Psychological risk prediction method, system, equipment and medium

Abstract

The invention relates to a psychological risk prediction method, system and device and a medium, and the psychological risk prediction method comprises the steps: obtaining behavior evidence of related survey reports of human factors in an aviation accident; constructing a psychological risk system of the aviation accident based on psychological risk factors corresponding to the behavior evidence; determining a danger level standard of the pilot in the psychological risk factors in the analysis sample; screening aviation accident cases meeting specific conditions as quantitative analysis samples, and obtaining accident psychological risk degree data according to the danger degree grade standard of psychological risk factors; extracting a first psychological risk factor weight based on partial least square discriminant analysis; outputting a second psychological risk factor weight according to a Lasso regularized logistic regression model; adopting an AHP analytic hierarchy process to obtain a third psychological risk factor weight; and obtaining the final weight of the psychological risk factor by adopting a range maximization method. Accurate prediction of accident psychological risk factors is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of processing psychological risk analysis. In particular, it relates to a method, system, device and medium for predicting psychological risks. Background Art

[0002] With the rapid development of technology, the causal structure of various accidents has changed significantly. Among them, human psychological risk factors play an important role. Psychological risk factors are the deep reasons for various human errors and accidents, and it is particularly important to analyze the psychological risk factors in various accidents. By reviewing accident reports, the identification and analysis of psychological risk factors can be achieved, and multiple psychological indicators and characteristics can be integrated for early warning of the psychological state of personnel in important positions.

[0003] Currently, the development of related technologies can include the following two aspects: (1) The quantitative analysis methods for the psychology of relevant staff in each professional field mainly focus on evaluating the psychological state of personnel through methods such as questionnaires, behavioral observations, and physiological monitoring. For example, standardized psychological assessment tools (such as the Self-Rating Anxiety Scale) are used to quantify the emotional state of personnel. However, these methods often cannot reflect the actual psychological state of people during the work process and may be affected by personal subjective evaluations. (2) The quantitative analysis of accident reports in each field is mainly based on the statistics of the occurrence frequency of a certain factor, and methods such as regression analysis and principal component analysis are carried out. However, these analysis techniques lack focus on the psychological level. Therefore, how to invent an accurate prediction technology for appropriate accident psychological risk factors is an urgent problem to be solved. Summary of the Invention

[0004] The present invention provides a method, system, device and medium for predicting psychological risks to solve the problem that the existing prediction technology is not applicable to the prediction of accident psychological risk factors.

[0005] To achieve the above object, in a first aspect, the present invention relates to a method for predicting psychological risks, including:

[0006] Obtaining the behaviors that map the psychological risks of pilots in the relevant investigation reports of human factors in aviation accidents as behavioral evidence;

[0007] Extracting and inductively defining multiple psychological risk factors corresponding to the behavioral evidence of the aviation accidents, and constructing a psychological risk system for the aviation accidents, where the psychological risk system includes psychological risk factors, the risk levels of the psychological risk factors, and the types and occurrence frequencies of behavioral evidence corresponding to the risk levels of the psychological risk factors;

[0008] Determine the danger level standard of the psychological risk factors of the pilots in the analysis sample based on the number and occurrence frequency of the behavioral evidences shown by the pilots in the aviation accidents under each of the psychological risk factors;

[0009] Screen aviation accident cases that meet specific conditions as the quantitative analysis sample, score the danger level of the psychological risk factors of the quantitative analysis sample according to the danger level standard of the psychological risk factors, and obtain the accident psychological risk degree data;

[0010] Extract the first psychological risk factor weights based on partial least squares discriminant analysis, use R2X, R2Y and Q2Y to evaluate the model fitting effect, and use the VIP value as the first psychological risk factor weights of each of the psychological risk factors; output the second psychological risk factor weights of each of the psychological risk factors according to the constructed Lasso-regularized logistic regression model; adopt the AHP (Analytic Hierarchy Process) to obtain the third psychological risk factor weights of each of the psychological risk factors;

[0011] Normalize the first psychological risk factor weights, the second psychological risk factor weights and the third psychological risk factor weights of the psychological risk factors, and use the range maximization method to obtain the final weights of the psychological risk factors.

