An arctic vessel risk analysis and decision system

Abstract

The application provides a polar region ship risk analysis and decision system, and belongs to the technical field of ship safety. The system comprises a user interaction layer, a data processing layer and a data storage layer, and further comprises a current situation risk assessment module, a loss calculation module and a system data updating module. The user interaction layer provides an interface for the three modules to interact with the crew. The data processing layer supports the construction and analysis calculation of the Bayesian network model of the current situation risk assessment module, and provides data processing capability for the calculation of various expenses in the loss calculation module. Meanwhile, the data processing layer realizes the transformation and storage of data in the system data updating module. The data storage layer provides data storage and reading services for the three modules, so as to ensure that the modules can obtain the required data. The application can evaluate the ice distress and ice damage risks that the ship may encounter, assist in decision-making and loss calculation, provide effective support and sailing suggestions for the crew, and provide support for the actual polar region ship risk management.

Description

Technical Field

[0001] This invention belongs to the field of ship safety technology, specifically, it relates to a risk analysis and decision-making system for ships in polar regions. Background Technology

[0002] As global temperatures continue to rise and polar glaciers continue to melt, the commercial, routine, and project-based operation of ships along the Arctic shipping route will become a reality. However, the harsh natural environment, unique geographical location, and complex navigation conditions of the Arctic pose significant threats and challenges to ship safety. Ice entrapment and ice damage are considered the most severe risks, potentially leading to ship damage, property loss, environmental damage, and even casualties. Current research on ship risk assessment in Arctic ice zones is mostly theoretical, providing guidance for practical operations but not directly applicable to actual ship navigation. Therefore, it is necessary to design and implement a ship risk assessment and decision-making system for Arctic ice zones. This system utilizes the Python programming language to compile a Bayesian network model, creating a user-interactive graphical interface to perform risk assessment, decision support, and loss calculation. While assessing ship risks under current conditions, it provides effective support and suggestions for crew operations and navigation decisions. Summary of the Invention

[0003] This invention addresses the two risk factors faced by polar vessels: ice trapping and ice damage. Using the Python language, a risk analysis and decision-making system for polar vessels is designed and established.

[0004] This invention is achieved through the following technical solution: a polar region ship risk analysis and decision-making system:

[0005] The system includes a user interaction layer, a data processing layer, and a data storage layer;

[0006] The user interaction layer is used to select the current risk node status, display reasoning results and auxiliary decision-making suggestions, and display economic loss results.

[0007] The data processing layer performs parameter learning and data inference through a Bayesian network model.

[0008] The data storage layer is used to store accident samples, recommended measures, and parameters.

[0009] Furthermore, the human-computer interface content displayed in the user interaction layer includes current situation risk assessment, loss calculation, system data updates, and system exit.

[0010] Furthermore, the parameter transmission path of the data processing layer is as follows: the human-machine interface converts ice zone and ship accident data into the computer engine, and through Python software, the input computer data is converted into Python type. Then, the pgmpy library is used to build a Bayesian network model and perform analysis and calculation.

[0011] The Bayesian network returns the inference data to the Python engine, which then transforms the data into the original data type of the computer engine and transmits it back to the computer engine. The results are then displayed on the human-computer interface, enabling the provision of risk analysis and decision support functions.

[0012] Furthermore, the system includes a current situation risk assessment module, a loss calculation module, and a system data update module;

[0013] The user interaction layer provides an interface for the three modules to interact with the crew, through which the crew can operate the various modules;

[0014] The data processing layer provides support for the construction, analysis and calculation of the Bayesian network model in the current situation risk assessment module, and also provides data processing capabilities for the calculation of various costs in the loss calculation module. At the same time, it realizes the transformation and storage of data in the system data update module.

[0015] The data storage layer provides data storage and retrieval services for the three modules, ensuring that the modules can obtain the data they need.

[0016] Furthermore, the current situation risk assessment module displays content including risk factor selection, Bayesian network inference results, and suggested measures, and has risk assessment and decision support functions;

[0017] The loss calculation module performs a brief calculation of the economic losses of a ship after it is trapped or damaged by ice. The parameters include whether the ship has docked at port after the accident, whether it has returned empty, the segment where the accident occurred, the remaining mileage of that segment, and the thrust loss after the accident.

