Transportation accident cause chain identification method, device and equipment

Abstract

The invention provides a transportation accident cause chain identification method, device and equipment, and relates to the technical field of transportation safety. The transportation accident cause chain identification method comprises the following steps: acquiring a cause classification data set according to transportation accident information to be identified and a cause classification model; the cause classification model is established according to an HFACS model and historical transportation accident information and comprises at least two levels, each level comprises at least one suspected cause, the suspected causes of different levels belong to different types, and the suspected cause of the previous level in two adjacent levels can directly influence the suspected cause of the next level; through a significance test method, obtaining a correlation degree between suspected causes of adjacent levels in the cause classification data set; obtaining a target cause chain according to the association degree; the target cause chain comprises suspected causes of at least two adjacent levels. According to the scheme, the cause chain between the hidden danger and the accident in the dangerous cargo transportation scene can be identified.

Description

Technical Field

[0001] This invention relates to the field of transportation safety technology, and in particular to a method, apparatus, and equipment for identifying the causal chain of a transportation accident. Background Technology

[0002] Hazardous chemicals, due to their flammability, explosiveness, reactivity, corrosiveness, toxicity, and radioactivity, are highly susceptible to safety accidents during transportation, seriously threatening human health, property safety, and environmental safety. Furthermore, accidents involving vehicles transporting hazardous materials can lead to secondary injuries due to the characteristics of the cargo itself, with incalculable consequences. The accident causal chains in the transportation of hazardous materials are more complex and unique than those in general transportation due to the unique characteristics of the goods, transportation conditions, and legal requirements. Currently, the prevention and management of hazardous materials transportation accidents lack causal chain identification algorithms, making it difficult to accurately assess the contribution of each factor to the accident, thus affecting the effectiveness and relevance of safety management measures. Summary of the Invention

[0003] The purpose of this invention is to provide a method, apparatus, and equipment for identifying the causal chain of transportation accidents, in order to solve the problem of the lack of analysis methods for the causal paths between hidden dangers and accidents in the context of hazardous goods transportation in the prior art.

[0004] To achieve the above objectives, the present invention is implemented as follows:

[0005] In a first aspect, embodiments of the present invention provide a method for identifying the causal chain of a transportation accident, comprising:

[0006] Based on the information of the transportation accident to be identified and the cause classification model, a cause classification dataset is obtained; wherein, the cause classification model is established based on the HFACS (Human Factors Analysis and Classification System) model and historical transportation accident information; the cause classification model includes at least two levels, each level includes at least one suspected cause, the suspected causes of different levels belong to different types, and the suspected cause of the upper level can directly affect the suspected cause of the lower level in two adjacent levels;

[0007] The degree of association between suspected causes at adjacent levels in the cause classification dataset is obtained by using a significance test method.

[0008] Based on the degree of correlation, a target causative chain is obtained; wherein, the target causative chain includes at least two adjacent levels of suspected causes.

[0009] Optionally, in the transportation accident causation chain identification method, the causation classification model includes a first level, a second level adjacent to the first level, a third level adjacent to the second level, and a fourth level adjacent to the third level; the suspected causation at the first level belongs to the management organization type, the suspected causation at the second level belongs to the unsafe supervision type, the suspected causation at the third level belongs to the unsafe behavior premise type, and the suspected causation at the fourth level belongs to the unsafe behavior type.

[0010] Optionally, in the transportation accident causation chain identification method, the step of obtaining the degree of association between suspected causes at adjacent levels in the causation classification dataset through a saliency test includes:

[0011] The first degree of association between suspected causes at adjacent levels in the cause classification dataset is obtained by using the chi-square test method.

[0012] The second degree of association between suspected causes at adjacent levels in the cause classification dataset is obtained by using the concession ratio test method.

[0013] Optionally, in the transportation accident causation chain identification method, the step of obtaining the target causation chain based on the degree of correlation includes:

[0014] Based on the first degree of correlation, at least one suspected causal chain is obtained;

[0015] Based on the second degree of association corresponding to the suspected causative chain, the target causative chain is obtained.

[0016] Optionally, in the transportation accident causation chain identification method, the step of obtaining the degree of association between suspected causes at adjacent levels in the causation classification dataset through a saliency test includes:

[0017] The degree of association between the suspected causes at the target level and the suspected causes indicated by the cause clustering results in the cause classification dataset is obtained by using a significance test method; wherein, the target level is the level adjacent to the level where the suspected cause indicated by the cause clustering results is located.