[0012] To achieve the above object, in a second aspect, the present invention relates to a psychological risk prediction system, including:

[0013] A behavioral evidence acquisition module, configured to obtain the behaviors that map the psychological risks of the pilots in the relevant investigation reports of the human factors in the aviation accidents as behavioral evidences;

[0014] A risk system construction module, configured to extract and summarize and define multiple psychological risk factors corresponding to the behavioral evidences of the aviation accidents, and construct the psychological risk system of the aviation accidents, wherein the psychological risk system includes psychological risk factors, the danger levels of the psychological risk factors, and the types and occurrence frequencies of the behavioral evidences corresponding to the danger levels of the psychological risk factors;

[0015] A danger level determination module, configured to determine the danger level standard of the psychological risk factors of the pilots in the analysis sample based on the number and occurrence frequency of the behavioral evidences shown by the pilots in the aviation accidents under each of the psychological risk factors;

[0016] A quantitative analysis scoring module, configured to screen aviation accident cases that meet specific conditions as the quantitative analysis sample, score the danger level of the psychological risk factors of the quantitative analysis sample according to the danger level standard of the psychological risk factors, and obtain the accident psychological risk degree data;

[0017] A psychological risk factor weight acquisition module, which is used to extract the first psychological risk factor weights based on partial least squares discriminant analysis, evaluate the model fitting effect using R2X, R2Y and Q2Y, and use the VIP value as the first psychological risk factor weights of each of the psychological risk factors; divide the data into a training set and a test set to construct a logistic regression model with Lasso regularization, and output the second psychological risk factor weights of each of the psychological risk factors according to the logistic regression model; adopt the AHP (Analytic Hierarchy Process) to obtain the third psychological risk factor weights of each of the psychological risk factors;

[0018] A final weight acquisition module, which is used to normalize the first psychological risk factor weights, the second psychological risk factor weights and the third psychological risk factor weights of the psychological risk factors, and use the range maximization method to obtain the final weights of the psychological risk factors.

[0019] To achieve the above object, in a third aspect, the present invention further relates to an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the above-mentioned psychological risk prediction method is implemented.

[0020] To achieve the above object, in a fourth aspect, the present invention further relates to a computer-readable storage medium, in which instructions are stored, and when the instructions run, the above-mentioned psychological risk prediction method is executed.

[0021] A psychological risk prediction method, system, device and medium provided by the present invention have the following beneficial effects compared with the prior art:

[0022] The method includes extracting psychological hazard source indicators and corresponding behavioral manifestations through accident case analysis, and using the range maximization method to comprehensively combine the results of the AHP (Analytic Hierarchy Process), partial least squares discriminant analysis (Partial Least Squares Discriminant Analysis, PLS-DA) method and the logistic regression analysis method with Lasso regularization to determine the index weights of psychological risk factors and construct a psychological risk factor analysis model. In this solution, based on the in-depth analysis of accident cases and considering the data characteristics, two types of subjective and objective method orientations are comprehensively combined to assign weights to psychological risk factors, which can not only anchor the core psychological hazard source indicators but also reduce the weight fluctuations caused by data changes.

[0023] The present invention effectively combines these three methods to comprehensively evaluate the weights of multiple psychological risk factors in accidents, thereby more comprehensively and accurately reflecting the specific impacts of various psychological risk factors on safety, and providing a solid theoretical basis for accident prevention. In summary, when evaluating the psychological risk factors in accidents, the method of the present invention combines the advantages of the existing technologies, constructs a scientific and comprehensive weight assignment system, and provides reliable data support for the risk assessment and safety management of accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 is a flowchart of a psychological risk prediction method in Embodiment 1 of the present invention;

[0025] Figure 2 is a schematic structural diagram of a psychological risk prediction system in Embodiment 2 of the present invention;

[0026] Figure 3 is a schematic structural diagram of an electronic device in Embodiment 3 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] The present invention will be further described in detail below with reference to the drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the present invention, rather than limiting the present invention. Additionally, it should be noted that for the sake of description, only the parts related to the present invention rather than all the structures are shown in the drawings.

