DEMATEL-AISM-based locomotive steward safety control influence factor analysis method

Abstract

The invention discloses a DEMATEL-AISM-based locomotive steward safety control influence factor analysis method. The method comprises the following steps: selecting influence factor indexes; identifying influence factors and constructing a locomotive crew safe driving influence factor index system; a DEMATEL method is constructed; the construction basis comprises the following steps: calculating the influence degree and the influenced degree of each element on other elements based on the logic relationship among the elements in the index system and the direct influence matrix; each element is a risk factor; an AISM model is constructed; the construction basis comprises the following steps: obtaining a relation matrix based on a decision matrix formed by the reason degree and centrality of each element, fusing game confrontation in a GAN (Generative Adversarial Network) on the basis of a classic interpretation structure model ISM, and obtaining a pair of simplest hierarchical topological graphs through opposite hierarchical extraction rules; and performing model analysis, namely analyzing the result, the hierarchy, the causal whole series and the loop, and finally giving suggestions according to the result. The invention further discloses a corresponding analysis method, electronic equipment and a computer readable storage medium.

Description

Technical Field

[0001] The invention relates to the technical field of railway operation safety assessment and monitoring, and in particular to a method for analyzing factors affecting safe operation of locomotive crew members based on DEMATEL-AISM. Background Art

[0002] With the rapid development of high-speed railways, while the scale of the road network and the quality of service are constantly improving, the safety assurance system is also continuously upgraded. Although the overall safety situation is improving, accidents still occur. These accidents may be caused by human negligence, equipment failure or other reasons. Among them, the proportion of accidents caused by human factors is becoming more and more prominent, gradually becoming the main constraint on improving railway driving safety and the biggest cause of fatal railway accidents. The role of locomotive crew members in railway transportation is vital, and their behavior and status are directly related to the safety of trains and passengers. Studies have shown that the lack of concentration of locomotive crew members is one of the common causes of accidents. Therefore, it is particularly important to explore the interaction and hierarchical relationship between the various risk factors affecting locomotive crew members.

[0003] At present, some scholars have adopted methods such as analytic hierarchy process (AHP), analytic network process (ANP), decision laboratory analysis (DEMATEL) or interpretive structural model (ISM) to analyze the complex relationship between various risk factors, and then assist decision makers in making trade-offs and choices. In general, DEMATEL and ISM provide relatively effective decision support for decision makers at different levels, which helps to deal with complex decision-making problems and has certain advantages in railway safety. However, they still cannot meet the requirements of accurate analysis of factors affecting safe operation. Summary of the invention

[0004] The purpose of the present invention is to design a method for analyzing factors affecting safe operation of locomotive crew members based on DEMATEL-AISM in view of the defects of the prior art. An extraction method is added to the ISM to obtain an adversarial interpretation structure model (AISM) to provide richer information. At the same time, the DEMATEL method and the AISM model are jointly analyzed to comprehensively understand the correlation and hierarchical relationship between various factors affecting the normal operation of the operator, so as to provide a scientific basis for formulating effective operation management strategies and improving railway driving safety.

[0005] The first aspect of the present invention is to provide a method for analyzing factors affecting safe operation of locomotive crew based on DEMATEL-AISM, comprising:

[0006] S1, selecting influencing factor indicators, wherein the selecting influencing factor indicators includes: preliminarily establishing an influencing factor framework through a large amount of literature research, wherein the influencing factor framework covers one or more of human factors, train factors, track environment, driving behavior, external and social factors, wherein different influencing factors will have different degrees of impact;

[0007] S2, identifying influencing factors based on the selected influencing factor indicators, and constructing an influencing factor indicator system for safe driving of locomotive crew members; S2 includes: adjusting the selected influencing factor indicators based on interviews with experts in the field of operating safety and / or questionnaire survey analysis, so as to obtain an influencing factor indicator system for safe driving of locomotive crew members;

[0008] S3, constructing a DEMATEL method based on the indicator system; wherein, the construction basis of the DEMATEL method includes: comparing each element in the indicator system in pairs, calculating the influence and influence of each element on other elements based on the logical relationship and direct influence matrix between the elements in the indicator system, thereby calculating the cause degree and centrality of each element and determining the causal relationship between the elements and the position of each element in the indicator system; wherein, each element is a risk factor;

[0009] S4, constructing an AISM model based on the indicator system; wherein the construction basis of the AISM model includes: obtaining a relationship matrix based on a decision matrix composed of the causality and centrality of each element, integrating the game confrontation in the generative adversarial network GAN on the basis of the classic explanatory structure model ISM, and obtaining a pair of simplest hierarchical topological graphs through opposite hierarchical extraction rules;

[0010] S5, conduct method analysis based on the DEMATEL method and the AISM model, and give corresponding suggestions; wherein, S5 includes: analyzing the results, levels, and the entire series and loops of cause and effect, and finally giving suggestions based on the results.