[0018] Furthermore, the accident loss calculation considers maintenance costs F1, which includes annual maintenance and single maintenance costs for propellers, increased fuel costs due to reduced thrust caused by propeller structural damage F2, transshipment port costs F3, and potential losses due to an empty return to the maintenance port F4, as shown in the following formula:

[0019] F l =F1+F2+F3+F4

[0020] The calculation of maintenance cost F1 is based on the annual maintenance rate of 1.606% for icebreakers;

[0021] The additional fuel cost F2 depends on the fuel consumption rate (f), fuel price (bp), and single-voyage travel time (time). Fuel cost is defined as:

[0022] F2 = f × time × b p

[0023] Assuming a fuel price of 500 USD / ton, the fuel consumption rate (f) can be calculated using the propeller law:

[0024] f = SFOC × BHP

[0025]

[0026] Wherein, SFOC represents a specific fuel consumption rate (unit: g / kW·h), with a fixed value of 185 g / kW·h; BHP is the engine braking horsepower (unit: kW); m is the proportional factor; Vi is the speed of the container ship traveling on different sections of the voyage; BHPmax is the maximum output power of the main engine; and Vmax is the maximum ship speed.

[0027] Transit port charges F3 mainly include (a) port fees, (b) container handling fees, (c) port pilotage fees, and (d) icebreaking fees for specific ports; these charges are generally calculated based on the vessel's net tonnage or container capacity and relevant rate standards, as shown in the following formula:

[0028]

[0029] (b) Container handling fee = Handling fee rate (unit: USD / TEU) × Vessel's container capacity (unit: TEU)

[0030] (c) Port pilotage fee = pilotage fee rate (unit: USD / ton) × net ton;

[0031] (d) Icebreaking fee for a specific port = Icebreaking fee rate (unit: USD / ton) × Gross tonnage

[0032] The icebreaking rates were obtained by the Russia Northern Sea Route Administration.

[0033] The loss F4 due to empty return to the port of repair is calculated based on the freight rate index, assuming a container freight rate of 900 USD / TEU.

[0034] Furthermore, the system data update module is divided into accident data update and accident loss parameter update sections. System users, based on newly reported ice trapping and ice damage accidents, discretize the data according to the identified risk factors and input it into the corresponding boxes in the accident data update section. Then, the software accident database is updated and updated. To address the issue that the losses after ice trapping and ice damage accidents vary for different ship types, the accident loss parameter update module modifies the corresponding parameters according to the specific type of the ship used by the system user, in order to obtain a more accurate estimate of economic losses.

[0035] Beneficial effects of the invention

[0036] This invention designs and develops a risk assessment and decision-making system that can be practically applied to vessels navigating in polar ice regions. It is simple, convenient, and fast to operate, capable of assessing, assisting in decision-making, and calculating losses for potential ice entrapment and ice damage risks encountered by vessels. It provides effective support and navigation advice to crew members, and supports practical risk management for vessels in polar regions. This invention, to a certain extent, broadens the practical application scope of risk assessment for vessels in Arctic ice regions. Attached Figure Description

[0037] Figure 1 This is a design diagram of the risk assessment and decision-making system of the present invention.

[0038] Figure 2 This is the interface for the risk analysis and decision-making system of the present invention.

[0039] Figure 3 This is a schematic diagram of the parameter passing path in the processing layer of the present invention.

[0040] Figure 4 This is the pre-context risk assessment module of the present invention.

[0041] Figure 5 This is the loss calculation module of the present invention.

[0042] Figure 6 This is the system data update module of the present invention.

[0043] Figure 7 This is the accident data update module of the present invention.

[0044] Figure 8 This is the loss parameter update module of the present invention.

[0045] Figure 9 This is a diagram showing the timing of the "oversight negligence" and "sea ice thickness" nodes in this invention.

[0046] Figure 10 This provides relevant recommendations for when the "poor ice conditions" node of this invention occurs.

[0047] Figure 11The sample calculation for this invention is shown in (a) for loss setting and (b) for calculation result. Detailed Implementation

[0048] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0049] Unless otherwise specified, the experimental methods used in the following examples are conventional methods. Unless otherwise specified, the materials, reagents, methods, and instruments used are all conventional materials, reagents, methods, and instruments in the art, and can be obtained commercially by those skilled in the art.