[0028] Embodiment 1

[0029] A psychological risk prediction method. Please refer to Figure 1 , the psychological risk prediction of an aviation accident is implemented by an electronic device with a central processing unit, which can be a personal computer, a smart terminal, a local area network, a server, etc. As Figure 1 shown, it includes the following steps: S101 to S106.

[0030] S101 Obtain the behavioral evidence reflecting the psychological risks of pilots in the relevant investigation reports on human factors in aviation accidents.

[0031] Before obtaining the behaviors reflecting the psychological risks of pilots in the relevant investigation reports on human factors in aviation accidents as behavioral evidence, it further includes: in order to facilitate the study of the mapping between individual behaviors and psychological characteristics, select the relevant investigation reports on human factors in aviation accidents as analysis samples to extract the behaviors reflecting the psychological risks of pilots as behavioral evidence;

[0032] Among them, the process of extracting behavioral evidence can be a manual analysis step, can also be extracted through a machine learning model, or can be completed through the cooperation of a machine learning model and manual assistance.

[0033] Specifically, compare the behaviors of the pilots in the analysis samples with the standardized operations, and refer to non-standardized scenarios to extract the behaviors that can reflect the psychological risks of the pilots in the accident as behavioral evidence. Compare the behaviors of individuals in each stage (such as the planning and preparation stage, the task execution stage, the reporting and feedback stage, etc.) in the relevant investigation reports of aviation accidents with the standardized operations, determine the deviations in the individual's behavioral operations, and extract the behavioral manifestations that can reflect their psychological risks in the accident to form a list of behavioral evidence. For example, a certain person did not pay attention to the surrounding environment conditions displayed in the system according to the operation standards during the planning and preparation stage, lacking the analysis and judgment of environmental factors, resulting in difficulty in effectively coping with sudden situations caused by the environment when encountered, and then inducing an accident. This reflects the deviation of the individual's planning and preparation behavior.

[0034] S102 Extract, summarize, and define the psychological risk factors corresponding to the behavioral evidence of multiple aviation accidents, and construct a psychological risk system for aviation accidents. Among them, the psychological risk system includes psychological risk factors, the risk levels of psychological risk factors, and the types and occurrence frequencies of behavioral evidence corresponding to the risk levels of psychological risk factors.

[0035] In this embodiment, first determine the psychological risk factors that induce the behavioral evidence. Specifically: Analyze all possible individual reasons for the generation of the behavioral evidence, speculate on the corresponding psychological risk factors based on the individual reasons, and discuss the association between the extracted behavioral evidence and the corresponding psychological risk factors by the expert group to summarize and define the psychological risk factors and the behavioral evidence corresponding to the psychological risk factors.

[0036] S102 may specifically include S121 to S123.

[0037] S121 Determine the psychological risk factors that induce the behavioral evidence. Specifically: Based on deductive reasoning, speculate on all possible psychological risk factors that induce the individual's behavioral evidence.

[0038] Specifically, the steps are as follows:

[0039] S1211 Identify risk factors based on behavioral evidence: For example, an experienced person did not check each preparation work (equipment, system, materials, personnel, etc.) item by item according to the checklist before the task execution, violating the standard operating procedure, which is an unsafe operating behavior.

[0040] S1212 Analyze all possible individual reasons for the behavior: This person may fail to do a good job in the preparation work due to factors such as time pressure, disregard for rules, or excessive reliance on personal experience.

[0041] S1213 Speculate on the corresponding psychological risk factors: This person may have psychological risk factors such as insufficient risk assessment, weak discipline awareness, and overconfidence, resulting in the failure to complete the inspection task of the plan checklist as required.

[0042] For each aviation accident case, at least two experts conduct back-to-back analyses, and hold group discussions on the points of disagreement in each case until a consensus is reached.

[0043] S123 A psychological risk factor system and corresponding behavioral evidence are formed by summarizing the analysis results of different cases, as shown in Table 1 below.

