Risk influence factor analysis method for marine traffic accidents

Abstract

The invention is applicable to the field of marine traffic accident research, and provides a marine traffic accident risk influence factor analysis method, which comprises the following steps of: mining an association relationship among risk influence factors in different ship type marine traffic accidents from different ship type marine traffic accident survey reports through an ARM technology; based on the association relationship, constructing a risk interaction network of risk influence factors in different ship type marine traffic accidents through a CN theory, and analyzing topological feature data of the risk interaction network; according to the topological characteristic data and a dynamic analysis model constructed based on a WINGS method, obtaining a causal relationship among risk influence factors in different ship type marine traffic accidents; and based on the causal relationship and an antagonism hierarchical topology model established based on an AISM method, determining the causal hierarchical relationship among the risk influence factors in the marine traffic accidents of different ship types.

Description

Technical Field

[0001] The present application belongs to the field of marine traffic accident research, and in particular to a method for analyzing risk influencing factors of marine traffic accidents. Background Art

[0002] Maritime transport is the backbone of globalized trade and plays an important role in international trade and global economic development. However, with the booming global maritime trade, the size of ships and the density of ship traffic are increasing, and the risk of ship accidents is also rising. According to the European Maritime Safety Agency (EMSA)'s "Annual Overview of Maritime Casualties and Accidents", a total of 23,814 maritime traffic accidents involving casualties were reported between 2014 and 2022. Considering that maritime traffic accidents often cause significant casualties, property losses and environmental damage, the International Maritime Organization (IMO) and other relevant departments have taken many measures to prevent maritime traffic accidents. Despite this, it is still difficult to completely avoid the occurrence of maritime traffic accidents. Therefore, in order to reduce the occurrence of accidents and improve the safety of maritime transportation, it is crucial to study the risks of maritime traffic accidents.

[0003] In the field of marine traffic accident risk analysis, traditional fault tree analysis (FTA) and Bayesian network (BN) are widely used. For example, some researchers combined FTA and fuzzy cognitive reliability error analysis (CREAM) to study the impact of human error on oil tanker grounding accidents, and conducted quantitative risk analysis of bulk carrier fire and explosion accidents through fuzzy FTA, presenting the results in the form of probability. In addition, some researchers proposed a model combining FTA and BN to provide specific preventive measures for bulk carriers with fire and explosion accidents, and proposed a cause assessment model for grounding accidents and ship-to-ship collision accidents, providing a useful tool for dynamic risk assessment and risk analysis.

[0004] However, the existing methods for risk analysis of maritime traffic accidents lack comprehensive insights into accident RIFs (risk influencing factors), and mainly focus on maritime traffic accidents involving specific ship types. Research on accident RIFs of different types of ships is relatively rare. At the same time, there is little research on the heterogeneous coupling relationship between RIFs at different levels, and the understanding of the transitivity of RIFs of maritime traffic accidents is not comprehensive and specific enough. Summary of the invention

[0005] The embodiment of the present application provides a method for analyzing risk influencing factors of maritime traffic accidents, which can solve the problems that the existing maritime traffic accident risk analysis methods in the prior art lack comprehensive insight into accident RIFs (risk influencing factors), and mainly focus on maritime traffic accidents of specific ship types, while research on accident RIFs of different types of ships is relatively rare. At the same time, there is little research on the heterogeneous coupling relationship between RIFs at various levels, and the transmission cognition of RIFs of maritime traffic accidents is not comprehensive and specific enough.

[0006] In a first aspect, an embodiment of the present application provides a method for analyzing risk influencing factors of a maritime traffic accident, comprising:

[0007] Through ARM technology, the correlation between risk influencing factors in marine traffic accidents of different types of ships is excavated in the marine traffic accident investigation reports of different types of ships;

[0008] Based on the association relationship, a risk interaction network of risk influencing factors in marine traffic accidents of different ship types is constructed through CN theory, and the topological characteristic data of the risk interaction network is analyzed;

[0009] Obtaining the causal relationship between risk influencing factors in marine traffic accidents of different ship types based on the topological feature data and the preset dynamic analysis model;

[0010] Based on the causal relationship and the adversarial hierarchical topological model established based on the AISM method, the causal hierarchical relationship between risk influencing factors in marine traffic accidents of different ship types is determined.

[0011] Optionally, the step of mining the correlation between risk influencing factors in marine traffic accidents of different types of ships in the marine traffic accident investigation reports of different types of ships by using ARM technology includes:

[0012] Based on the marine traffic accident investigation reports of different ship types, a database of risk influencing factors is established, wherein the risk influencing factors are factors that affect marine traffic accidents of ships;

[0013] Through the preset mining rules in the ARM technology, the correlation between the risk influencing factors in marine traffic accidents of different ship types is mined in the database. The mathematical expression of the preset mining rules is:

[0014]

[0015]

[0016] Among them, Support(E) represents the support of item E, L E represents the number of occurrences of item E in the data, and L represents the number of all data; Indicates the confidence level from item E to item F, Indicates the degree of improvement from item E to item F.

