should calculate in the exact time. Did you do any interpolation during the calculation of vessel distribution? If yes, how many seconds or minutes interval did you used to create the new position?

Reply：We agree with it. We have preprocessed AIS data interpolation with a 2 second interval.

Line 149: “in order to ensure that the own ship’s position is the coordinate origin” but in Figure 3(b), own ship position is not in the center of the coordinate origin. Some explanations in the Line 175 but not clear.

Reply：We agree with it. This is a human annotation error, and we have made modifications to the Figure 3(b).

Line 170: “shape and size of the own ship domain can be obtained by analyzing density map data.” Is the shape and size obtained from density map data? How did you used the equation (1) Please describe how you did that?

Reply：By using the thresholds in Figures 10 and 11, we can obtain the sizes of the major and minor axes of the ellipse. By substituting them into equation (1), we can obtain the size of the ship domain.

Line 175: “The center of the ellipse in the ship field is located to the left rear of the center of the ship.” Why did you deviate the ellipse from the center? According to the Figure 3a, vessels are close on the starboard side rather than port side. Does this not show you the vessels tend more distance on port side?

Reply：Figure 3a is a distribution map of all real data plotted together, without distinguishing between data in the same direction and in the opposite direction, resulting in vessels close on the starboard side rather than port side.

Line 220-223: Is there any explanation to use +-20 degree to calculate overtaking or head situation? For example, why you did not consider the +-30 or 40 data of the relative bearing when calculating the ship domain?

Reply：Due to the narrow channel of the Yangtze River, the relative heading value of the target ship during overtaking situation is close to 0. Figure 4 shows the relative heading distribution of surrounding ships based on the own ship's heading, in which 77.5% of the ships have relative heading between -180 degrees to -160 degrees, -20 degrees to 20 degrees, and 160 degrees to 180 degrees. Therefore, the relative heading of the ship is selected with data from -20 degrees to 20 degrees as the same direction data, and data from -180 degrees to -160 degrees and 160 degrees to 180 degrees as the opposite direction data for distribution density statistical analysis of ship. The amount of data for +-30 or +-40 data of the relative bearing is relatively small, and due to the presence of ship turning, the data in this interval does not provide clear heading division. So it does not affect our research results.

.

Figure 4. Distribution of relative heading of ships.

Figure 5: Which vessel did you used to create this scatter distribution map? For example, if you used all vessels in the traffic, some vessels would overtake the other vessels from port side and scatter distribution for those vessels will be more on starboard side. But at the same time, other vessels will be overtaken from port side and scatter distribution for those vessels will be more on port side. When we look at all data, we must see more equal distribution on both sides.

Reply：We agree with it. We used the vessel Navigating in the Wuhan reach of the Yangtze River, which comply with the Regulations for Preventing Collisions in Inland Rivers of the People's Republic of China. According to this regulation, ships should overtake from near the middle line of the waterway. And select AIS data from the same direction for analysis, and the own ship has a faster speed and more behavior of overtaking the target ship. Therefore, it is shown in Figure 5.

Line 232-235: Why you mirrored the data in Figure 6? This may lead false inference than the actual situation.

Reply：According to the Regulations for Preventing Collisions in Inland Rivers of the People's Republic of China, when the flow velocity changes on the left and right sides of the channel, there may be changes in the up and down channels on the left and right sides. Ships should overtake from near the middle line of the waterway Therefore, it is necessary to form equivalent discrimination criteria after mirroring to cope with the changes in the up and down channels on the left and right sides.

Figure 7: Why there are some data on the starboard side in the opposite direction scatter distribution map? Is it because some vessels are not following the inland rules? Or because of the error in the AIS data?

Reply：This is because, according to the Regulations for Preventing Collisions in Inland Rivers of the People's Republic of China, when the flow velocity changes on the left and right sides of the channel, there may be changes in the up and down channels on the left and right sides.

Line 245-246: Why you mirrored the data in Figure 8? This may lead false inference than the actual situation.

Reply：According to the Regulations for Preventing Collisions in Inland Rivers of the People's Republic of China, when the flow velocity changes on the left and right sides of the channel, there may be changes in the up and down channels on the left and right sides. Ships should overtake from near the middle line of the waterway Therefore, it is necessary to form equivalent discrimination criteria after mirroring to cope with the changes in the up and down channels on the left and right sides.

Line 274-291: It is not clear why you choose %40, %15 etc. I.e., how did you decide that %40 is appropriate threshold value?