[0044] The specific steps are as follows: S1231 to S1235.

[0045] S1231 Organize all the psychological risk factors and corresponding behavioral manifestations that appear in different cases.

[0046] S1232 Combine similar psychological risk factors in different cases.

[0047] S1333 Extract a general expression of behavioral evidence based on the common characteristics of aviation accident scenarios, and combine and determine similar expressions.

[0048] S1234 Organize the combined psychological risk factors and corresponding behavioral manifestations, and construct a framework for the psychological risk system.

[0049] S1235 Apply the summarized framework of the psychological risk system to the analysis of new cases to test its applicability.

[0050] The following is an example to illustrate the psychological risk factors and corresponding behavioral evidence:

[0051] Psychological risk factor: Weak sense of responsibility. The corresponding behavioral evidence is:

[0052] Behavioral evidence 1: Shirking responsibility - often trying to shift the blame to others / equipment / systems when problems occur;

[0053] Behavioral evidence 2: Not taking the initiative to solve problems - not taking timely action when problems are found (such as equipment warnings);

[0054] Behavioral evidence 3: Ignoring bad work habits, having the same type of operation mistakes more than once, but never trying to improve them.

[0055] S103; Based on the number and frequency of behavioral evidence shown by pilots in each psychological risk factor in aviation accidents, determine the risk level criteria for pilots in the analysis sample in terms of psychological risk factors.

[0056] Based on the number of behavioral evidence and the performance frequency / cycle shown by an individual under each psychological index, define the degree of danger of a certain psychological risk factor of the individual, which is divided into four levels: no abnormality, mild, moderate, and severe.

[0057] If the psychological risk factor i includes four behavioral manifestations: i-1, i-2, i-3, and i-4, the risk level of this psychological risk factor for individual W is defined as follows:

[0058]

[0059] Table 1

[0060] S104: Screen aviation accident cases that meet specific conditions as quantitative analysis samples, and score the risk level of the psychological risk factors in the quantitative analysis samples according to the risk level standard of psychological risk factors to obtain accident psychological risk level data.

[0061] Among them, screening aviation accident cases that meet specific conditions as quantitative analysis samples specifically includes:

[0062] Select more than or equal to 50 relevant investigation reports on human factors that meet the first screening standard as quantitative analysis samples. Among them, the first screening standard is: it is determined that the occurrence factors of the aviation accident include human factors and the relevant investigation reports of the aviation accident include pilot behavior descriptions. The aviation accident cases screened here can overlap with the aviation accident data in S101 and can be from the same sample.

[0063] In this embodiment, score the risk level of the psychological risk factors in the quantitative analysis samples. Among them, use the psychological risk factors as independent variables, and score the severity level of the psychological risk factors according to the risk level standard of the psychological risk factors in Table 1 above. The score range is 0 points, 1 point, 2 points, and 3 points: 0 points correspond to no abnormality, 1 point corresponds to mild, 2 points correspond to moderate, and 3 points correspond to severe. Use the accident severity level data as the response variable: mark them as 1, 2, and 3 according to unsafe events, serious symptoms, and accidents respectively.

[0064] S105: Extract the weight of the first psychological risk factor based on partial least squares discriminant analysis, use R2X, R2Y, and Q2Y to evaluate the fitting effect of the model, and use the VIP value as the weight of the first psychological risk factor for each psychological risk factor; output the weight of the second psychological risk factor for each psychological risk factor according to the constructed Lasso-regularized logistic regression model; use the AHP (Analytic Hierarchy Process) to obtain the weight of the third psychological risk factor for each psychological risk factor.

[0065] In this embodiment, in the process of obtaining the weight of the first psychological risk factor, use the VIP value as the corresponding weight of the weighted psychological risk factor, which specifically includes: screening and determining the weighted psychological risk factors with VIP value > 1, and using the VIP value as the corresponding weight of the weighted psychological risk factor. The specific process of obtaining the weight of the first psychological risk factor is: S1511 to S1515.