[0017] Optionally, the step of constructing a risk interaction network of risk influencing factors in marine traffic accidents of different ship types based on the association relationship through CN theory, and analyzing the topological feature data of the risk interaction network includes:

[0018] Based on the association relationship, a risk interaction network of risk influencing factors in marine traffic accidents of different ship types is constructed through CN theory, and the risk interaction network is characterized by an adjacency matrix to obtain a target matrix;

[0019] Calculate the node dissimilarity between different network nodes in the risk interaction network according to the target matrix and the first calculation formula;

[0020] Calculate the edge dissimilarity between different edges in the risk interaction network according to the target matrix and the second calculation formula;

[0021] Using the node dissimilarity and the edge dissimilarity as topological feature data of the risk interaction network;

[0022] Wherein, the first calculation formula is:

[0023]

[0024] Where N1 is the number of nodes in the first risk interaction network, N2 is the number of nodes in the second risk interaction network, is the average node degree of the first risk interaction network, is the average node degree of the second risk interaction network, d 1i is the node degree of the ith node in the first risk interaction network, d 2i is the node degree of the ith node in the second risk interaction network;

[0025] The second calculation formula is:

[0026]

[0027] Where M1 is the number of edges in the first risk interaction network, M2 is the number of edges in the second risk interaction network, is the average edge weight in the second risk interaction network, is the average edge weight in the second risk interaction network, w 1i is the edge weight of the ith edge in the first risk interaction network, w 2i is the edge weight of the ith edge in the second risk interaction network.

[0028] Optionally, before the step of obtaining the causal relationship between risk influencing factors in marine traffic accidents of different ship types according to the topological feature data and the preset dynamic analysis model, the method further includes:

[0029] The preset dynamic analysis model is constructed based on data-driven complex network data. The dynamic analysis model is a model based on data-driven and used to execute the WINGS algorithm.

[0030] Optionally, the step of obtaining the causal relationship between risk influencing factors in marine traffic accidents of different ship types based on the topological feature data and a preset dynamic analysis model includes:

[0031] By using the WINGS algorithm in the preset dynamic analysis model, the influence degree, the influence degree, the centrality degree and the cause degree among the risk influencing factors in marine traffic accidents of different ship types are obtained according to the topological feature data;

[0032] Determine the causal relationship between risk influencing factors in marine traffic accidents of different ship types based on the influence degree, affected degree, centrality degree and cause degree;

[0033] The mathematical expression of the WINGS algorithm is:

[0034]

[0035] CT i = l i +m i

[0036] RT i = l i -m i

[0037] Among them, t ij represents the comprehensive impact of risk influencing factor i on risk influencing factor j; l i Indicates the influence of risk factor i; m i Indicates the degree of influence of risk influencing factor i; CT i represents the centrality of risk influencing factor i; RT i Represents the causal degree of risk influencing factor i.

[0038] Optionally, the step of determining the causal hierarchical relationship between risk influencing factors in marine traffic accidents of different ship types based on the causal relationship and the adversarial hierarchical topological model established based on the AISM algorithm includes:

[0039] Inputting the causal relationship into the adversarial hierarchical topology model, so that the adversarial hierarchical topology model uses the causal relationship as a guide for the guided AISM algorithm to generate UP type classification and DOWN type classification;

[0040] Based on the UP type classification and the DOWN type classification, the causal hierarchical relationship between risk influencing factors in marine traffic accidents of different ship types is determined.

[0041] Optionally, the mathematical representation of the adversarial hierarchical topological model is:

[0042] Rs(x i )={x j |n ij =1}

[0043] Qs(x j )={x i |n ji =1}

[0044] Ts(Ha i )=RS(Ha i )∩QS(Ha i )

[0045] Among them, Rs(x i ) represents x i The reachable set reaches; Qs(x j ) represents x i Starting from the predecessor set; Ts(x i ) represents x i The intersection of the reachable set and the antecedent set.

[0046] In a second aspect, the present application provides a device for analyzing risk influencing factors of marine traffic accidents, the device comprising:

[0047] The information receiving module is used to mine the correlation between the risk influencing factors of marine traffic accidents of different types of ships in the marine traffic accident investigation reports of different types of ships through ARM technology;

[0048] A first analysis module is used to construct a risk interaction network of risk influencing factors in marine traffic accidents of different ship types based on the association relationship through CN theory, and analyze the topological characteristic data of the risk interaction network;

[0049] The second analysis module is used to obtain the causal relationship between risk influencing factors in marine traffic accidents of different ship types based on the topological feature data and the preset dynamic analysis model;

[0050] The third analysis module is used to determine the causal hierarchical relationship between risk influencing factors in marine traffic accidents of different ship types based on the causal relationship and the adversarial hierarchical topological model established based on the AISM method.