[0050] This invention discloses a polar region ship risk analysis and decision-making system, based on the Python language, which consists of three layers: a user interaction layer, a data processing layer, and a data storage layer. Figure 1 As shown.

[0051] The user interaction layer is used for selecting the current risk node status, displaying reasoning results and auxiliary decision-making suggestions, and displaying economic loss results. User interaction, simply put, refers to communication between humans and computers. The five main interactive behavioral elements widely recognized and valued in the field of interaction are: person, action, purpose, medium, and scenario. Specifically, in this invention, it mainly involves the interaction between the polar vessel risk analysis and decision-making system and the crew. As a polar vessel risk analysis and decision-making system, the interface is as follows... Figure 2 As shown, the human-computer interface of the user interaction layer displays content including current situation risk assessment, loss calculation, system data update, and system exit.

[0052] The interface design uses the Tkinter library, a third-party library in Python. Tkinter is a standard GUI (Graphical User Interface) library in Python that supports cross-platform GUI program development. It is a module in Python specifically designed for window design and features simple construction, high compatibility, and suitability for writing small programs.

[0053] The data processing layer performs parameter learning and data inference through a Bayesian network model. Data processing is a key step in the realization of a system. In the process of risk assessment, the parameter transmission path of the data processing layer is as follows: the human-machine interface converts ice zone and ship accident data into the computer engine, the input computer data is converted into Python type through Python software, and then the pgmpy library is used to build a Bayesian network model and perform analysis and calculation.

[0054] The Bayesian network returns the inference data to the Python engine, which then transforms the data into its original format for the computer engine and transmits it back. The results are then displayed on the user interface, providing risk analysis and decision support. The processing layer works as follows: Figure 3 As shown.

[0055] The data storage layer is used to store accident samples, recommended measures, and parameters.

[0056] The software data storage of this invention uses the Pandas library in Python. Pandas is a third-party library in Python, a tool based on NumPy, primarily used for data analysis tasks. The `read_csv()` and `to_csv()` commands in Pandas can read and save CSV files from a specified path, and are used as the data storage layer in this software.

[0057] Correspondingly, the system of the present invention includes a current situation risk assessment module, a loss calculation module, and a system data update module;

[0058] The user interaction layer provides an interface for the three modules to interact with the crew, through which the crew can operate the various modules;

[0059] The data processing layer provides support for the construction, analysis and calculation of the Bayesian network model in the current situation risk assessment module, and also provides data processing capabilities for the calculation of various costs in the loss calculation module. At the same time, it realizes the transformation and storage of data in the system data update module.

[0060] The data storage layer provides data storage and retrieval services for the three modules, ensuring that the modules can obtain the data they need.

[0061] Current situation risk assessment module, such as Figure 4As shown, the main content includes risk factor selection, Bayesian network inference results, and suggested measures, providing risk assessment and decision support functions. The Bayesian network model is primarily built using the pgmpy library in Python. Pgmpy is an open-source project, a Python library for processing probabilistic graphical models. The Bayesian network model structure is constructed using the BayesianModel function, followed by the use of the TabularCPD function to build the conditional probability distribution (CPD) table. Finally, the CPD parameter data is added to the Bayesian network structure to complete the Bayesian network model.

[0062] The loss calculation module mainly performs a brief calculation of the economic losses after a ship is trapped in ice or damaged by ice, such as... Figure 5 As shown, the main parameters include whether the vessel docked at a port after the accident, whether it returned empty, the segment where the accident occurred, the remaining mileage of that segment, and the thrust loss after the accident. The accident loss calculation of this system considers maintenance costs F1 (including annual maintenance and single maintenance costs for propellers, etc.), increased fuel costs F2 due to reduced thrust caused by propeller structural damage, transit port costs F3, and possible losses due to returning to the maintenance port empty F4, as shown in the following formula.

[0063] F l =F1+F2+F3+F4

[0064] The calculation of maintenance cost F1 is based on the following assumptions: It is assumed that the annual maintenance cost of container ships is 1.095% of the ship's construction cost. According to relevant literature, it is assumed that the annual maintenance cost rate for ice-resistant ships is approximately 10%-25% higher than that of ordinary ships of the same size, and that the construction cost of ice-resistant ships is approximately 25% higher than that of ordinary ships of the same size. Therefore, the annual maintenance cost rate for ice-resistant ships is assumed to be 1.606%.