[0018] Optionally, in the method for identifying the causal chain of a transportation accident, before obtaining the degree of association between suspected causes at adjacent levels in the causal classification dataset through a saliency test, the method further includes:

[0019] Based on the causation classification dataset, cluster analysis is performed on the suspected causes corresponding to the transportation accident information to be identified, and the causation clustering results are obtained.

[0020] Secondly, embodiments of the present invention also provide a transportation accident causation chain identification device, comprising:

[0021] The first acquisition module is used to acquire a cause classification dataset based on the transportation accident information to be identified and the cause classification model; wherein, the cause classification model is established based on the Human Factors Analysis and Classification System (HFACS) model and historical transportation accident information; the cause classification model includes at least two levels, each level includes at least one suspected cause, the suspected causes of different levels belong to different types, and the suspected cause of the upper level can directly affect the suspected cause of the lower level in two adjacent levels;

[0022] The second acquisition module is used to acquire the degree of association between suspected causes at adjacent levels in the cause classification dataset through a significance test method.

[0023] The third acquisition module is used to acquire the target causal chain based on the degree of correlation; wherein the target causal chain includes at least two adjacent levels of suspected causes.

[0024] Thirdly, embodiments of the present invention also provide a transportation accident causation chain identification device, comprising: a processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the transportation accident causation chain identification method as described in the first aspect.

[0025] Fourthly, embodiments of the present invention also provide a readable storage medium storing a program that, when executed by a processor, implements the transportation accident causation chain identification method as described in the first aspect.

[0026] Fifthly, embodiments of the present invention also provide a computer program product, including computer instructions, which, when executed by a processor, implement the transportation accident causation chain identification method as described in the first aspect.

[0027] The beneficial effects of the above-described technical solution of the present invention are as follows:

[0028] In this embodiment of the invention, a causation classification dataset is obtained based on the information of the transportation accident to be identified and the causation classification model. The causation classification model is established based on the HFACS model and historical transportation accident information, including at least two levels. Each level includes at least one suspected causation. Suspected causations at different levels belong to different types, and the suspected causation of the upper level can directly influence the suspected causation of the lower level in adjacent levels. The degree of correlation between suspected causations at adjacent levels in the causation classification dataset is obtained through a significance test. Based on the degree of correlation, a target causation chain is obtained. The target causation chain includes suspected causations at least two adjacent levels. In this way, the causal paths between hidden dangers and accidents in hazardous goods transportation scenarios can be effectively obtained, providing scientific support for accident prevention and decision support for further improving the safety level of the hazardous goods industry. Attached Figure Description

[0029] Figure 1 This is a flowchart illustrating the method for identifying the causal chain of a transportation accident in an embodiment of the present invention.

[0030] Figure 2 This is a schematic diagram illustrating the principle of dangerous goods transportation accidents in an embodiment of the present invention;

[0031] Figure 3 This is a schematic diagram of a derivative accident chain of hazardous chemicals in an embodiment of the present invention;

[0032] Figure 4 This is a diagram illustrating the analytical framework of the HFACS model in this embodiment of the invention.

[0033] Figure 5 This is a schematic diagram of the causal chain in an embodiment of the present invention.

[0034] Figure 6 This is a schematic diagram of the structure of the transportation accident causation chain identification device in an embodiment of the present invention;

[0035] Figure 7 This is a hardware block diagram of the transportation accident causation chain identification device in an embodiment of the present invention. Detailed Implementation

[0036] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0037] In various embodiments of the present invention, it should be understood that the sequence number of each process described below does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0038] In addition, the terms "system" and "network" are often used interchangeably in this article.

[0039] The terms "first," "second," etc., used in the specification and claims of this invention are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, not limited in number; for example, the first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0040] Embodiments of the present invention, such as Figure 1 As shown, a method for identifying the causal chain of a transportation accident is provided, including:

[0041] S101, Based on the information of the transportation accident to be identified and the cause classification model, obtain the cause classification dataset; wherein, the cause classification model is established based on the HFACS model and historical transportation accident information; the cause classification model includes at least two levels, each level includes at least one suspected cause, the suspected causes of different levels belong to different types, and the suspected cause of the upper level can directly affect the suspected cause of the lower level in two adjacent levels.