[0011] Preferably, the S1 comprises:

[0012] S11, determining the primary index, including: confirming the factors affecting the operator's safe operation through literature research, the factors affecting the operator's safe operation include one or more of human factors, train factors, track environment, driving behavior, external factors and social factors; based on the factors affecting the operator's safe operation, confirming four primary indexes affecting the operator's safe operation, the four primary indexes include human factors, train factors, external factors and social factors;

[0013] S12, determining the secondary indicators, including: determining the secondary indicators corresponding to the four primary indicators.

[0014] Preferably, the secondary indicators of S12 include:

[0015] For the first-level indicator human factors, the second-level indicators include: fatigue level, emotional state and psychological factors of the operator, physiological state, bad operating behavior, attitude and sense of responsibility of the operator, decision-making ability and cognitive ability of the operator, information and communication, and safety awareness;

[0016] For the first-level indicator train factors, the second-level indicators include: technical condition of the train, mechanical failures and equipment damage, new safety technologies and innovative applications, and intelligent operation assistance systems;

[0017] For the first-level indicator external factors, the second-level indicators include: animals or pedestrians and other trains;

[0018] For the first-level indicator social factors, the second-level indicators include: social environment, cultural background, economic status, work environment, education level, and social relations and interpersonal communication.

[0019] Preferably, S2 includes:

[0020] S21, based on the expert interview and questionnaire analysis method in the field of operation safety, the primary and secondary indicators are adjusted to establish a preliminary index system of factors affecting safe driving of locomotive crew members, and multiple influencing factor indicators are selected from the primary and secondary indicators; wherein the preliminary index system of factors affecting safe driving of locomotive crew members corresponds to the factors affecting safe operation of operators described in S11 and is classified into six major analysis indicators: human factors, train factors, track environment factors, driving behavior factors, social factors and external factors;

[0021] S22, deleting some unqualified indicators from the preliminary locomotive crew safe driving influencing factor indicator system, combining similar indicators, and modifying some indicators.

[0022] Preferably, S3 includes:

[0023] S31, based on risk factor F i The degree of influence on the risk factor Fj is encoded as the direct influence matrix O ij , the specific encoding method is O ij =F i F j The degree of influence is stipulated that when i = j, O ij =0, where i=1,2,3,…,n; j=1,2,3,…,n;

[0024] S32, normalizing the direct influence matrix O based on formula (1) to obtain the standard influence matrix M;

[0025]

[0026] S33, calculate the comprehensive influence matrix T(t ij ), where I represents the identity matrix;

[0027] T(t ij )=M(IM) -1 (2);

[0028] S34, based on the comprehensive influence matrix T(t ij ) Calculate the influence D i 、Influence C i 、Center degree M i R i , which are the measurement criteria for the influence of the four measurement factors in the method, and the formulas are shown in (3)-(6) below, where i = 1, 2, 3, ..., n; j = 1, 2, 3, ..., n; t (ij) is the element of the comprehensive influence matrix T;

[0029]

[0030] M i =D i +C i (5);

[0031] R i =D i -C i (6).

[0032] Preferably, S4 includes:

[0033] S41, based on the cause degree and centrality of each element, a decision matrix is ​​formed, a threshold λ (λ∈[0,1]) is introduced, and a relationship matrix A is obtained according to the partial order rule in the following formula (7); wherein the relationship matrix A is composed of a ij composition:

[0034]

[0035] S42, to further consider the influence of the factors themselves, the overall relationship matrix B is calculated according to the following formula (8); where I represents the unit matrix;

[0036] B = A + I (8);

[0037] S43, performing multiplication on the overall relationship matrix B based on the following formula (9) to calculate the reachable matrix R;

[0038] R=B k+1 =B k ≠B k-1(9);

[0039] Where k refers to the kth power of the overall relationship matrix B;

[0040] S44, based on the extraction rules of the result priority principle and the cause priority principle, the reachability matrix R is divided into levels and regions, and the skeleton matrix S is calculated based on formula (10);

[0041] S=R-(RI) 2 -I (10);

[0042] S45, drawing an adversarial multi-level hierarchical explanatory structural model, and then performing analysis based on the adversarial multi-level hierarchical explanatory structural model.

[0043] Preferably, S5 analyzes the results, levels, and the entire series and loops of causes and effects, and finally gives suggestions based on the results, including:

[0044] S51, analyze the results, including causal degree analysis, centrality analysis, influence analysis and influenced degree analysis; wherein: the causal degree analysis is used to reflect the contribution of a factor to the method as a whole. The higher the causal degree of a factor, the greater its influence on the method, and it is considered to be a key factor in the method; the centrality analysis represents the position of a factor in the entire causal network, that is, its status and importance in the causal relationship; the influence analysis includes analyzing the influence of a certain factor on other factors; in DEMATEL, the influence between factors is determined by quantifying the degree of correlation between influencing factors; the influenced degree analysis is used to analyze the degree to which a factor is influenced by other factors, so as to determine the factors that are more influenced by other factors.