[0051] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for analyzing risk influencing factors of marine traffic accidents as described above is implemented.

[0052] In a fourth aspect, the present application provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the risk influencing factor analysis method for marine traffic accidents as described above.

[0053] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0054] Through ARM technology, the correlation between risk influencing factors in marine traffic accidents of different ship types is mined in the marine traffic accident investigation reports of different ship types; based on the correlation, the risk interaction network of risk influencing factors in marine traffic accidents of different ship types is constructed through CN theory, and the topological feature data of the risk interaction network is analyzed; according to the topological feature data and the preset dynamic analysis model, the causal relationship between risk influencing factors in marine traffic accidents of different ship types is obtained; based on the causal relationship and the adversarial hierarchical topological model established based on the AISM method, the causal hierarchical relationship between risk influencing factors in marine traffic accidents of different ship types is determined. A comprehensive analysis model for marine traffic accidents composed of association rule mining (ARM), complex network (CN), nonlinear weighted influence measurement system (WINGS) and adversarial interpretation structure model (AISM) is proposed, which fills the research gap in the field of analyzing risk influencing factors of marine traffic accidents of different ship types. Extending the multidimensional factor analysis method from static to dynamic greatly enhances the understanding of the development dynamics of marine traffic accidents, so that preventive measures can be more accurately applied to specific ship types. This dynamic analysis method not only improves the effectiveness of accident prevention, but also provides a reliable basis for formulating more accurate prevention plans for marine traffic accidents of different types of ships. In addition, the present invention proposes and applies a new comprehensive analysis model, and constructs RINs for different types of marine traffic accidents with this model. By deeply analyzing the topological characteristics of these networks, the present invention explores the key RIFs in various types of marine traffic accidents and reveals the complex relationship between the RIFs. By analyzing the causal relationship of RIFs of marine traffic accidents of various types of ships from the perspective of dynamic information transmission, the present invention can more accurately reveal the impact of different RIFs on marine traffic accidents. This new comprehensive analysis model not only determines the key RIFs of marine traffic accidents of different types of ships, but also provides a new perspective for understanding the transmissibility of RIFs. These achievements have important reference value for formulating heterogeneous prevention plans for marine traffic accidents of different types of ships, and significantly improve the scientificity and effectiveness of maritime safety risk management. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0056] Figure 1 It is a flowchart of a method for analyzing risk influencing factors of marine traffic accidents provided by an embodiment of the present application;

[0057] Figure 2 This is a schematic diagram of the structure of a device for analyzing risk influencing factors of marine traffic accidents according to an embodiment of the present application;

[0058] Figure 3 is a structural block diagram of a computer device in one embodiment;

[0059] Figure 4 It is a structural block diagram of a computer device in another embodiment. DETAILED DESCRIPTION

[0060] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0061] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0062] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0063] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0064] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0065] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0066] Figure 1 A schematic flow chart of a method for analyzing risk factors of marine traffic accidents provided by the present application is shown. An embodiment of the present application provides a method for analyzing risk factors of marine traffic accidents, including:

[0067] S101. Through ARM technology, the correlation between risk influencing factors in marine traffic accidents of different types of ships can be found in the marine traffic accident investigation reports of different types of ships;

[0068] In a possible implementation, the step of mining the correlation between risk influencing factors in marine traffic accidents of different types of ships in marine traffic accident investigation reports of different types of ships by using ARM technology includes:

[0069] Based on the marine traffic accident investigation reports of different ship types, a database of risk influencing factors is established, wherein the risk influencing factors are factors that affect marine traffic accidents of ships;

[0070] Through the preset mining rules in the ARM technology, the correlation between the risk influencing factors in marine traffic accidents of different ship types is mined in the database. The mathematical expression of the preset mining rules is:

[0071]

[0072] Among them, Support(E) represents the support of item E, L E represents the number of occurrences of item E in the data, and L represents the number of all data; Indicates the confidence level from item E to item F, Indicates the degree of improvement from item E to item F.

[0073] For example, according to the marine traffic accident investigation reports of different ship types provided by the maritime investigation agency, bulk carriers, container ships, fishing boats and oil tankers are the most frequent types of ships in marine traffic accidents. ARM is used to mine the above-mentioned marine traffic accident investigation reports of different ship types. Specifically, after many experiments, the minimum support threshold and the minimum confidence threshold are determined to be 0.3 and 0.5 respectively, and then the Apriori algorithm is used to mine the association relationship of the accident data (marine traffic accident investigation reports of different ship types) of bulk carriers, containers, fishing boats and oil tankers.

[0074] S102, based on the association relationship, constructing a risk interaction network of risk influencing factors in marine traffic accidents of different ship types through CN theory, and analyzing the topological characteristic data of the risk interaction network;

[0075] For example, CN is a network structure that abstracts the elements in a complex system as nodes and the relationships between the elements as edges. It is widely used to check the safety and reliability of safety systems, for example, risk coupling related to ships, personnel, management and environment.