[0042] In this embodiment of the invention, optionally, the transportation accident information to be identified is hazardous goods transportation accident information, and correspondingly, the historical transportation accident information is historical hazardous goods transportation accident information. Optionally, the hazardous goods include dangerous goods or hazardous chemicals. The causation classification dataset is used to indicate the frequency of each suspected causation in the transportation accident information to be identified corresponding to each causation classification model.

[0043] It should be noted that, prior to S101, the method further includes:

[0044] A causation classification model is established based on historical transportation accident information and HFACS. Specifically, the historical transportation accident information is analyzed, classified, and summarized to obtain the historical transportation accident causation statistics shown in Table 1 below. Table 1 includes four columns: type, unsafe cause, frequency statistics, and specific cause. Then, a causation classification model is established based on the historical transportation accident causation statistics and the HFACS model.

[0045]

[0046]

[0047] Table 1

[0048] Furthermore, such as Figure 2 As shown, causal analysis of historical transportation accident information reveals that the transportation system, composed of factors such as people, vehicles, roads, hazardous goods, environment, and management, is more complex, prone to accidents, and has more severe consequences than ordinary freight transportation due to the interaction and influence of these factors. Among these, personnel, vehicles, equipment, and hazardous goods are internal causes of accidents, roads and the environment are external causes, and management influences other factors through its control. When conditions, behaviors, or external stimuli exceed safety limits, the entire system loses control, leading to undesirable consequences, i.e., accidents.

[0049] Different types of hazardous chemical road transport accidents can cause varying degrees of social harm depending on the physical and chemical properties of the transported chemicals, weather conditions, population density, and surrounding road structure. For example... Figure 3 As shown, when analyzing derivative accidents of road accidents involving hazardous chemicals, the following categories should be reconsidered:

[0050] (1) Accident types: collision, rollover, and fall.

[0051] (2) Properties of hazardous chemicals: diffusion, flammability, and toxicity.

[0052] (3) Quantity of hazardous chemicals: More, Normal, Less.

[0053] (4) Weather: Normal, windy, rainy / snowy, foggy.

[0054] (5) Surrounding population density: high, medium, low.

[0055] (6) Surrounding road structure: expressways, national highways, provincial highways, urban roads, and rural roads.

[0056] (7) Sensitive surrounding locations: schools, hospitals, densely populated areas, important protective targets, gas stations, residential buildings, etc.

[0057] (8) Derivative accidents caused by this: explosion, fire, toxic gas leak, traffic congestion, environmental pollution.

[0058] Optionally, the causation classification model includes a first level, a second level adjacent to the first level, a third level adjacent to the second level, and a fourth level adjacent to the third level. The suspected causations in the first level belong to the organizational influence type, the suspected causations in the second level belong to the unsafe supervision type, the suspected causations in the third level belong to the precondition type of unsafe behavior, and the suspected causations in the fourth level belong to the unsafe behavior type.

[0059] It is understood that the first level is the level above the second level, the second level is the level below the first level, the second level is the level above the third level, the third level is the level below the second level, the third level is the level above the fourth level, and the fourth level is the level above the third level.

[0060] It should be noted that the HFACS model is a comprehensive human error analysis method that summarizes four levels of accidents, with each level including at least one potential cause, such as... Figure 4 As shown, these are the first level (level 1), the second level (level 2), the third level (level 3), and the fourth level (level 4). Since the suspected causes at level 1 belong to the organizational influence type, level 1 can also be called the organizational influence level; since the suspected causes at level 2 belong to the unsafe monitoring type, level 2 can also be called the unsafe monitoring level; since the suspected causes at level 3 belong to the preconditions for unsafe behavior type, level 3 can also be called the preconditions for unsafe behavior level; and since the suspected causes at level 4 belong to the unsafe behavior type, level 4 can also be called the unsafe behavior level. The following is a detailed explanation of each level:

[0061] Organizational hierarchy: While high-level decisions may not appear to bear direct responsibility for accidents, flaws in decision-making that are not immediately identified can easily lead to them. Organizational impacts can be categorized into three types: resource management, referring to executive-level decisions related to the allocation and maintenance of organizational resources, such as human resources, funding, and facility allocation; organizational climate, referring to factors that affect work efficiency within the organization, such as organizational structure, culture, and policies; and organizational workflows, referring to policies and regulations that constrain and define organizational work, such as organizational work speed and work procedure standards.