[0045] S52, analyze the whole series of levels and causalities, including root cause analysis, transition cause analysis and adjacent cause analysis; wherein: the root cause level in the root cause analysis is the union of the lowest factors in the confrontation hierarchy diagram, and the essential cause level factors only send out directed arrows, but do not receive directed arrows, that is, they only affect other factors but are not affected by other factors; the transition cause analysis includes factors located in the middle layer of the confrontation hierarchy diagram after removing the essential cause level and adjacent cause level factors, which both receive directed arrows and send out directed arrows, and can be affected by other factors and affect other factors; the adjacent cause level in the adjacent cause analysis is the union of the uppermost factors in the confrontation hierarchy diagram, and the adjacent cause level factors only receive directed arrows, but do not send out directed arrows, that is, they are only affected by other factors but do not affect other factors, and the adjacent cause level factors are the most direct factors affecting the operator's safe operation;

[0046] S53, analyzing the loop, wherein the loop is a strong connection between the influencing factors;

[0047] S54, based on the analysis of S51-S53, take comprehensive and systematic measures to promote the safe operation of locomotive crew members and improve the overall safety operation level.

[0048] The second aspect of the present invention further provides a method for analyzing factors affecting safe operation of locomotive crew based on DEMATEL-AISM, comprising:

[0049] An influencing factor index selection module is used to select influencing factor indexes; wherein, the selection of influencing factor indexes includes: preliminarily establishing an influencing factor framework through a large amount of literature research, wherein the influencing factor framework covers one or more of human factors, train factors, track environment, driving behavior, external and social factors, wherein different influencing factors will have different degrees of influence;

[0050] A locomotive crew safe driving influencing factor index system construction module is used to identify influencing factors based on the selected influencing factor indexes and construct a locomotive crew safe driving influencing factor index system; the locomotive crew safe driving influencing factor index system construction module adjusts the selected influencing factor index based on interviews with experts in the field of operating safety and / or questionnaire survey analysis, thereby obtaining a locomotive crew safe driving influencing factor index system;

[0051] A DEMATEL method construction module is used to construct a DEMATEL method based on the indicator system; wherein, the construction basis of the DEMATEL method includes: comparing each element in the indicator system in pairs, calculating the influence and the influence of each element on other elements based on the logical relationship and direct influence matrix between the elements in the indicator system, thereby calculating the cause degree and centrality of each element and determining the causal relationship between the elements and the status of each element in the indicator system; wherein, each element is a risk factor;

[0052] An AISM model construction module is used to construct an AISM model based on the indicator system; wherein, the construction basis of the AISM model includes: obtaining a relationship matrix based on a decision matrix composed of the causal degree and centrality of each element, integrating the game confrontation idea in the generative adversarial network GAN on the basis of the classic explanatory structure model ISM, and obtaining a pair of simplest hierarchical topological graphs through opposite hierarchical extraction rules;

[0053] The influencing factor analysis module is used to perform method analysis based on the DEMATEL method and the AISM model and give corresponding suggestions; wherein the influencing factor analysis module analyzes the results, levels, and the entire series and loops of cause and effect, and finally gives suggestions based on the results.

[0054] A third aspect of the present invention provides an electronic device, comprising a processor and a memory, wherein the memory stores a plurality of instructions, and the processor is configured to read the instructions and execute the method described in the first aspect.

[0055] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a plurality of instructions, and the plurality of instructions can be read by a processor to execute the method described in the first aspect.

[0056] Beneficial effects of the method of the present invention:

[0057] The present invention provides a scientific method to evaluate and optimize the key factors of locomotive crew safety operation, thereby achieving objective identification and analysis of locomotive crew safety operation factors, improving the objectivity and scientificity of the analysis, and improving the safety and efficiency of railway transportation. At the same time, it helps relevant departments or individuals to better understand the current situation and potential risks of locomotive crew safety operation, so as to take corresponding measures. Compared with traditional safety analysis methods, the present invention achieves objective identification and analysis of locomotive crew safety operation factors, improves the objectivity and scientificity of the analysis; constructs a comprehensive analysis method based on DEMATEL-AISM, so that various factors can be comprehensively analyzed, providing decision makers with more targeted references, and providing a more scientific and effective means to improve the safety level of locomotive crew. Specifically including:

[0058] 1. Comprehensively consider the identification of factors affecting locomotive crew operation and the construction of an index system

[0059] Through interviews with experts in the field of railway operation safety, questionnaire surveys and analysis, a large number of repeated factors and other factors affecting train operators were adjusted and distinguished. Specifically, indicators with unclear analysis purposes and weak scientificity were deleted, indicators with strong similarity were merged, and some indicators were modified to make them easier to understand and more operational. Finally, an indicator framework for factors affecting safe driving of locomotive crew members was obtained, which includes 4 first-level indicators and 23 second-level indicators.