[0076] Specifically, based on the mined association rules, RIN (risk interaction network) was established for the four ship types, and RIN was used to analyze the coupling relationship between RIFs.

[0077] S103, obtaining the causal relationship between risk influencing factors in marine traffic accidents of different ship types according to the topological feature data and the preset dynamic analysis model;

[0078] In a possible implementation manner, before the step of obtaining the causal relationship between risk influencing factors in marine traffic accidents of different ship types according to the topological feature data and the preset dynamic analysis model, the step further includes:

[0079] The preset dynamic analysis model is constructed based on data-driven complex network data. The dynamic analysis model is a model based on data-driven and used to execute the WINGS algorithm.

[0080] For example, the WINGS method is a method that evaluates the intrinsic strength and impact strength of complex relationships between system components. This method can be used to analyze the RIFs characteristics and causal relationships of marine traffic accidents of different ship types, and reveal the intertwined relationships between RIFs in complex systems. The WINGS model (preset dynamic analysis model) is constructed using data-driven complex network data. Compared with the traditional WINGS method, the data-driven WINGS model can significantly reduce the subjectivity brought by expert scoring, thereby improving the objectivity and credibility of the analysis. On the other hand, the application of the WINGS model can effectively solve the problems of complex networks in analyzing nonlinear correlations and influences between RIFs, thereby more accurately revealing the interactions between RIFs in complex systems.

[0081] S104. Based on the causal relationship and the adversarial hierarchical topological model established based on the AISM method, determine the causal hierarchical relationship between risk influencing factors in marine traffic accidents of different ship types.

[0082] In a possible implementation, the steps of constructing a risk interaction network of risk influencing factors in marine traffic accidents of different ship types based on the association relationship through CN theory, and analyzing the topological feature data of the risk interaction network include:

[0083] Based on the association relationship, a risk interaction network of risk influencing factors in marine traffic accidents of different ship types is constructed through CN theory, and the risk interaction network is characterized by an adjacency matrix to obtain a target matrix;

[0084] Calculate the node dissimilarity between different network nodes in the risk interaction network according to the target matrix and the first calculation formula;

[0085] Calculate the edge dissimilarity between different edges in the risk interaction network according to the target matrix and the second calculation formula;

[0086] Using the node dissimilarity and the edge dissimilarity as topological feature data of the risk interaction network;

[0087] Wherein, the first calculation formula is:

[0088]

[0089] Where N1 is the number of nodes in the first risk interaction network, N2 is the number of nodes in the second risk interaction network, is the average node degree of the first risk interaction network, is the average node degree of the second risk interaction network, d 1i is the node degree of the ith node in the first risk interaction network, d2i is the node degree of the ith node in the second risk interaction network;

[0090] The second calculation formula is:

[0091]

[0092] Where M1 is the number of edges in the first risk interaction network, M2 is the number of edges in the second risk interaction network, is the average edge weight in the second risk interaction network, is the average edge weight in the second risk interaction network, w 1i is the edge weight of the ith edge in the first risk interaction network, w 2i is the edge weight of the ith edge in the second risk interaction network.

[0093] For example, the node dissimilarity and edge dissimilarity between RINs of different ship types are analyzed based on CN theory. The node dissimilarity and edge dissimilarity between RINs of different ship types are shown in Table 1. As can be seen from Table 1, the node dissimilarity and edge dissimilarity of fishing vessel RIN and oil tanker RIN are the largest, which indicates that the two types of ships have large differences in RIN structure. To a certain extent, it can be reflected that there are significant differences in the RIFs of marine traffic accidents between these two types of ships. This difference may be due to the large differences between fishing vessels and oil tankers in terms of ship size, route selection and crew quality. As can be seen from Table 1, the node dissimilarity of RIN of bulk carriers and RIN of container ships is the smallest, indicating that they have certain similarities in RIN structure. This similarity may mean that there are some common RIFs in marine traffic accidents of bulk carriers and container ships. This phenomenon may be because both types of ships are cargo ships and their safety management strategies are very similar. It can be found from Table 1 that the edge dissimilarity of the RIN of bulk carriers and the RIN of oil tankers is the smallest, which indicates that they have a high similarity in the edge connection mode of RIN. This similarity shows that although there are certain differences in the composition of the nodes of RIN between bulk carriers and oil tankers, the correlation between their accident RIFs may be relatively similar. Based on this finding, the present invention can further explore the possible reasons for this situation. For example, although they have differences in cargo types and ship operations, their ship structures and technical characteristics of navigation operations may be relatively similar, and they may use similar types of engines, navigation equipment, etc. Therefore, although they have certain differences in the composition of nodes in RINs, the edge connection mode is very similar. This difference research helps the relevant maritime departments to better understand the characteristics and risks of different types of ships, and then formulate targeted safety management policies to reduce the occurrence of accidents.