[0066] S1511: Extract the weights of psychological hazard sources indicators based on Partial Least Squares Discriminant Analysis.

[0067] S1512: Centralize and normalize the variance of the independent variables and response variables obtained through S103.

[0068] S1513: Based on the partial least squares method, construct a PLS-DA model to establish the relationship between independent variables and response variables.

[0069] S1514: Use R2X, R2Y, and Q2Y to evaluate the model fitting effect, and perform a permutation test to determine whether the model is overfitted.

[0070] S1515: Anchor the discriminative psychological indicators (VIP value > 1), and use the VIP value weight as the corresponding weight value of the indicator.

[0071] In this embodiment, before obtaining the weights of the second psychological risk factors, it is necessary to: divide the data into a training set and a test set to construct a logistic regression model with Lasso regularization. In this embodiment, obtaining the weights of the second psychological risk factors specifically includes: S1521 - S1524.

[0072] S1521: Data preprocessing: Standardize the data set obtained based on 3 to eliminate the influence of dimensions, and randomly divide the data set into a training set and a test set according to a ratio of 7:3 to ensure the generalization ability of the model on unseen data.

[0073] S1522: Model construction: Construct a logistic regression model with Lasso regularization (L1) on the training set, and perform feature selection by adding an L1 penalty term to the loss function.

[0074] S1523: Model evaluation: Use the test set to predict the constructed model, and calculate indicators such as prediction accuracy, sensitivity, and specificity; use confusion matrices, ROC curves, AUC values, etc. to evaluate the model performance to ensure the prediction ability of the model.

[0075] S1524: Extract important indicators: According to the model, output the weights of each psychological risk factor, and indicators with a weight of zero can be regarded as unimportant indicators.

[0076] In this embodiment, obtaining the weights of the third psychological risk factors specifically includes: S1531 - S1536.

[0077] S1531: Multiple experts evaluate whether the psychological risk factors obtained through the psychological risk system should be retained, and the evaluation criterion is the correlation with the occurrence of accidents;

[0078] S1532 constructs a comparison matrix to conduct pairwise comparisons of psychological risk factors.

[0079] Preferably, the 1-9 scale method is used.

[0080] The definitions of the scales are shown in Table 2 below:

[0081] Scale Meaning 1 Index i is as important as index j 3 Index i is slightly more important than index j 5 Index i is significantly more important than index j 7 Index i is strongly more important than index j 9 Index i is extremely more important than index j 2,4,6,8 The importance degree of index i compared to index j is between the above two adjacent scales Reciprocal If the scale of i compared to j is x, then the scale of j compared to i is 1 / x

[0082] An example of the judgment matrix in Table 2 is shown in Table 3 below:

[0083]

[0084] Table 3

[0085] S1533: Calculate the consistency ratio of each expert and the average consistency ratio of all experts; for the consistency ratio CR of each expert, if CR is less than 0.1, it is considered that the scoring of this expert is consistent. The calculation formula for the CR value is as follows: where CI = (λ max -n) / (n - 1), where λmax is the maximum eigenvalue of the judgment matrix; n is the order of the judgment matrix; RI is the random consistency index, which is a constant related to the order of the matrix and is obtained by looking up the table.

[0086] S1534: Assign weights to the experts who pass the consistency test. The weight calculation formula for each expert is as follows:

[0087] where wi is the weight of the i-th expert; CI i is the consistency index of the i-th expert; m is the total number of experts;

[0088] S1535: The average consistency ratio GCR of all experts is the mean of the CR of each expert. If GCR is less than 0.1, it is considered that the overall scoring is consistent.

[0089] The eigenvalue method is used to calculate the weights of the psychological risk factors of each expert, and the weight value of the third psychological risk factor is calculated where Wj is the comprehensive weight of the psychological risk factor j, wi is the weight of the i-th expert, Wij is the weight of the psychological risk factor j given by the i-th expert, and m is the total number of experts.

[0090] To ensure that the comprehensive sum of the comprehensive weights is 1, normalize W j as follows: W j ′ is the normalized weight value of the third psychological risk factor of the psychological risk factor j, and n is the total number of psychological risk factors.