Reply：According to research conducted with experienced navigators, the safe distance between the own ship and the target ship ahead in the same direction is about 200m in the Yangtze River. In Figure 10 (a), the Normalized ship distribution density corresponding to a distance of 200 from the target ship ahead is about 40%. Therefore, we chose 40% in Figure 10 and Figure 11. But in Figure 10 (b), choosing 40% is clearly inappropriate, so a moderate 15% was chosen. The main purpose here is to determine the eccentric position of the ellipse. In order to ensure the availability of the final ship field, adjustments will be made in the subsequent formulas(2) through k_a and k_b.

.

(a) the same direction                        (b) the opposite direction.

Figure 10. Density curve of the ship under x₀=0 section for ship navigation.

Line 444-446: “Overall, our method's collision risk alarm judgment is consistent with the navigator's judgment, so our method is effective in identifying the collision risk during navigation and in line with the navigator's cognitive habits.” Can we say this with one example of one navigator? How did you define “TAlarm is 60s.”?

Reply：The yellow line in Figure 18 shows the steering behavior record of the navigator, which is consistent with the navigator's cognition through comparison with our method. The TAlarm can be adjusted based on the actual navigation and traffic conditions in different water areas. We determined define “TAlarm is 60s.” by observing 10 scenarios of steering and based on the needs of experienced navigators of the Yangtze River.

Minor corrections:

Dots should be after that citation brackets. Ie. “[1].”

Reply：We have made modifications in the manuscript.

 

AIS data coverage of the study area can be shown in the article to understand the waterway characteristics.

Reply：We have replaced Figure 14 in the manuscript, which shows the AIS data coverage of the study area.

Figure 14. Electronic chart with AIS data of the Luoyang reach

Line 24-25: For the sentence “correct collision avoidance decisions can be made with the Regulations for Preventing Collisions in Inland Rivers of the People's Republic of China”, what kind of differences in these regulations rather than universal standard like COLREG?

Reply：We have cited the Regulations for Preventing Collisions in Inland Rivers of the People's Republic of China in the references and provided a PDF file of the rules in the attachment. Compared to COLREG, this rule is specific to the complex waterways of the Yangtze River in China.

Figure 1: There is only head-on and overtaking situation in the flowchart. Crossing situation?

Reply：Due to the limited number of Crossing situation in the complex waterways of the Yangtze River, according to the "Regulations for Preventing Collisions in Domestic Rivers of the People's Republic of China", ships mainly encounter two situations: head-on and overtaking. Therefore, there are only two situations in the flowchart.

Line 138: Why did you used 0.5m/s to determine the vessels on stopped position. For example, vessel with 2 knots speed also might be on drifting or doing anchoring maneuver.

Reply: Our manuscript selects a smaller speed to determine whether the ship is moving. On the one hand, there is a dedicated berthing area for ships in the complex waterway of the Yangtze River. On the other hand, this article studies collision avoidance during ship navigation, which has a large amount of AIS data and reduces the amount of invalid data. At the same time, the target ship moving at a speed of 2 knots, even in stopped position, needs to avoid collision. Therefore, the selection of 0.5m/s to determine the vessels on stopped position does not affect the research results.

3.2. Scatter distribution map of the target ships: Which data did you used to calculate orientation, heading etc.

Reply: The relative orientation is calculated from the  longitude, latitude, heading of the own ship, and the longitude and latitude of the target ship.

Line 202: “Crossing” is better than “crosswalk”.

Reply: We agree with it. And we have revised it in manuscript.

Line 240: “60” will probably be “160”.

Reply: We agree with it. And we have revised it in manuscript.

Line 256: “the front is long and the back long front and short back” these words can be better explained.

Reply: We agree with it. And we have reinterpreted this sentence.

Line 299: “Equation B” or “Equation 2”?

Reply: We intended to define formulas (3) as Equation B. The improper expression in the manuscript has caused ambiguity, and we have made revisions

Line 338, 345: There is no Equation L or Equation B

Reply: We intended to define formulas (3) as Equation B and define formulas (11) as Equation B. The improper expression in the manuscript has caused ambiguity, and we have made revisions.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The paper is related to risk identification method for ship navigation. The topic is interesting to the academic community. My comments:

1.       The authors should conduct a comprehensive comparison with existing methods and emphasize how their approach extends current techniques. It's a good idea to expand your literature review, for example:

-   Pietrzykowski Z., Wołejsza P., Nozdrzykowski Ł. et al. „The autonomous navigation system of a sea-going vessel” Ocean Engineering vol. 261, art. no. 112104, 2022.