[0094] Table 1 Dissimilarity between networks

[0095]

[0096] Specifically, node dissimilarity: Node dissimilarity Nd is a characteristic used to compare the degree of difference between nodes in two networks. It reveals the similarities and differences between different network structures by comparing the degree distribution, neighbor set, feature vector, and structural similarity of nodes between the two networks. Suppose the two networks are Web1 and Web2, and the node degrees of Web1 and Web2 are d 1i and d 2i The comparison of node dissimilarity can be expressed using the standard deviation of node degree (node ​​degree refers to the number of edges associated with the node). The calculation formula (first preset formula) is:

[0097]

[0098] Where N1 and N2 are the number of nodes in the networks Web1 and Web2 respectively. and are the average node degrees in Web1 and Web2, respectively.

[0099] Edge dissimilarity: Edge dissimilarity Ed is a metric that measures the degree of difference in the edges (connections) between two networks. The existence of dissimilarity between the edges of two networks may mean that there is a significant difference in the distribution of connection weights between their edges. Suppose the edge weights in Web1 and Web2 are w 1i and w 2i The calculation formula (the second preset formula) is:

[0100]

[0101] Among them, M1 and M2 are the number of edges in the networks Web1 and Web2 respectively. and are the average edge weights in Web1 and Web2, respectively.

[0102] In a possible implementation, the step of obtaining the causal relationship between risk influencing factors in marine traffic accidents of different ship types based on the topological feature data and the preset dynamic analysis model includes:

[0103] By using the WINGS algorithm in the preset dynamic analysis model, the influence degree, the influence degree, the centrality degree and the cause degree among the risk influencing factors in marine traffic accidents of different ship types are obtained according to the topological feature data;

[0104] Determine the causal relationship between risk influencing factors in marine traffic accidents of different ship types based on the influence degree, affected degree, centrality degree and cause degree;

[0105] The mathematical expression of the WINGS algorithm is:

[0106]

[0107] CT i = l i +m i

[0108] RT i = l i -m i

[0109] Among them, t ij represents the comprehensive impact of risk influencing factor i on risk influencing factor j; l i Indicates the influence of risk factor i; m i Indicates the degree of influence of risk influencing factor i; CT i represents the centrality of risk influencing factor i; RT i Represents the causal degree of risk influencing factor i.

[0110] For example, the centrality (CT) and causality (RT) of each RIF are obtained by the WINGS method, aiming to accurately analyze the dynamic causal relationship between RIFs. The centrality (CT) is an indicator to evaluate the importance and influence of each RIF in a marine traffic accident. The causality (RT) distinguishes whether RIFs are the initiators or recipients of the impact. When RT>0, it indicates that this RIF is a causal factor affecting other RIFs. The larger the value of RT, the easier it is for this RIF to affect other RIFs in a complex system. When RT<0, it indicates that this RIF is a result factor susceptible to the influence of other RIFs. The smaller the value of RT, the more susceptible these RIFs are to the influence of other RIFs in a complex system. Therefore, when comprehensively considered, the causal factors with larger centrality values ​​are the key RIFs and driving factors for the occurrence of accidents. Taking the oil tanker marine traffic accident as an example (as shown in Table 2), including CC2 (good company safety culture), TAS2 (5≤ crew time at sea <10 years), TRA2 (adequate company training), and TD2 (high traffic density). These RIFs play an important role in the accident chain and can serve as initial triggers before an accident occurs. Different RIFs work in different ways and may have adverse effects throughout the voyage. For example, company management factors represented by CC2 (good company safety culture) and TRA2 (adequate company training) pose significant safety risks in ship operations and may hinder emergency response, thereby facilitating accidents. It is crucial to strengthen the management and prevention measures of these RIFs to minimize the risk of accidents and ensure maritime safety. At the same time, the result factors with large centrality values ​​play a vital role in the risk transmission within the system. Taking the fishing vessel maritime traffic accident as an example (as shown in Table 3), including PF2 (guarantee PSC / FSC inspection), SM1 (sufficient manning on board), VIS2 (no administrative violations in supervision), SFC1 (complete and valid seafarer certificate), R2 (appropriate management regulations), SC1 (complete and valid ship certificate). These RIFs are largely affected by other RIFs, but due to their large centrality values, any changes in them may have a wide impact on the evolution of maritime traffic accidents. Therefore, in order to avoid the negative impact of these RIFs and prevent them from further exacerbating the occurrence of marine traffic accidents, special attention needs to be paid to these RIFs (such as safety management systems and personnel's educational background). In order to minimize the risk of accidents and ensure maritime safety, it is necessary to ensure the stability and compliance of these key RIFs by taking necessary measures.