[0060] 2. Combination of DEMATEL method and AISM model

[0061] The DEMATEL method and the AISM model each have their limitations when analyzing factors that affect the safe operation of locomotive crews. The DEMATEL method can quantitatively analyze the relationship between factors, but it may ignore the competition and confrontation between factors, resulting in an incomplete explanation of complex methods. Although the AISM model takes into account the idea of ​​confrontation and can more deeply display the relationship between internal elements, it may lack the quantitative analysis ability of the DEMATEL method and it is difficult to accurately evaluate the degree of influence between factors. Based on this, the DEMATEL method in the DEMATEL-AISM model proposed in this paper provides the ability of quantitative analysis, which can accurately evaluate the degree of mutual influence between factors; while the AISM model takes into account the confrontation relationship and can more comprehensively understand the competition and confrontation relationship between factors. By combining the two, we can more comprehensively grasp the complexity of the method, identify more accurate and comprehensive influencing factors, and provide more effective solutions and guarantees for improving railway driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the related technologies, the drawings required for use in the specific embodiments or the related technical descriptions will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0063] Figure 1 A flow chart of a method provided according to an embodiment of the present invention;

[0064] Figure 2 Establish a flow chart for the DEMATEL-AISM model provided according to an embodiment of the present invention;

[0065] Figure 3 A structural diagram of an electronic device provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0066] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0068] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0069] Embodiment 1

[0070] See also Figure 1 This embodiment provides a method for analyzing factors affecting safe operation of locomotive crew based on DEMATEL-AISM, including:

[0071] S1, selecting influencing factor indicators; wherein, the selecting influencing factor indicators includes: preliminarily establishing an influencing factor framework through a large amount of literature research, wherein the influencing factor framework covers one or more of human factors, train factors, track environment, driving behavior, external and social factors, wherein different influencing factors will have different degrees of impact;

[0072] S2, identifying influencing factors based on the selected influencing factor indicators, and constructing an influencing factor indicator system for safe driving of locomotive crew members; S2 includes: adjusting the selected influencing factor indicators based on interviews with experts in the field of operating safety and / or questionnaire survey analysis, so as to obtain an influencing factor indicator system for safe driving of locomotive crew members;

[0073] S3, constructing a DEMATEL (Decision-making Trial and Evaluation Laboratory) model based on the indicator system; wherein, the construction basis of the DEMATEL method includes: comparing each element in the indicator system in pairs, calculating the influence and influence of each element on other elements based on the logical relationship and direct influence matrix between the elements in the indicator system, thereby calculating the causal degree and centrality of each element and determining the causal relationship between the elements and the status of each element in the indicator system; wherein, each element is a risk factor;

[0074] S4, constructing an AISM model based on the indicator system; wherein, the construction basis of the AISM model includes: obtaining a relationship matrix based on a decision matrix composed of the causality and centrality of each element, integrating the adversarial idea in the generative adversarial network (GAN) on the basis of the classic explanatory structural model (ISM), and obtaining a pair of simplest hierarchical topological graphs through opposite hierarchical extraction rules;

[0075] S5, conduct method analysis based on the DEMATEL method and the AISM model, and give corresponding suggestions; wherein, S5 includes: analyzing the results, levels, and the entire series and loops of cause and effect, and finally giving suggestions based on the results.

[0076] As a preferred embodiment, the S1 includes:

[0077] S11, determining the first-level indicators, including: confirming the factors affecting the operator's safe operation through a large amount of literature research, the factors affecting the operator's safe operation include one or more of human factors, train factors, track environment, driving behavior, external factors and social factors; based on the factors affecting the operator's safe operation, confirming four first-level indicators affecting the operator's safe operation, the four first-level indicators include human factors, train factors, external factors and social factors;

[0078] S12, determining the secondary indicators, including: determining the secondary indicators corresponding to the four primary indicators; wherein:

[0079] For the first-level indicator human factors, the second-level indicators include: fatigue level, operator's emotional state and psychological factors, physiological state, bad operating behavior, operator's attitude and sense of responsibility, operator's decision-making ability and cognitive ability, information and communication, and safety awareness; (8 items)

[0080] For the first-level indicator train factors, the second-level indicators include: technical condition of the train, mechanical failure and equipment damage, new safety technology and innovative applications, and intelligent operation assistance systems; (4)

[0081] For the first-level indicator external factors, the second-level indicators include: animals or pedestrians and other trains; (3)

[0082] For the first-level indicator social factors, the second-level indicators include: social environment, cultural background, economic status, work environment, education level, and social relations and interpersonal communication; (8 items)

[0083] As a preferred embodiment, the S2 includes:

[0084] S21, based on the expert interview and questionnaire analysis method in the field of operation safety, the primary and secondary indicators are adjusted to establish a preliminary index system of factors affecting safe driving of locomotive crew members, and multiple influencing factor indicators are selected from the primary and secondary indicators; wherein the preliminary index system of factors affecting safe driving of locomotive crew members corresponds to the factors affecting safe operation of operators described in S11 and is classified into six major analysis indicators: human factors, train factors, track environment factors, driving behavior factors, social factors and external factors;

[0085] S22, from the preliminary analysis system of factors affecting safe driving of locomotive crew, the indicators with unclear analysis purpose and weak scientificity were deleted, the indicators with strong similarity were merged, and some indicators were modified to make them easier to understand and more operational. Finally, the analysis system of factors affecting safe driving of locomotive crew was obtained. The index system of factors affecting safe driving of locomotive crew includes 4 first-level indicators and 23 second-level indicators. The specific description of the indicators is shown in Table 1 below:

[0086] Table 1

[0087]

[0088] like Figure 2 As shown, as a preferred implementation, S3 constructs a DEMATEL method based on the index system; wherein, the construction basis of the DEMATEL method includes: comparing each element in the index system in pairs, calculating the influence and influence of each element on other elements based on the logical relationship and direct influence matrix between the elements in the index system, thereby calculating the cause degree and centrality of each element and determining the causal relationship between the elements and the status of each element in the index system; wherein, each element is a risk factor, including:

[0089] S31, based on risk factor F i The degree of influence on the risk factor Fj is encoded as the direct influence matrix O ij (This embodiment is O 23×23 ), the specific encoding method is O ij =Fi F j The degree of influence is stipulated that when i = j, O ij =0, where i=1,2,3,…,n; j=1,2,3,…,n;

[0090] S32, normalize the direct influence matrix O to obtain the standard influence matrix M. Normalization can eliminate the dimension effect and reduce the numerical range. Here, the row and maximum value method is used, as shown in formula (1);

[0091]

[0092] S33, calculate the comprehensive influence matrix T(t ij ), where I represents the identity matrix;

[0093] T(t ij )=M(IM) -1 (2);

[0094] S34, based on the comprehensive influence matrix T(t ij ) Calculate the influence D i 、Influence C i 、Center degree M i R i , which are the measurement criteria for the influence of the four measurement factors in the method, and the formulas are shown in (3)-(6) below, where i = 1, 2, 3, ..., n; j = 1, 2, 3, ..., n; t (ij) is the element of the comprehensive influence matrix T;

[0095]

[0096] R i =D i -C i (6).

[0097] AISM (Adversarial Interpretative Structural Model) is derived from the game interpretation structural model. On the basis of the hierarchical extraction rule of ISM (Interpretative Structural Model) with priority on results, the opposing hierarchical extraction rule of priority on causes is added to establish a set of adversarial hierarchical topological graphs. This method of explaining the order of factors (causality) in the form of paired topological hierarchical graphs is called the adversarial interpretation structural model.

[0098] As a preferred implementation, S4 constructs an AISM model based on the index system; wherein, the construction basis of the AISM model includes: obtaining a relationship matrix based on a decision matrix composed of the causality and centrality of each element, integrating the game confrontation (Adversarial) idea in the generative adversarial network (GAN) on the basis of the classic explanatory structural model (ISM), and obtaining a pair of simplest hierarchical topological graphs through opposite hierarchical extraction rules, including:

[0099] S41, based on the cause degree and centrality of each element, a decision matrix is ​​formed, a threshold λ (λ∈[0,1]) is introduced, and a relationship matrix A is obtained according to the partial order rule in the following formula (7); wherein the relationship matrix A is composed of a ij composition:

[0100]

[0101] S42, to further consider the influence of the factors themselves, the overall relationship matrix B is calculated according to the following formula (8); where I represents the unit matrix;

[0102] B = A + I (8);

[0103] S43, performing multiplication on the overall relationship matrix B based on the following formula (9) to calculate the reachable matrix R;

[0104] R=B k+1 =B k ≠B k-1 (9);

[0105] Where k refers to the kth power of the overall relationship matrix B;

[0106] S44, based on the extraction rules of the result priority principle (UP type) and the cause priority principle (DOWN type), the reachability matrix R is divided into levels and regions, and the skeleton matrix S is calculated based on formula (10);

[0107] S=R-(RI) 2 -I (10);

[0108] S45, drawing an adversarial multi-level hierarchical explanatory structural model, and then performing analysis based on the adversarial multi-level hierarchical explanatory structural model.

[0109] As a preferred implementation, S5 analyzes the results, levels, and the entire series and loops of cause and effect, and finally gives suggestions based on the results, including:

[0110] S51, analyzing the results, including cause degree analysis, centrality analysis, influence degree analysis and impact degree analysis; wherein:

[0111] (1) Cause analysis

[0112] Causality indicates the degree of influence of one factor on other factors; it reflects the contribution of a factor to the method as a whole, that is, whether this factor is the main reason for changes in other factors or is passively affected by other factors; factors with higher causality have greater influence on the method and are usually considered to be the key factors in the method.

[0113] (2) Centrality analysis

[0114] Centrality indicates the position of a factor in the entire causal network, that is, its position and importance in the causal relationship. The centrality of a factor depends on the number and importance of other factors it is connected to. Factors with high centrality play a core role in the causal network and may be a bridge or key node connecting other factors.

[0115] (3) Impact analysis

[0116] The influence degree indicates the degree of influence of a factor on other factors. In DEMATEL, the influence degree between factors is determined by quantifying the degree of association between influencing factors. In DEMATEL, the influence degree can help determine the causal relationship between factors and their importance in the entire influencing factor network.

[0117] (4) Impact analysis

[0118] The degree of influence refers to the degree to which a factor is affected by other factors. It is a supplement to the degree of influence and helps determine the degree to which a factor is affected by other factors. By analyzing the degree of influence, we can determine which factors are more affected by other factors, so as to consider the role of these factors more comprehensively in the decision-making process and make decisions and analysis more effectively.