-   Lokukaluge P. Perera, C. Guedes Soares, Collision risk detection and quantification in ship navigation with integrated bridge systems, Ocean Engineering, Volume 109, 2015 (344-354)

-   Ulku Ozturk, Kadir Cicek, Individual collision risk assessment in ship navigation: A systematic literature review, Ocean Engineering, Volume 180, 2019 (130-143)

2.       Why were such values of ka kb adopted? Please explain.

3.       How were external disturbances taken into account in the analysis? Please refer to literature, for example, as:

-   G. Budak, S. Beji, Controlled course-keeping simulations of a ship under external disturbances, Ocean Engineering, Volume 218, 2020, 108126.

-   Borkowski, P. Numerical Modeling of Wave Disturbances in the Process of Ship Movement Control. Algorithms 2018, 11, 130

4.       What is the reason why the authors have not performed the same study in different scenarios? The number of experimental cases should be increased. I would suggest the authors demonstrate the effectiveness of the research method by adding more instances of ship encounters and comparing and analyzing different situations.

5.       The authors are suggested to have Discussion section to investigate the weakness, strength, and potential enhancement of proposed scheme.

Author Response

Dear Reviewer:

Thank you very much for your comments concerning our manuscript entitled “Risk Identification Method for Ship Navigation in the Complex Waterways via Consideration of Ship Domain”. Those comments are all valuable and very helpful for revising and improving our paper, and have an important guiding significance to our researches as well. We have checked the manuscript and revised it according to the comments. The main corrections in the paper and the replies to the comments are as follows.

The paper is related to risk identification method for ship navigation. The topic is interesting to the academic community. My comments:

1. The authors should conduct a comprehensive comparison with existing methods and emphasize how their approach extends current techniques. It's a good idea to expand your literature review, for example:

-Pietrzykowski Z., Wołejsza P., Nozdrzykowski Ł. et al. „The autonomous navigation system of a sea-going vessel” Ocean Engineering vol. 261, art. no. 112104, 2022.

-Lokukaluge P. Perera, C. Guedes Soares, Collision risk detection and quantification in ship navigation with integrated bridge systems, Ocean Engineering, Volume 109, 2015 (344-354)

-Ulku Ozturk, Kadir Cicek, Individual collision risk assessment in ship navigation: A systematic literature review, Ocean Engineering, Volume 180, 2019 (130-143)

Reply：We agree with it. And we have conducted a comparison with existing methods by expanding our literature review in Section 1.

2. Why were such values of ka kb adopted? Please explain.

Reply：In order to ensure the availability of the final ship field, adjustments will be made in the subsequent formulas(3) through k_a and k_b. The front, back, left, and right of the ship domain are different, so the magnification is different. The values of a and b should be determined based on the actual situation.

.

(a) the same direction                        (b) the opposite direction.

Figure 10. Density curve of the ship under x₀=0 section for ship navigation.

 

3. How were external disturbances taken into account in the analysis? Please refer to literature, for example, as:

-G. Budak, S. Beji, Controlled course-keeping simulations of a ship under external disturbances, Ocean Engineering, Volume 218, 2020, 108126.

-Borkowski, P. Numerical Modeling of Wave Disturbances in the Process of Ship Movement Control. Algorithms 2018, 11, 130

Reply：We agree that analyzing external interference is important. Due to the lack of actual flow field data, poor ship informatization conditions, and the lack of equipment such as recorders and meteorological instruments, our research has not yet considered external disturbances, which is also a deficiency of current research. We will further carry out relevant research in the future. We have further discussed it in the discussion section of the manuscript.

4. What is the reason why the authors have not performed the same study in different scenarios? The number of experimental cases should be increased. I would suggest the authors demonstrate the effectiveness of the research method by adding more instances of ship encounters and comparing and analyzing different situations.

Reply：We agree with it. We have added a comparison with traditional methods (DCPA and TCPA) to validate our proposed method.

Figure 19. Comparison between our method and TCPA (the blue line is T_ca and The yellow line is the TCPA)

According to traditional methods, DCPA less than 200m and TCPA less than 80s are used as alarm judgment conditions, where DCPA is less than 100 meters throughout the entire process, so only the changes in TCPA are considered[].