[0062] Unsafe oversight levels: Unsafe oversight refers to the management-level causes of accidents, and is divided into four types: Inadequate oversight, which may manifest as a lack of guidance for crew members, failure to provide standardized operating procedures, or insufficient training in responding to flight incidents; Inappropriate planning and tasks, mainly referring to organizational-level plans that may lead to accidents, such as unreasonable task assignments and staffing; Failure to correct known problems, which refers to allowing problems to persist despite issues in personnel, equipment, training, and other safety-related areas; and Regulatory violations, which refer to supervisory personnel intentionally violating regulatory provisions or regulations.

[0063] Unsafe behavior prerequisites hierarchy: The prerequisites for unsafe behavior are the direct causes that lead to it, including two types: the operator's substandard individual state and the operator's substandard task execution. The operator's substandard individual state is divided into three categories: adverse mental state, referring to mental states that affect the performance of work tasks, such as mental fatigue, anxiety, loss of situational awareness, etc.; adverse physical state, referring to pathological or physiological states that hinder safe operation, such as hallucinations, disorientation, or a cold, etc.; and physiological and psychological limitations, referring to the inherently limited range of abilities of a person in terms of physiology and psychology. In the aviation field, this specifically refers to the pilot's inability to meet the requirements for completing certain tasks, such as limited night vision, insufficient comprehension and reaction speed, etc.

[0064] Unsafe behavior levels: Unsafe behavior is the most direct cause of accidents and is divided into two types: mistakes and violations. Mistakes refer to a person's psychological state or actions failing to meet task requirements, and are divided into three categories: skill-related mistakes, which refer to errors in skill-related behaviors, such as improper attention allocation or memory errors; decision-making mistakes, which refer to actions that do not conform to the requirements of the current situation, and are further divided into process errors, selection errors, and problem-solving errors, covering all stages of planning and decision-making in flight missions; and cognitive errors, which refer to errors caused by pilots' improper understanding of information in the current situation, such as misunderstandings of visual or spatial information leading to incorrect judgments.

[0065] Based on historical transportation accident information and the HFACS model, this invention establishes a causal classification model as shown in Table 2 below, thereby providing a more comprehensive analysis of the causes and influencing factors of human error in dangerous goods transportation accidents.

[0066]

[0067]

[0068] Table 2

[0069] S102, using a significance test method, obtain the degree of association between suspected causes at adjacent levels in the cause classification dataset.

[0070] In this embodiment of the invention, the causal classification model is used to analyze, classify and summarize the transportation accident information to be identified, and the frequency of each suspected causal cause in the causal classification model is counted to obtain a suspected causal dataset.

[0071] By using a significance test method, the frequency of suspected causes corresponding to adjacent levels in the cause classification dataset is statistically analyzed to obtain the degree of association between the suspected causes in the cause classification dataset.

[0072] Optionally, obtaining the degree of association between suspected causes at adjacent levels in the cause classification dataset through a significance test method includes one of the following implementation methods:

[0073] Implementation method 1: Obtain the first degree of association between suspected causes at adjacent levels in the cause classification dataset by using the chi-square test method;

[0074] The second degree of association between suspected causes at adjacent levels in the cause classification dataset is obtained by using the concession ratio test method.

[0075] It should be noted that the chi-square test is used to determine the causal relationship between suspected causes at adjacent levels in the causal classification dataset. The odds-radio (OR) test is used to analyze whether a suspected cause at a higher level in the causal classification dataset increases the probability of a suspected cause occurring at a lower level.

[0076] Implementation Method 2: Using a significance test method, obtain the degree of correlation between the suspected causes at the target level in the causal classification dataset and the suspected causes indicated by the causal clustering results. The target level is the level adjacent to the level where the suspected causes indicated by the causal clustering results are located.

[0077] Regarding this second embodiment, in S102, before obtaining the degree of association between suspected causes at adjacent levels in the cause classification dataset using a significance test method, the method further includes:

[0078] Based on the causation classification dataset, cluster analysis is performed on the suspected causes corresponding to the transportation accident information to be identified, and the causation clustering results are obtained.

[0079] It should be noted that before S102, the causal clustering results of the transportation accident to be identified need to be obtained based on the frequency of each suspected causal factor in the causal classification dataset. Then, in S102, the degree of correlation between the suspected causal factors at the target level and the suspected causal factors indicated by the causal clustering results can be obtained through a significance test method.