[0119] S52, analysis of the entire hierarchy and causal chain, including root cause analysis, transition cause analysis and proximate cause analysis; among which:

[0120] (1) Root cause analysis

[0121] The fundamental causal level is the union of the lowest factors in the confrontation hierarchy diagram. The essential causal level factors only send out directed arrows, but do not receive directed arrows, that is, they only affect other factors but are not affected by other factors. The fundamental causal level factors are the most fundamental reasons that affect the operator's safe operation.

[0122] (2) Analysis of the causes of transition

[0123] The transitional causal level, that is, the factors located in the middle layer of the confrontational hierarchy diagram after removing the essential causal level and the adjacent causal level factors, both receive and send directed arrows, and can be affected by other factors and affect other factors. The transitional causal level factors not only play the role of propagating (transitional) influence in the method, but also can become the source of influence and affect other factors. Therefore, it is very necessary to control the transitional causal level factors, especially the factors with high centrality.

[0124] (3) Proximal cause analysis

[0125] The adjacent causal level is the union of the top factors in the confrontation hierarchy diagram. The adjacent causal level factors only receive directed arrows but do not send out directed arrows, that is, they are only affected by other factors but do not affect other factors. This shows that the adjacent causal level factors are the most direct factors affecting the operator's safe operation. However, the adjacent causal level factors are "vulnerable to the influence of other factors. Therefore, when controlling the adjacent causal level factors, we should also pay attention to the control of their antecedent factors, or cut off their connection with the antecedent factors.

[0126] S53, analyzing the loop, wherein the loop is a strong connection between the influencing factors;

[0127] Loops are bidirectional arrows between influencing factors, indicating strong connections between these factors and causal relationships between them. The existence of these loops indicates that there are complex interactions between the various factors in the method, which may affect the stability and safety of the entire method. Managers need to carefully analyze and monitor these loops so that timely measures can be taken to prevent the occurrence of adverse dynamic feedback effects, thereby ensuring safe operation of locomotive crews.

[0128] S54, based on the analysis of S51-S53, the government, enterprises and all sectors of society need to work together to take comprehensive and systematic measures to promote the safe operation of locomotive crew members from multiple aspects such as laws and regulations, education and training, social environment, weather conditions, and management systems. At the same time, we should pay attention to factors with high centrality, such as the decision-making and cognitive abilities of operators, information and communication, attitudes and sense of responsibility of operators, etc., strengthen relevant training and management, and improve the overall level of safe operation.

[0129] Embodiment 2

[0130] This embodiment provides a method for analyzing factors affecting safe operation of locomotive crew based on DEMATEL-AISM, including:

[0131] An influencing factor index selection module is used to select influencing factor indexes; wherein, the selection of influencing factor indexes includes: preliminarily establishing an influencing factor framework through a large amount of literature research, wherein the influencing factor framework covers one or more of human factors, train factors, track environment, driving behavior, external and social factors, wherein different influencing factors will have different degrees of influence;

[0132] A locomotive crew safe driving influencing factor index system construction module is used to identify influencing factors based on the selected influencing factor indexes and construct a locomotive crew safe driving influencing factor index system; the locomotive crew safe driving influencing factor index system construction module adjusts the selected influencing factor index based on interviews with experts in the field of operating safety and / or questionnaire survey analysis, thereby obtaining a locomotive crew safe driving influencing factor index system;

[0133] A DEMATEL method construction module is used to construct a DEMATEL method based on the indicator system; wherein, the construction basis of the DEMATEL method includes: comparing each element in the indicator system in pairs, calculating the influence and the influence of each element on other elements based on the logical relationship and direct influence matrix between the elements in the indicator system, thereby calculating the cause degree and centrality of each element and determining the causal relationship between the elements and the status of each element in the indicator system; wherein, each element is a risk factor;

[0134] An AISM model construction module is used to construct an AISM model based on the indicator system; wherein, the construction basis of the AISM model includes: obtaining a relationship matrix based on a decision matrix composed of the causality and centrality of each element, integrating the adversarial idea of ​​the game in the generative adversarial network (GAN) on the basis of the classic explanatory structural model (ISM), and obtaining a pair of simplest hierarchical topological graphs through opposite hierarchical extraction rules;

[0135] The influencing factor analysis module is used to perform method analysis based on the DEMATEL method and the AISM model and give corresponding suggestions; wherein the influencing factor analysis module analyzes the results, levels, and the entire series and loops of cause and effect, and finally gives suggestions based on the results.

[0136] The present invention also provides a memory storing a plurality of instructions, wherein the instructions are used to implement the method as in the first embodiment.

[0137] like Figure 3As shown, the present invention also provides an electronic device, including a processor 301 and a memory 302 connected to the processor 301, the memory 302 stores a plurality of instructions, and the instructions can be loaded and executed by the processor so that the processor can execute the method as in the first embodiment.