Figure 19 shows that our proposed method is more accurate than the traditional DCPA and TCPA judgment methods for early warning. Firstly, the traditional DCPA and TCPA judgment methods lag by 8 seconds before starting to alarm. Secondly, after the navigation risk alarm, the traditional methods always maintain the alarm state for a long time, and only stop the alarm after the ship has passed and cleared, which is inconsistent with the driver's judgment of navigation risk and cannot accurately reflect the navigation situation.

5. The authors are suggested to have Discussion section to investigate the weakness, strength, and potential enhancement of proposed scheme.

Reply： We agree with it. And we have added discussion.

“In this paper, a new method for identifying ship navigation risks by combining the ship domain with AIS data is proposed, in which the collision time with the target ship is calculated based on the collision hazard detection line and safety distance boundary, and formed a method for dividing the danger level of the ship navigation situation.

The research focuses on the Wuhan reach of the Yangtze River, where ships navigate in accordance with the Regulations for Preventing Collisions in Inland Rivers of the People's Republic of China. In the complex waterway of the Yangtze River, the simulation results of our method are consistent with the actual navigation behavior and superior to the traditional TCPA method. This also indicates that our method can accurately reflect the consolidation risk of the target ship relative to the own ship, and the warning data is stable and has good real-time performance, which can effectively assist the navigator in identifying consolidation risks during navigation.

However, our method did not consider the hydrodynamic interaction between the ships. The external hydrodynamic interference, such as waves and currents, is one of the important influencing factors for ship navigation decisions. Our method has limitations by determining a reasonable size of the ship domain and avoiding the impact of hydrodynamic interference through navigators. Due to the lack of actual flow field data, poor ship informatization conditions, and the lack of equipment such as recorders and meteorological instruments, our research cannot consider external disturbances. In addition, hydrodynamic simulation requires a certain amount of computational power support, which may improve the accuracy of early warning but reduce the speed of early warning.

In the future, we will conduct research on rapid simulation and early warning methods that consider hydrodynamic interaction based on our proposed method.”

 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

1. Why do you set AIS interval 2second?

2. Please check reference related to navigator’s safe distance 200m

 

Requested comments have been well incorporated.

Author Response

Dear Reviewer:

Thank you very much for your comments concerning our manuscript entitled “Risk Identification Method for Ship Navigation in the Complex Waterways via Consideration of Ship Domain”. Those comments are all valuable and very helpful for revising and improving our paper, and have an important guiding significance to our researches as well. We have checked the manuscript and revised it according to the comments. The main corrections in the paper and the replies to the comments are as follows.

1. Why do you set AIS interval 2second?

Reply：Due to the fact that we perceive ship location information through radar and AIS devices, AIS data is interpolated and fused with radar data, and the interval time needs to be consistent. The radar is stable and scans 30 times per minute, which means it takes 2 seconds to scan once. Therefore, by setting AIS interval of 2 seconds, it can be more convenient to fuse with radar perception data.

2. Please check reference related to navigator’s safe distance 200m

Reply：The "Channel 1" researched is a command ship for channel work, with a total length of 49.60 meters. According to Wen's research, the ratio of ship length to ship domain in typical inland waterway is between 2.5 and 3.5, which means the ship domain is between 124m and 173.6m. The longer the ship, the relatively smaller the ship domain, and the minimum ship length studied by Wen is greater than 60m. Therefore, the ship domain of our own ship with a length of 49.60m should be greater than 173.6m. In addition, based on the investigation of experienced navigators, a reasonable safe distance of 200m was obtained, which is consistent with Wen's research findings.

19. Wen, Y.Q.; Li, T.; Zheng, H.T.; Huang, L.; Zhou, C.H.; Xiao, C.S. Characteristics of ship domain in typical inland waters. NAVIGATION OF CHINA 2018, 41(1), 43-47.

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for updating the manuscript. I hope review improved your paper

Comments on the Quality of English Language

No comments

Author Response

Dear Reviewer:

      Thank you very much for your comments concerning our manuscript entitled “Risk Identification Method for Ship Navigation in the Complex Waterways via Consideration of Ship Domain” and agree to acceptance of our manuscript. And our manuscript has undergone further grammar polishing.

Reviewer 3 Report

Comments and Suggestions for Authors

The authors addressed my previous comments. I am satisfied with this revision.

Author Response

Dear Reviewer:

       Thank you very much for your comments concerning our manuscript entitled “Risk Identification Method for Ship Navigation in the Complex Waterways via Consideration of Ship Domain” and agree to acceptance of our manuscript.