[0111] From the perspective of causality, the present invention can deeply understand the dynamic information transmission between RIFs. By analyzing the impact of each RIF on other RIFs, the causal chain between key RIFs in the accident process can be revealed. This analysis helps to understand the dynamic mechanism of accident occurrence, as well as the interaction and influence between different RIFs in the accident chain. Taking the oil tanker marine traffic accident as an example, it can be observed that SFS2 (no defects in the company's safety management system) is one of the most influential RIFs. In the causal analysis, the present invention can further explore the impact of this RIF on other RIFs. For example, a good company safety management system can promote regular inspection and maintenance of ship equipment, thereby reducing the risk of accidents. On the contrary, defects in the management system may cause the crew to ignore safety regulations and increase the possibility of accidents. In addition, another important RIF is EB2 (good crew educational background). In the causal analysis, the impact of crew educational background on the occurrence of accidents can be explored. A good educational background makes the crew more capable of dealing with emergencies, thereby reducing the occurrence of accidents. On the contrary, crew members with lower education levels may not be able to properly deal with emergencies, increasing the risk of accidents. By analyzing the causal relationship between these RIFs, the dynamic process of accident occurrence can be better understood. Specifically, defects in the management system may lead to incomplete implementation of safety regulations, which in turn affects the behavior and response capabilities of crew members, ultimately increasing the risk of accidents. Analyzing the transmission of this dynamic information helps to reveal the root causes of accidents and provides an important reference for formulating effective prevention strategies.

[0112] Table 2 Influence of factors on oil tankers

[0113]

[0114]

[0115] Table 3 The influence degree of factors on fishing vessels

[0116]

[0117] In a possible implementation, the step of determining the causal hierarchical relationship between risk influencing factors in marine traffic accidents of different ship types based on the causal relationship and the adversarial hierarchical topological model established based on the AISM algorithm includes:

[0118] Inputting the causal relationship into the adversarial hierarchical topology model, so that the adversarial hierarchical topology model uses the causal relationship as a guide for the guided AISM algorithm to generate UP type classification and DOWN type classification;

[0119] Based on the UP type classification and the DOWN type classification, the causal hierarchical relationship between risk influencing factors in marine traffic accidents of different ship types is determined.

[0120] Optionally, the mathematical representation of the adversarial hierarchical topological model is:

[0121] Rs(x i )={x j |n ij =1}

[0122] Qs(x j )={x i |n ji =1}

[0123] Ts(Ha i )=RS(Ha i )∩QS(Ha i )

[0124] Among them, Rs(x i ) represents x i The reachable set reaches; Qs(x j ) represents x i Starting from the predecessor set; Ts(x i ) represents x i The intersection of the reachable set and the antecedent set.

[0125] For example, the AISM method uses the reachable set, the antecedent set, and the intersection of the reachable set and the antecedent set to construct UP type and DOWN type classifications respectively based on cause and effect. The calculation formulas are shown in equations (13) to (15).

[0126] Rs(x i )={x j |n ij =1} (1)

[0127] Qs(x j )={x i |n ji =1} (2)

[0128] Ts(Ha i )=RS(Ha i )∩QS(Ha i ) (3)

[0129] Where Rs(x i ) represents x i The reachable set reaches; Qs(x j ) represents x i Starting from the predecessor set; Ts(x i ) represents x iThe intersection of the reachable set and the antecedent set.

[0130] For UP type classification, when Rs(x i )=Ts(x i ), take out Rs(x i ) and put these factors at the top level, then delete all the reachable sets that are related to Ts(x i ) intersects to form a new reachable set and intersection. For DOWN type, when Qs(x j )=Ts(x j ), take out Qs(x j ) and put these factors at the bottom, then delete all the factors in the previous set that are related to Ts(x i ) to form a new predecessor set and intersection set. Iterate in sequence until all factors are taken out.

[0131] For example, in the confrontation hierarchical topology, the lower the RIFs are, the more root-causing they are, and the higher the RIFs are, the more direct they are. As can be seen from Table 4, the RIFs of marine traffic accidents involving bulk carriers, fishing boats, and oil tankers are divided into three levels: direct influencing factors (D-RIFs), indirect influencing factors (I-RIFs), and fundamental influencing factors (U-RIFs). The RIFs of marine traffic accidents involving container ships are divided into two levels: direct influencing factors (D-RIFs) and fundamental influencing factors (U-RIFs).

[0132] Table 4 Results of the adversarial hierarchy analysis for bulk carriers

[0133]

[0134]

[0135] In order to conduct a comprehensive analysis of D-RIFs, I-RIFs and U-RIFs, this paper takes a complete hierarchical topology map as an example for research. Figure 2 As shown, there is a complete system in the fishing boat DOWN type confrontation hierarchical topology diagram, namely L3→L2→L1.