[0065] The additional fuel cost F2 depends on the fuel consumption rate (f), fuel price (bp), and single-voyage travel time (time). Fuel cost is defined as:

[0066] F2 = f × time × b p

[0067] Assuming a fuel price of 500 USD / ton, according to the propeller law, the fuel consumption rate (f) can be calculated using the following formula:

[0068] f = SFOC × BHP

[0069]

[0070] Wherein, SFOC represents a specific fuel consumption rate (unit: g / kW·h), fixed at 185 g / kW·h; BHP is engine braking horsepower (unit: kW); m is the proportional factor; Vi is the speed of the container ship on different sections of the voyage; BHPmax is the maximum output power of the main engine; Vmax is the maximum ship speed.

[0071] Transit port charges F3 mainly include (a) port fees, (b) container handling fees, (c) port pilotage fees, and (d) icebreaking fees for specific ports. These charges are generally calculated based on the vessel's net tonnage or container capacity and relevant rate standards, as shown in the following formula:

[0072]

[0073] (b) Container handling fee = Handling fee rate (unit: USD / TEU) × Vessel's container capacity (unit: TEU)

[0074] (c) Port pilotage fee = pilotage fee rate (unit: USD / ton) × net ton;

[0075] (d) Icebreaking fee for a specific port = Icebreaking fee rate (unit: USD / ton) × Gross tonnage

[0076] The icebreaking rates are obtained by the Russia Northern Sea Route Administration.

[0077] The loss F4 due to empty return to the repair port is calculated based on the China Container Freight Index released by the Shanghai Shipping Exchange from 2017 to 2019, assuming a container freight rate of 900 USD / TEU.

[0078] System data update module, such as Figure 6 As shown, this module is mainly divided into accident data update and accident loss parameter update sections, such as... Figure 7 and Figure 8 As shown. System users can discretize newly reported ice-trapping and ice-damage incidents according to the 24 risk factors categorized by the system and input them into the corresponding boxes in the incident data update section. Then, clicking the "Incident Data Update" button below will update and add to the software's incident database. To address the varying losses after ice-trapping and ice-damage incidents for different ship types, the incident loss parameter update module can modify the corresponding parameters based on the specific type of vessel used by the system user, in order to obtain a more accurate estimate of economic losses.

[0079] The following example illustrates this: a ship departing from a port on the East China Sea coast and traveling through the Northeast Passage to the port of Rotterdam in the Netherlands.

[0080] Suppose that the crew experiences lookout negligence in a certain area, and the sea ice thickness in that area is known to be 1.1m. In the "Current Risk Assessment" module, the "Lookout Negligence" node can be set to "True," and the sea ice thickness set to 1.1m. Figure 9 As shown, clicking "Update Data" will perform risk analysis and decision support for the current status. After Bayesian network calculation, risk factors with increased probability of occurrence will be displayed in the blank space, while corresponding navigation precautions and suggestions will be displayed in the blank space on the right. Figure 10 As shown, nodes such as "Insufficient Information," "Insufficient Hazard Assessment," "Fatigue," and "Visibility" are displayed sequentially in the blank "Risk Factors" box in the middle of the software due to their increased probability of occurrence. The display order is: the nodes that have already occurred, the two types of risk nodes, and other nodes. Clicking on the node name in the middle risk factor box will display corresponding suggestions on the right side of the software interface to assist the crew in making decisions. The figure shows the relevant suggestions given by the software when the "Poor Ice Conditions" situation occurs.

[0081] The loss calculation module primarily performs a simplified calculation of the economic losses following ice entrapment or ice damage incidents involving ships. It assumes the sample ship has a 100% load rate and a planned return eastward voyage with a 50% load rate. The ship encountered ice damage in the middle of the Vilkitsky Strait, resulting in propeller damage that reduced its speed to 80% of its original speed with the same engine power. After stopping at Murmansk to inspect the damage, the ship continued to Rotterdam to unload cargo and then returned empty to a port on the East China Sea coast for repairs. The system settings are as follows: Figure 11 As shown in Figure a, the output result after clicking to calculate the result is as follows. Figure 11 As shown in b, the system anticipates losses of $5.68 million from the accident.