[0080] Further, in the second embodiment described above: obtaining the degree of association between the suspected causes at the target level in the causative classification dataset and the suspected causes indicated by the causative clustering results through a significance test method includes:

[0081] The first degree of association between the suspected causes at the target level in the cause classification dataset and the suspected causes indicated by the cause clustering results is obtained by using the chi-square test method.

[0082] The second degree of association between the suspected causes at the target level in the cause classification dataset and the suspected causes indicated by the cause clustering results is obtained by using the concession ratio test method.

[0083] Compared to the first embodiment, this second embodiment can locate the cause of the transportation accident to be identified more quickly, thereby improving the efficiency of obtaining the target cause chain in S103.

[0084] Specifically, the chi-square test method is explained in conjunction with Table 3 below. Table 3 is used to indicate whether there is a significant causal relationship between the suspected cause of the previous level and the suspected cause of the next level in adjacent levels, that is, whether there is a strong association.

[0085]

[0086] Table 3

[0087] The formulas for calculating the chi-square values ​​in Table 3 above are as follows:

[0088]

[0089] Where f represents the frequency and n represents the sum of rows or columns.

[0090] It should be noted that after calculating the chi-square value, the degrees of freedom between the suspected causative factors at the previous and next levels can also be calculated. The mathematical formula for calculating the degrees of freedom is as follows:

[0091] df = (r-1)(c-1)

[0092] Where r represents a row and c represents a column.

[0093] The mathematical formula for calculating the concession ratio is as follows:

[0094]

[0095] It should be noted that when the OR value is greater than 1, it means that the suspected cause at the upper level in an adjacent hierarchy will increase the probability of the suspected cause at the lower level occurring; when the OR value is less than 1, it means that the suspected cause at the upper level in an adjacent hierarchy will not increase the probability of the suspected cause at the lower level occurring.

[0096] S103, based on the degree of correlation, obtain the target causative chain, which includes at least two adjacent levels of suspected causes.

[0097] Optionally, S103, obtaining the target causal chain based on the degree of correlation includes:

[0098] Based on the first degree of correlation, at least one suspected causal chain is obtained;

[0099] Based on the second degree of association corresponding to the suspected causative chain, the target causative chain is obtained.

[0100] In this embodiment of the invention, at least one suspected causal chain is first obtained based on the first degree of correlation between suspected causes at adjacent levels in the causal classification dataset. A suspected causal chain includes two suspected causes located at adjacent levels, such as suspected cause A at the first level and suspected cause B at the second level. A suspected causal chain indicates the transmission relationship from suspected cause A to suspected cause B, where the two have a strong correlation; suspected cause A can directly affect suspected cause B, leading to the occurrence of suspected cause B. At least one suspected causal chain is a causal chain with a first degree of correlation greater than a first correlation threshold, which is a preset value used to identify suspected causal chains with a strong correlation.

[0101] Furthermore, based on the second degree of association corresponding to the suspected causal chain, a target causal chain with a second degree of association greater than a second association threshold is obtained; the second association threshold is a preset value used to obtain the final target causal chain from at least one suspected causal chain.

[0102] Next, taking the information of 143 dangerous goods transportation accidents to be identified as an example, the information of the transportation accidents to be identified is decomposed and encoded according to the cause classification model. If the information of the transportation accidents to be identified includes the suspected cause in the cause classification model, it is encoded as "1"; if the information of the transportation accidents to be identified does not include the suspected cause in the cause classification model, it is encoded as "0", thereby establishing a suspected cause dataset. The suspected cause dataset can be represented by a matrix of 143 rows and 19 columns of 0s and 1s, as shown in the following formula, and can be illustrated in Table 4 below.

[0103]

[0104]

[0105]

[0106] Table 4

[0107] Then, the chi-square value, p-value, OR value, and 95% confidence interval between the suspected causes at the upper and lower levels were calculated, and all suspected causes with p less than 0.5, OR greater than 1, and chi-square value greater than the chi-square threshold were compiled, as shown in Table 5 below.