[0136] There are complex interactions between the RIFs of L1, L2, and L3. By analyzing the RIFs of the L3 layer, it can be found that most of them belong to the company's safety management factors. The RIFs of the L3 layer are the core RIFs that affect maritime traffic accidents. It can be seen that the company's safety management factors are the fundamental factors affecting maritime traffic accidents of fishing vessels. The L2 layer covers RIFs in many aspects such as human, environmental, and management, indicating the diversity of I-RIFs. Therefore, it will be a very complex project to effectively prevent and control these I-RIFs involving many aspects. Although these I-RIFs do not act directly, they play an intermediary role in the process of information transmission. Therefore, by effectively managing this layer, the information transmission channel from U-RIFs to D-RIFs can be blocked. In the L3 layer, each RIF is combined into a loop. D-RIFs are the direct RIFs of maritime traffic accidents. Taking WL3 (channel width / ship length ≥ 2) in this loop as an example, when WL3 is disturbed, that is, when the ratio of channel width to ship length changes, the risk of ship capsizing, grounding or mechanical damage may increase. Solving D-RIFs can reduce the probability of accidents or mitigate the severity of accidents more quickly and effectively.

[0137] In the second aspect, the present application provides a device for analyzing risk factors affecting marine traffic accidents, such as Figure 2 As shown, the device comprises:

[0138] The information receiving module 201 is used to mine the correlation between the risk influencing factors of marine traffic accidents of different types of ships from the marine traffic accident investigation reports of different types of ships through ARM technology;

[0139] The first analysis module 202 is used to construct a risk interaction network of risk influencing factors in marine traffic accidents of different ship types based on the association relationship through CN theory, and analyze the topological feature data of the risk interaction network;

[0140] The second analysis module 203 is used to obtain the causal relationship between risk influencing factors in marine traffic accidents of different ship types based on the topological feature data and the preset dynamic analysis model;

[0141] The third analysis module 204 is used to determine the causal hierarchical relationship between risk influencing factors in marine traffic accidents of different ship types based on the causal relationship and the adversarial hierarchical topological model established based on the AISM method.

[0142] In one embodiment, an electronic device 300 is provided, such as Figure 3As shown, it includes a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the computer program 311, the following steps are implemented: by using ARM technology, the correlation between the risk influencing factors in marine traffic accidents of different types of ships is mined in the marine traffic accident investigation reports of different types of ships; based on the correlation, a risk interaction network of the risk influencing factors in marine traffic accidents of different types of ships is constructed by CN theory, and the topological feature data of the risk interaction network is analyzed; according to the topological feature data and the preset dynamic analysis model, the causal relationship between the risk influencing factors in marine traffic accidents of different types of ships is obtained; based on the causal relationship and the adversarial hierarchical topological model established based on the AISM method, the causal hierarchical relationship between the risk influencing factors in marine traffic accidents of different types of ships is determined.

[0143] In one embodiment, a computer readable storage medium 400 is provided, such as Figure 4 As shown, a computer program 411 is stored thereon, and the computer program 411 implements the following steps when executed by the processor: through ARM technology, the correlation relationship between the risk influencing factors in marine traffic accidents of different ship types is mined in the marine traffic accident investigation reports of different ship types; based on the correlation relationship, a risk interaction network of the risk influencing factors in marine traffic accidents of different ship types is constructed through CN theory, and the topological feature data of the risk interaction network is analyzed; according to the topological feature data and the preset dynamic analysis model, the causal relationship between the risk influencing factors in marine traffic accidents of different ship types is obtained; based on the causal relationship and the adversarial hierarchical topological model established based on the AISM method, the causal hierarchical relationship between the risk influencing factors in marine traffic accidents of different ship types is determined.

[0144] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0145] In the embodiments provided in the present application, it should be understood that the disclosed devices / network equipment / terminal equipment and methods can be implemented in other ways. For example, the device / network equipment / terminal equipment embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0146] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0147] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for analyzing risk factors affecting marine traffic accidents, characterized in that: The method comprises: Through ARM technology, the correlation between risk influencing factors in marine traffic accidents of different types of ships is excavated in the marine traffic accident investigation reports of different types of ships; Based on the association relationship, a risk interaction network of risk influencing factors in marine traffic accidents of different ship types is constructed through CN theory, and the topological characteristic data of the risk interaction network is analyzed; Obtaining the causal relationship between risk influencing factors in marine traffic accidents of different ship types based on the topological feature data and the preset dynamic analysis model; Based on the causal relationship and the adversarial hierarchical topological model established based on the AISM method, the causal hierarchical relationship between risk influencing factors in marine traffic accidents of different ship types is determined.

2. The method for analyzing risk factors affecting marine traffic accidents according to claim 1, characterized in that: The steps of mining the correlation between the risk influencing factors in marine traffic accidents of different types of ships in the marine traffic accident investigation reports of different types of ships by using the ARM technology include: Based on the marine traffic accident investigation reports of different ship types, a database of risk influencing factors is established, wherein the risk influencing factors are factors that affect marine traffic accidents of ships; Through the preset mining rules in the ARM technology, the correlation between the risk influencing factors in marine traffic accidents of different ship types is mined in the database. The mathematical expression of the preset mining rules is: Among them, Support(E) represents the support of item E, L E represents the number of occurrences of item E in the data, and L represents the number of all data; Indicates the confidence level from item E to item F, Indicates the degree of improvement from item E to item F.