[0082] Therefore, the risk assessment and decision-making system designed in this invention can be practically applied to ships navigating in polar ice areas. It is simple, convenient and fast to operate, and can assess, assist in decision-making and calculate losses for the risks that ships may encounter such as ice entrapment and ice damage. It provides effective support and navigation advice for the crew and supports the actual risk management of ships in polar areas.

[0083] The above provides a detailed description of the polar region ship risk analysis and decision-making system proposed in this invention, and elucidates the principles and implementation methods of this invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A polar region ship risk analysis and decision-making system, characterized in that: The system includes a user interaction layer, a data processing layer, and a data storage layer; The user interaction layer is used to select the current risk node status, display reasoning results and auxiliary decision-making suggestions, and display economic loss results. The data processing layer performs parameter learning and data inference through a Bayesian network model. The specific parameter transmission path of the data processing layer is as follows: the human-machine interface converts ice zone and ship accident data into the computer engine, and through Python software, the input computer data is converted into Python type. Then, the pgmpy library is used to build a Bayesian network model and perform analysis and calculation. The Bayesian network returns the inference data to the Python engine, which then transforms the data into the original data type of the computer engine and transmits it back to the computer engine. The results are then displayed on the human-computer interface, enabling the provision of risk analysis and decision support functions. The data storage layer is used to store accident samples, recommended measures, and parameters; The system includes a current situation risk assessment module, a loss calculation module, and a system data update module; The user interaction layer provides an interface for the three modules to interact with the crew, through which the crew can operate the various modules; The data processing layer provides support for the construction, analysis and calculation of the Bayesian network model in the current situation risk assessment module, and also provides data processing capabilities for the calculation of various costs in the loss calculation module. At the same time, it realizes the transformation and storage of data in the system data update module. The data storage layer provides data storage and retrieval services for the three modules, ensuring that the modules can obtain the data they need; The current situation risk assessment module displays content including risk factor selection, Bayesian network inference results, and suggested measures, and has risk assessment and decision support functions; The loss calculation module performs a brief calculation of the economic losses of a ship after it is trapped or damaged by ice. The parameters include whether the ship has docked at port after the accident, whether it has returned empty, the segment where the accident occurred, the remaining mileage of that segment, and the thrust loss after the accident.

2. The system according to claim 1, characterized in that: The human-computer interface of the user interaction layer displays content including current situation risk assessment, loss calculation, system data update, and system exit.

3. The system according to claim 2, characterized in that: The accident loss calculation considers maintenance cost F1, which includes annual maintenance and single maintenance costs for propellers, increased fuel costs due to reduced thrust caused by propeller structural damage F2, transshipment port costs F3, and potential losses due to empty return to the maintenance port F4, as shown in the following formula: The calculation of maintenance cost F1 is based on the annual maintenance rate of 1.606% for icebreakers; The additional fuel cost F2 depends on the fuel consumption rate (f), fuel price (bp), and single-voyage travel time (time). Fuel cost is defined as: F2= Assuming a fuel price of 500 USD / ton, the fuel consumption rate (f) can be calculated using the propeller law: Wherein, SFOC represents a specific fuel consumption rate (unit: g / kW·h), with a fixed value of 185 g / kW·h; BHP is the engine braking horsepower (unit: kW); m is the proportional factor; Vi is the speed of the container ship traveling on different sections of the voyage; BHPmax is the maximum output power of the main engine; and Vmax is the maximum ship speed. Transit port charges F3 mainly include (a) port fees, (b) container handling fees, (c) port pilotage fees, and (d) icebreaking fees for specific ports; these charges are generally calculated based on the vessel's net tonnage or container capacity and relevant rate standards, as shown in the following formula: The icebreaking rates were obtained by the Russia Northern Sea Route Administration. The loss F4 due to empty return to the port of repair is calculated based on the freight rate index, assuming a container freight rate of 900 USD / TEU.

4. The system according to claim 3, characterized in that: The system data update module is divided into accident data update and accident loss parameter update. System users, based on newly reported ice-related accidents and ice damage, discretize the risk factors and input them into the corresponding boxes in the accident data update section. Then, the software accident database is updated and added. To address the varying losses associated with ice trapping and ice damage across different ship types, the accident loss parameter update module modifies relevant parameters based on the specific type of vessel used by the system user, thereby obtaining a more accurate estimate of economic losses.

Images