[0138] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for analyzing factors affecting safe operation of locomotive crew based on DEMATEL-AISM, characterized in that: include: S1, selecting influencing factor indicators, wherein the selecting influencing factor indicators includes: preliminarily establishing an influencing factor framework through literature research, wherein the influencing factor framework includes one or more of human factors, train factors, track environment, driving behavior, external and social factors, wherein different influencing factors will have different degrees of impact; S2, identifying influencing factors based on the selected influencing factor indicators, and constructing an influencing factor indicator system for safe driving of locomotive crew members; S2 includes: adjusting the selected influencing factor indicators based on interviews with experts in the field of operating safety and / or questionnaire survey analysis, so as to obtain an influencing factor indicator system for safe driving of locomotive crew members; S3, constructing a DEMATEL method based on the indicator system; wherein, the construction basis of the DEMATEL method includes: comparing each element in the indicator system in pairs, calculating the influence and influence of each element on other elements based on the logical relationship and direct influence matrix between the elements in the indicator system, thereby calculating the cause degree and centrality of each element and determining the causal relationship between the elements and the position of each element in the indicator system; wherein, each element is a risk factor; S4, constructing an AISM model based on the indicator system; wherein the construction basis of the AISM model includes: obtaining a relationship matrix based on a decision matrix composed of the causality and centrality of each element, integrating the game confrontation in the generative adversarial network GAN on the basis of the classic explanatory structure model ISM, and obtaining a pair of simplest hierarchical topological graphs through opposite hierarchical extraction rules; S5, conduct method analysis based on the DEMATEL method and the AISM model, and give corresponding suggestions; wherein, S5 includes: analyzing the results, levels, and the entire series and loops of cause and effect, and finally giving suggestions based on the results.

2. The method for analyzing factors affecting safe operation of locomotive crew based on DEMATEL-AISM according to claim 1, characterized in that: The S1 includes: S11, determining the primary index, including: confirming the factors affecting the operator's safe operation through literature research, the factors affecting the operator's safe operation include one or more of human factors, train factors, track environment, driving behavior, external factors and social factors; based on the factors affecting the operator's safe operation, confirming four primary indexes affecting the operator's safe operation, the four primary indexes include human factors, train factors, external factors and social factors; S12, determining the secondary indicators, including: determining the secondary indicators corresponding to the four primary indicators.

3. The method for analyzing factors affecting safe operation of locomotive crew based on DEMATEL-AISM according to claim 2, characterized in that: The secondary indicators of S12 include: For the first-level indicator human factors, the second-level indicators include: fatigue level, emotional state and psychological factors of the operator, physiological state, bad operating behavior, attitude and sense of responsibility of the operator, decision-making ability and cognitive ability of the operator, information and communication, and safety awareness; For the first-level indicator train factors, the second-level indicators include: technical condition of the train, mechanical failures and equipment damage, new safety technologies and innovative applications, and intelligent operation assistance systems; For the first-level indicator external factors, the second-level indicators include: animals or pedestrians and other trains; For the first-level indicator social factors, the second-level indicators include: social environment, cultural background, economic status, work environment, education level, and social relations and interpersonal communication.

4. The method for analyzing factors affecting safe operation of locomotive crew based on DEMATEL-AISM according to claim 3, characterized in that: The S2 includes: S21, based on the expert interview and questionnaire analysis method in the field of operation safety, the primary and secondary indicators are adjusted to establish a preliminary index system of factors affecting safe driving of locomotive crew members, and multiple influencing factor indicators are selected from the primary and secondary indicators; wherein the preliminary index system of factors affecting safe driving of locomotive crew members corresponds to the factors affecting safe operation of operators described in S11 and is classified into six major analysis indicators: human factors, train factors, track environment factors, driving behavior factors, social factors and external factors; S22, deleting some unqualified indicators from the preliminary locomotive crew safe driving influencing factor indicator system, combining similar indicators, and modifying some indicators.

5. The method for analyzing factors affecting safe operation of locomotive crew based on DEMATEL-AISM according to claim 4, characterized in that: The S3 includes: S31, based on risk factor F i The degree of influence on the risk factor Fj is encoded as the direct influence matrix O ij , the specific encoding method is O ij =F i F j The degree of influence is stipulated that when i = j, O ij =0, where i=1,2,3,…,n; j=1,2,3,…,n; S32, normalizing the direct influence matrix O based on formula (1) to obtain the standard influence matrix M; S33, calculate the comprehensive influence matrix T(t ij ), where I represents the identity matrix; T(t ij )=M(I-M) -1 (2); S34, based on the comprehensive influence matrix T(t ij ) Calculate the influence D i 、Influence C i 、Center degree M i R i , which are the measurement criteria for the influence of the four measurement factors in the method, and the formulas are shown in (3)-(6) below, where i = 1, 2, 3, ..., n; j = 1, 2, 3, ..., n0; t(ij) is the comprehensive influence matrix T(t ij ) elements; M i =D i +C i (5); R i =D i -C i (6)。 6. The method for analyzing factors affecting safe operation of locomotive crew based on DEMATEL-AISM according to claim 5, characterized in that: The S4 includes: S41, based on the cause degree and centrality of each element, a decision matrix is ​​formed, a threshold λ (λ∈[0,1]) is introduced, and a relationship matrix A is obtained according to the partial order rule in the following formula (7); wherein the relationship matrix A is composed of a ij composition: S42, to further consider the influence of the factors themselves, the overall relationship matrix B is calculated according to the following formula (8); where I represents the unit matrix; B = A + I (8); S43, performing multiplication on the overall relationship matrix B based on the following formula (9) to calculate the reachable matrix R; R=B k+1 =B k ≠B k-1 (9); Where k refers to the kth power of the overall relationship matrix B; S44, based on the extraction rules of the result priority principle and the cause priority principle, the reachability matrix R is divided into levels and regions, and the skeleton matrix S is calculated based on formula (10); S=R-(R-I) 2 -I (10); S45, drawing an adversarial multi-level hierarchical explanatory structural model, and then performing analysis based on the adversarial multi-level hierarchical explanatory structural model.