[0091] S106: Normalize the first psychological risk factor weight, the second psychological risk factor weight, and the third psychological risk factor weight of the psychological risk factors, and use the range maximization method to obtain the final weight of the psychological risk factors.

[0092] In this embodiment, the final weight of the psychological risk factors is obtained by using the range maximization method, which specifically includes:

[0093] S161 Normalize the first psychological risk factor weight, the second psychological risk factor weight, and the third psychological risk factor weight of the psychological risk factors to obtain a weight matrix.

[0094] S162 Calculate the range of each psychological risk factor. For each psychological risk factor j (j = 1, 2,..., n), calculate the maximum weight Max j and the minimum weight Min j :

[0095]

[0096] where, is the weight of the first psychological risk factor, is the weight of the second psychological risk factor, is the weight of the third psychological risk factor;

[0097] Calculate the range of the maximum weight and the minimum weight:

[0098] [Range j = Max j - Min j

[0099] S163: Calculate the final weight of the psychological risk factors:

[0100]

[0101] where, (k j ) is:

[0102]

[0103] is the weight representing the weight source of the psychological risk factor.

[0104] In some embodiments, after S106, it further includes S107: Based on the obtained degree of danger of the psychological risk factors of the aviation accident and the final weights of the psychological risk factors, predict the degree of psychological risk of the pilot during aviation flight.

[0105] Specifically, from the final weights (W​final,j ) and the scores of each psychological risk factor obtained in S104 (i.e., the degree of danger of the psychological risk factor X i ), to obtain the total psychological risk degree (Y) during the pilot's navigation, that is: Y = W final,1 X 1 +……W final,n X n。

[0106] In summary, a psychological risk prediction method proposed by the present invention, through the identification of behavioral evidence in accidents, and based on logical reasoning to penetrate the psychological risk factors that induce accidents, makes up for the deficiency of the existing accident analysis in the insufficient in-depth and comprehensive analysis of psychological risk factors, and provides a theoretical basis for safety management from the perspective of psychology. At the same time, this solution comprehensively uses the AHP analytic hierarchy process, PLS-DA and L1-regularized logistic regression for combination to weight the psychological risk factors: First, considering the characteristics of the psychological risk factors, subjective analysis is incorporated to make the final model more discriminative; second, PLS-DA analysis is introduced. Compared with other regression models, this method is more suitable for the sample characteristics (small samples) and index characteristics (the dependent variable is a categorical variable) of this solution; third, L1-regularized logistic regression is adopted to avoid underfitting or overfitting of the model and improve the model performance; fourth, different methods are used to weight the psychological risk factors, and the reliability of the analysis method is ensured through the comparison and combination of the evaluation results of different methods, and the analysis quality of the psychological risk factors is improved.

[0107] Through the above method, not only can key psychological risk factors be accurately identified and quantified, but also the prediction ability of accident occurrence can be improved, the safety management and training of various industries can be improved, the development of decision support systems can be promoted, and the understanding of the importance of psychological risk factors can be enhanced at the industry level. These effects together contribute to improving the safety management level of various industries and reducing the accident rate.

[0108] Embodiment 2

[0109] A psychological risk prediction system for predicting the psychological risk of aviation accidents, which is implemented by the hardware of an electronic device with a central processing unit, and can be implemented by a personal computer, a smart terminal, a local area network, a server, etc. In this embodiment, please refer to Figure 2 , including a behavioral evidence acquisition module 61, a risk system construction module 62, a danger level determination module 63, a quantitative analysis scoring module 64, a psychological risk factor weight acquisition module 65, and a final weight acquisition module 66.