[0108]

[0109] Table 5

[0110] According to Table 5 above, we can obtain the following: Figure 5The suspected causal chain shown by the black dashed line is further analyzed by obtaining the target suspected causal chain based on the OR value corresponding to the suspected causal chain, such as... Figure 5 The three target causal chains are shown by the black solid lines in the image:

[0111] The causal chain of the target is as follows: A2 Lack of safety culture → B3 Failure to correct problems → D2 Poor physiological state → F2 Decision-making errors. The enterprise's safety production organization atmosphere is insufficient, with production being prioritized over safety and safety awareness being low. As a result, the enterprise has failed to correct problems, such as failing to effectively implement pre-departure safety checks for drivers and failing to pay timely attention to the physical health of personnel, leading to improper emergency response by drivers and accidents.

[0112] Target causal chain ②: B1 Insufficient supervision → E2 Poor technical environment → F2 Decision-making errors, inadequate management of enterprise vehicle driving safety, failure to effectively implement "three inspections a day", failure to conduct regular technical level assessments of vehicles, resulting in vehicles being driven on the road with defects, leading to accidents.

[0113] Target causal chain ③: D1 Poor mental state → F3 Cognitive error, driver's lack of concentration, fatigue driving, resulting in poor mental state of the person, unable to effectively process situational information, leading to the accident.

[0114] Alternatively, cluster analysis can be performed on the causal classification dataset as shown in Table 4 to obtain the causal clustering results shown in Table 6 below. Three accident groups are obtained from the causal classification dataset. The most probable suspected cause corresponding to accident group 1 is D2 poor physiological state, the most probable suspected cause corresponding to accident group 2 is F1 skill error, and the most probable suspected cause corresponding to accident group 3 is B4 supervision violation. Thus, the suspected cause chain and the target cause chain can be obtained based on the suspected cause corresponding to each accident group.

[0115]

[0116] Table 6

[0117] In summary, the embodiments of the present invention provide a method for identifying the causal chain of transportation accidents, specifically a method for identifying the causal chain of multi-cause, multi-level hazardous goods transportation accidents based on the HFACS model, which solves the problem of the lack of analysis methods for the causal paths between hidden dangers and accidents in hazardous goods transportation scenarios in the prior art.

[0118] like Figure 6 As shown, this embodiment of the invention also provides a transportation accident causation chain identification device, comprising:

[0119] The first acquisition module 601 is used to acquire a cause classification dataset based on the transportation accident information to be identified and the cause classification model; wherein, the cause classification model is established based on the Human Factors Analysis and Classification System (HFACS) model and historical transportation accident information; the cause classification model includes at least two levels, each level includes at least one suspected cause, the suspected causes of different levels belong to different types, and the suspected cause of the upper level can directly affect the suspected cause of the lower level in two adjacent levels;

[0120] The second acquisition module 602 is used to acquire the degree of association between suspected causes at adjacent levels in the cause classification dataset through a significance test method.

[0121] The third acquisition module 603 is used to acquire the target causal chain based on the degree of correlation; wherein the target causal chain includes at least two adjacent levels of suspected causes.

[0122] Optionally, in the aforementioned transportation accident causation chain identification device, the causation classification model includes a first level, a second level adjacent to the first level, a third level adjacent to the second level, and a fourth level adjacent to the third level; the suspected causation at the first level belongs to the management organization type, the suspected causation at the second level belongs to the unsafe supervision type, the suspected causation at the third level belongs to the unsafe behavior premise type, and the suspected causation at the fourth level belongs to the unsafe behavior type.

[0123] Optionally, in the aforementioned transportation accident causation chain identification device, the second acquisition module 602 is specifically used for:

[0124] The first degree of association between suspected causes at adjacent levels in the cause classification dataset is obtained by using the chi-square test method.

[0125] The second degree of association between suspected causes at adjacent levels in the cause classification dataset is obtained by using the concession ratio test method.

[0126] Optionally, in the aforementioned transportation accident causation chain identification device, the third acquisition module 603 is specifically used for:

[0127] Based on the first degree of correlation, at least one suspected causal chain is obtained;

[0128] Based on the second degree of association corresponding to the suspected causative chain, the target causative chain is obtained.

[0129] Optionally, in the aforementioned transportation accident causation chain identification device, the second acquisition module 602 is specifically used for:

[0130] The degree of association between the suspected causes at the target level and the suspected causes indicated by the cause clustering results in the cause classification dataset is obtained by using a significance test method; wherein, the target level is the level adjacent to the level where the suspected cause indicated by the cause clustering results is located.