7. The method for analyzing factors affecting safe operation of locomotive crew based on DEMATEL-AISM according to claim 6, characterized in that: The S5 analyzes the results, levels, and entire series and loops of cause and effect, and ultimately provides recommendations based on the results, including: S51, analyzing the results, including causal degree analysis, centrality analysis, influence analysis and influenced degree analysis; wherein: the causal degree analysis is used to reflect the contribution of a factor to the method as a whole. The higher the causal degree of a factor, the greater its influence on the method, and it is considered to be a key factor in the method; the centrality analysis indicates the position of a factor in the entire causal network, that is, its position and importance in the causal relationship; the influence analysis includes analyzing the influence of a certain factor on other factors; in DEMATEL, the influence between the factors is determined by quantifying the correlation between the influencing factors; the influenced degree analysis is used to analyze the degree to which a certain factor is affected by other factors, so as to determine the factors that are greatly affected by other factors; S52, analyze the whole series of levels and causalities, including root cause analysis, transition cause analysis and adjacent cause analysis; wherein: the root cause level in the root cause analysis is the union of the lowest factors in the confrontation hierarchy diagram, and the essential cause level factors only send out directed arrows, but do not receive directed arrows, that is, they only affect other factors but are not affected by other factors; the transition cause analysis includes factors located in the middle layer of the confrontation hierarchy diagram after removing the essential cause level and adjacent cause level factors, which both receive directed arrows and send out directed arrows, and can be affected by other factors and affect other factors; the adjacent cause level in the adjacent cause analysis is the union of the uppermost factors in the confrontation hierarchy diagram, and the adjacent cause level factors only receive directed arrows, but do not send out directed arrows, that is, they are only affected by other factors but do not affect other factors, and the adjacent cause level factors are the most direct factors affecting the operator's safe operation; S53, analyzing the loop, wherein the loop is a strong connection between the influencing factors; S54, based on the analysis of S51-S53, take comprehensive and systematic measures to promote the safe operation of locomotive crew members and improve the overall safety operation level.

8. A method for analyzing factors affecting safe operation of locomotive crew based on DEMATEL-AISM, used to implement the analysis method according to any one of claims 1 to 7, characterized in that: include: An influencing factor index selection module is used to select influencing factor indexes; wherein, the selection of influencing factor indexes includes: preliminarily establishing an influencing factor framework through a large amount of literature research, wherein the influencing factor framework covers one or more of human factors, train factors, track environment, driving behavior, external and social factors, wherein different influencing factors will have different degrees of influence; A locomotive crew safe driving influencing factor index system construction module is used to identify influencing factors based on the selected influencing factor indexes and construct a locomotive crew safe driving influencing factor index system; the locomotive crew safe driving influencing factor index system construction module adjusts the selected influencing factor index based on interviews with experts in the field of operating safety and / or questionnaire survey analysis, thereby obtaining a locomotive crew safe driving influencing factor index system; A DEMATEL method construction module is used to construct a DEMATEL method based on the indicator system; wherein, the construction basis of the DEMATEL method includes: comparing each element in the indicator system in pairs, calculating the influence and the influence of each element on other elements based on the logical relationship and direct influence matrix between the elements in the indicator system, thereby calculating the cause degree and centrality of each element and determining the causal relationship between the elements and the status of each element in the indicator system; wherein, each element is a risk factor; An AISM model construction module is used to construct an AISM model based on the indicator system; wherein, the construction basis of the AISM model includes: obtaining a relationship matrix based on a decision matrix composed of the causal degree and centrality of each element, integrating the game confrontation idea in the generative adversarial network GAN on the basis of the classic explanatory structure model ISM, and obtaining a pair of simplest hierarchical topological graphs through opposite hierarchical extraction rules; The influencing factor analysis module is used to perform method analysis based on the DEMATEL method and the AISM model and give corresponding suggestions; wherein the influencing factor analysis module analyzes the results, levels, and the entire series and loops of cause and effect, and finally gives suggestions based on the results.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a plurality of instructions, and the processor is used to read the instructions and execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, and the plurality of instructions can be read by a processor to execute the method according to any one of claims 1 to 7.

Images