3. The method for analyzing risk factors affecting marine traffic accidents according to claim 1, characterized in that: The steps of constructing a risk interaction network of risk influencing factors in marine traffic accidents of different ship types based on the association relationship through CN theory, and analyzing the topological characteristic data of the risk interaction network include: Based on the association relationship, a risk interaction network of risk influencing factors in marine traffic accidents of different ship types is constructed through CN theory, and the risk interaction network is characterized by an adjacency matrix to obtain a target matrix; Calculate the node dissimilarity between different network nodes in the risk interaction network according to the target matrix and the first calculation formula; Calculate the edge dissimilarity between different edges in the risk interaction network according to the target matrix and the second calculation formula; Using the node dissimilarity and the edge dissimilarity as topological feature data of the risk interaction network; Wherein, the first calculation formula is: Where N1 is the number of nodes in the first risk interaction network, N2 is the number of nodes in the second risk interaction network, is the average node degree of the first risk interaction network, is the average node degree of the second risk interaction network, d1i is the node degree of the ith node in the first risk interaction network, and d2i is the node degree of the ith node in the second risk interaction network; The second calculation formula is: Where M1 is the number of edges in the first risk interaction network, M2 is the number of edges in the second risk interaction network, is the average edge weight in the second risk interaction network, is the average edge weight in the second risk interaction network, w1i is the edge weight of the ith edge in the first risk interaction network, and w2i is the edge weight of the ith edge in the second risk interaction network.

4. The method for analyzing risk factors affecting marine traffic accidents according to claim 1, characterized in that: Before the step of obtaining the causal relationship between risk influencing factors in marine traffic accidents of different ship types according to the topological feature data and the preset dynamic analysis model, the method further includes: The preset dynamic analysis model is constructed based on data-driven complex network data. The dynamic analysis model is a model based on data-driven and used to execute the WINGS algorithm.

5. The method for analyzing risk factors affecting marine traffic accidents according to claim 4, characterized in that: The step of obtaining the causal relationship between risk influencing factors in marine traffic accidents of different ship types based on the topological feature data and the preset dynamic analysis model includes: By using the WINGS algorithm in the preset dynamic analysis model, the influence degree, the influence degree, the centrality degree and the cause degree among the risk influencing factors in marine traffic accidents of different ship types are obtained according to the topological feature data; Determine the causal relationship between risk influencing factors in marine traffic accidents of different ship types based on the influence degree, affected degree, centrality degree and cause degree; The mathematical expression of the WINGS algorithm is: CT i =l i +m i RT i =l i -m i Among them, t ij represents the comprehensive impact of risk influencing factor i on risk influencing factor j; l i Indicates the influence of risk factor i; m i Indicates the degree of influence of risk influencing factor i; CT i represents the centrality of risk influencing factor i; RT i Represents the causal degree of risk influencing factor i.

6. The method for analyzing risk factors affecting marine traffic accidents according to claim 1, characterized in that: The step of determining the causal hierarchical relationship between risk influencing factors in marine traffic accidents of different ship types based on the causal relationship and the adversarial hierarchical topological model established based on the AISM algorithm includes: Inputting the causal relationship into the adversarial hierarchical topology model, so that the adversarial hierarchical topology model uses the causal relationship as a guide for the guided AISM algorithm to generate UP type classification and DOWN type classification; Based on the UP type classification and the DOWN type classification, the causal hierarchical relationship between risk influencing factors in marine traffic accidents of different ship types is determined.

7. The method for analyzing risk factors affecting marine traffic accidents according to claim 6, characterized in that: The mathematical representation of the adversarial hierarchical topological model is: Rs(x i )={x j |n ij =1} Qs(x j )={x i |n ji =1} Ts(Ha i )=RS(Ha i )∩QS(Ha i ) Among them, Rs(x i ) represents x i The reachable set reaches; Qs(x j ) represents x i Starting from the predecessor set; Ts(x i ) represents x i The intersection of the reachable set and the antecedent set.

8. A risk influencing factor analysis device for marine traffic accidents, characterized in that: The device comprises: The information receiving module is used to mine the correlation between the risk influencing factors of marine traffic accidents of different types of ships in the marine traffic accident investigation reports of different types of ships through ARM technology; A first analysis module is used to construct a risk interaction network of risk influencing factors in marine traffic accidents of different ship types based on the association relationship through CN theory, and analyze the topological characteristic data of the risk interaction network; The second analysis module is used to obtain the causal relationship between risk influencing factors in marine traffic accidents of different ship types based on the topological feature data and the preset dynamic analysis model; The third analysis module is used to determine the causal hierarchical relationship between risk influencing factors in marine traffic accidents of different ship types based on the causal relationship and the adversarial hierarchical topological model established based on the AISM method.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method for analyzing risk influencing factors of marine traffic accidents as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for analyzing risk influencing factors of marine traffic accidents as described in any one of claims 1 to 7 is implemented.

Images