[0131] Optionally, the transportation accident causation chain identification device further includes:

[0132] The acquisition module is used to perform cluster analysis on the suspected causes corresponding to the transportation accident information to be identified based on the cause classification dataset, and obtain the cause clustering results.

[0133] It should be noted that the transportation accident causation chain identification device provided in this embodiment of the invention is a device capable of executing the above-described transportation accident causation chain identification method. Therefore, all embodiments of the above-described transportation accident causation chain identification method are applicable to this device and can achieve the same or similar technical effects.

[0134] like Figure 7 As shown, this embodiment of the invention also provides a transportation accident causation chain identification device, including: a processor 701; and a memory 702 connected to the processor 701 via a bus interface, the memory 702 being used to store programs and data used by the processor 701 when performing operations, and the processor 701 calling and executing the programs and data stored in the memory 702.

[0135] Processor 701 is used to read the program from memory 702 and execute the following procedures:

[0136] Based on the information of the transportation accident to be identified and the cause classification model, a cause classification dataset is obtained; wherein, the cause classification model is established based on the Human Factors Analysis and Classification System (HFACS) model and historical transportation accident information; the cause classification model includes at least two levels, each level includes at least one suspected cause, the suspected causes of different levels belong to different types, and the suspected cause of the upper level can directly affect the suspected cause of the lower level in two adjacent levels;

[0137] The degree of association between suspected causes at adjacent levels in the cause classification dataset is obtained by using a significance test method.

[0138] Based on the degree of correlation, a target causative chain is obtained; wherein, the target causative chain includes at least two adjacent levels of suspected causes.

[0139] The transportation accident causation chain identification device also includes a transceiver 703, which is used to receive and send data in the controller of the processor 701.

[0140] Among them, Figure 7 In this context, the bus architecture can include any number of interconnected buses and bridges, specifically linking various circuits of one or more processors represented by processor 701 and memory represented by memory 702 together. The bus architecture can also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. The bus interface provides an interface. Transceiver 703 can be multiple elements, including transmitters and transceivers, providing a unit for communicating with various other devices over a transmission medium. For different user equipment, the user interface 704 can also be an interface capable of connecting external or internal devices, including but not limited to keypads, displays, speakers, microphones, joysticks, etc.

[0141] The processor 701 is responsible for managing the bus architecture and general processing, while the memory 702 can store the data used by the processor 501 when performing operations.

[0142] Optionally, the cause classification model includes a first level, a second level adjacent to the first level, a third level adjacent to the second level, and a fourth level adjacent to the third level; the suspected causes in the first level belong to the management organization type, the suspected causes in the second level belong to the unsafe supervision type, the suspected causes in the third level belong to the unsafe behavior premise type, and the suspected causes in the fourth level belong to the unsafe behavior type.

[0143] Optionally, the processor 701 is specifically used to read the program and perform the following steps:

[0144] The first degree of association between suspected causes at adjacent levels in the cause classification dataset is obtained by using the chi-square test method.

[0145] The second degree of association between suspected causes at adjacent levels in the cause classification dataset is obtained by using the concession ratio test method.

[0146] Optionally, the processor 701 is specifically used to read the program and perform the following steps:

[0147] Based on the first degree of correlation, at least one suspected causal chain is obtained;

[0148] Based on the second degree of association corresponding to the suspected causative chain, the target causative chain is obtained.

[0149] Optionally, the processor 701 is specifically used to read the program and perform the following steps:

[0150] The degree of association between the suspected causes at the target level and the suspected causes indicated by the cause clustering results in the cause classification dataset is obtained by using a significance test method; wherein, the target level is the level adjacent to the level where the suspected cause indicated by the cause clustering results is located.

[0151] Optionally, the processor 701 is also used to read the program and perform the following steps:

[0152] Based on the causation classification dataset, cluster analysis is performed on the suspected causes corresponding to the transportation accident information to be identified, and the causation clustering results are obtained.

[0153] A specific embodiment of the present invention also provides a computer-readable storage medium storing a computer program thereon. When the program is executed by a processor, it implements the steps in the above-described method for identifying the causal chain of transportation accidents and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0154] In addition, embodiments of the present invention also provide a computer program product, including computer instructions, which, when executed by a processor, implement the above-described... Figure 1 The various processes of the method embodiments shown can achieve the same technical effect, and will not be described again here to avoid repetition.

[0155] In the several embodiments provided by this invention, it should be understood that the disclosed methods and apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between devices or units, and may be electrical, mechanical, or other forms.

[0156] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can be physically comprised separately, or two or more units can be integrated into one unit. The integrated unit described above can be implemented in hardware or in the form of hardware plus software functional units.

[0157] The integrated units implemented as software functional units described above can be stored in a computer-readable storage medium. These software functional units, stored in a storage medium, include several instructions that cause a computer device (which may be a personal computer, server, or network device, etc.) to execute some steps of the transmission and reception methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0158] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for identifying the causal chain of a transportation accident, characterized in that, include: Based on the information of the transportation accident to be identified and the cause classification model, a cause classification dataset is obtained; wherein, the cause classification model is established based on the Human Factors Analysis and Classification System (HFACS) model and historical transportation accident information; the cause classification model includes at least two levels, each level includes at least one suspected cause, the suspected causes of different levels belong to different types, and the suspected cause of the upper level can directly affect the suspected cause of the lower level in two adjacent levels; The degree of association between suspected causes at adjacent levels in the cause classification dataset is obtained by using a significance test method. Based on the degree of correlation, a target causative chain is obtained; wherein, the target causative chain includes at least two adjacent levels of suspected causes.

2. The method for identifying the causal chain of a transportation accident according to claim 1, characterized in that, The cause classification model includes a first level, a second level adjacent to the first level, a third level adjacent to the second level, and a fourth level adjacent to the third level; the suspected causes of the first level belong to the management organization type, the suspected causes of the second level belong to the unsafe supervision type, the suspected causes of the third level belong to the unsafe behavior premise type, and the suspected causes of the fourth level belong to the unsafe behavior type.

3. The method for identifying the causal chain of a transportation accident according to claim 1, characterized in that, The step of obtaining the degree of association between suspected causes at adjacent levels in the cause classification dataset through a significance test method includes: The first degree of association between suspected causes at adjacent levels in the cause classification dataset is obtained by using the chi-square test method. The second degree of association between suspected causes at adjacent levels in the cause classification dataset is obtained by using the concession ratio test method.

4. The method for identifying the causal chain of a transportation accident according to claim 3, characterized in that, The step of obtaining the target causal chain based on the degree of correlation includes: Based on the first degree of correlation, at least one suspected causal chain is obtained; Based on the second degree of association corresponding to the suspected causative chain, the target causative chain is obtained.

5. The method for identifying the causal chain of a transportation accident according to claim 1, characterized in that, The step of obtaining the degree of association between suspected causes at adjacent levels in the cause classification dataset through a significance test method includes: The degree of association between the suspected causes at the target level and the suspected causes indicated by the cause clustering results in the cause classification dataset is obtained by using a significance test method; wherein, the target level is the level adjacent to the level where the suspected cause indicated by the cause clustering results is located.

6. The method for identifying the causal chain of a transportation accident according to claim 5, characterized in that, Before obtaining the degree of association between suspected causes at adjacent levels in the cause classification dataset through a significance test, the method further includes: Based on the causation classification dataset, cluster analysis is performed on the suspected causes corresponding to the transportation accident information to be identified, and the causation clustering results are obtained.

7. A device for identifying the causal chain of a transportation accident, characterized in that, include: The first acquisition module is used to acquire a cause classification dataset based on the transportation accident information to be identified and the cause classification model; wherein, the cause classification model is established based on the Human Factors Analysis and Classification System (HFACS) model and historical transportation accident information; the cause classification model includes at least two levels, each level includes at least one suspected cause, the suspected causes of different levels belong to different types, and the suspected cause of the upper level can directly affect the suspected cause of the lower level in two adjacent levels; The second acquisition module is used to acquire the degree of association between suspected causes at adjacent levels in the cause classification dataset through a significance test method. The third acquisition module is used to acquire the target causal chain based on the degree of correlation; wherein the target causal chain includes at least two adjacent levels of suspected causes.

8. A device for identifying the causal chain of a transportation accident, characterized in that, include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the transportation accident causal chain identification method as described in any one of claims 1 to 6.

9. A readable storage medium, characterized in that, The readable storage medium stores a program that, when executed by a processor, implements the transportation accident causation chain identification method as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, The method includes computer instructions that, when executed by a processor, implement the method for identifying the causal chain of a transportation accident as described in any one of claims 1 to 6.

Images