[0110] The behavioral evidence acquisition module 61 is used to obtain the behaviors that map the psychological risks of pilots in the relevant investigation reports of human factors in aviation accidents as behavioral evidence;

[0111] A risk system construction module 62 is used to extract and summarize the psychological risk factors corresponding to the behavioral evidence of multiple aviation accidents, and construct a psychological risk system for aviation accidents. Among them, the psychological risk system includes psychological risk factors, the risk levels of psychological risk factors, and the types and occurrence frequencies of behavioral evidence corresponding to the risk levels of psychological risk factors;

[0112] A risk level determination module 63 is used to determine the risk level criteria of the psychological risk factors of the pilots in the analysis sample based on the number and occurrence frequency of the behavioral evidence shown by the pilots in each psychological risk factor in the aviation accident;

[0113] A quantitative analysis scoring module 64 is used to screen aviation accident cases that meet specific conditions as quantitative analysis samples, and score the risk level of the psychological risk factors of the quantitative analysis samples according to the risk level criteria of the psychological risk factors to obtain accident psychological risk degree data;

[0114] A psychological risk factor weight acquisition module 65 is used to extract the first psychological risk factor weights based on partial least squares discriminant analysis, use R2X, R2Y, and Q2Y to evaluate the model fitting effect, and use the VIP value as the first psychological risk factor weight of each psychological risk factor; divide the data into a training set and a test set to construct a logistic regression model with Lasso regularization, and output the second psychological risk factor weight of each psychological risk factor according to the logistic regression model; use the AHP analytic hierarchy process to obtain the third psychological risk factor weight of each psychological risk factor;

[0115] A final weight acquisition module 66 is used to normalize the first psychological risk factor weights, the second psychological risk factor weights, and the third psychological risk factor weights of the psychological risk factors, and use the range maximization method to obtain the final weights of the psychological risk factors.

[0116] In some embodiments, screening aviation accident cases that meet specific conditions as quantitative analysis samples specifically includes:

[0117] Selecting no less than 50 relevant investigation reports on human factors that meet the first screening criteria as quantitative analysis samples, where the first screening criteria are: determining that the occurrence factors of the aviation accident include human factors and the relevant investigation reports of the aviation accident include pilot behavior descriptions.

[0118] In some embodiments, scoring the risk level of the psychological risk factors of the quantitative analysis samples according to the risk level criteria of the psychological risk factors to obtain accident psychological risk degree data specifically includes:

[0119] Score the severity level of psychological risk factors according to the risk level standard of psychological risk factors, where 0 points correspond to no abnormality, 1 point corresponds to mild, 2 points correspond to moderate, and 3 points correspond to severe.

[0120] In some embodiments, extract the behaviors in the accident that can reflect the pilot's psychological risk as behavioral evidence. Specifically: compare the pilot's behaviors in the analysis sample with the standardized operations to extract the behaviors in the accident that can reflect the pilot's psychological risk as behavioral evidence.

[0121] In some embodiments, use the VIP value as the corresponding weight of the weighted psychological risk factor, specifically including: screening and determining the weighted psychological risk factors with VIP value > 1, and using the VIP value as the corresponding weight of the weighted psychological risk factor.

[0122] In some embodiments, it further includes a risk prediction module 67 (not shown in the drawings), which is used to obtain the final weight of the psychological risk factor by using the range maximization method, and then based on the risk level of the psychological risk factor of the obtained aviation accident and the final weight of the psychological risk factor, predict the psychological risk level of the pilot during aviation flight.

[0123] In some embodiments, use the range maximization method to obtain the final weight of the psychological risk factor, specifically including:

[0124] Normalize the first psychological risk factor weight value, the second psychological risk factor weight value, and the third psychological risk factor weight value of the psychological risk factor to obtain a weight matrix;

[0125] Calculate the range of each psychological risk factor. For each psychological risk factor j (j = 1, 2,..., n), calculate the maximum weight Max j and the minimum weight Min j :

[0126]

[0127] where, is the first psychological risk factor weight, is the second psychological risk factor weight, is the third psychological risk factor weight;

[0128] Calculate the range of the maximum weight and the minimum weight:

[0129] [Range j = Max j - Min j

[0130] Calculate the final weight of the psychological risk factor:

[0131] ​

[0132] Among them, (k j ) is:

[0133]

[0134] is the weight representing the weight source of the psychological risk factor.

[0135] In some instances, using the AHP (Analytic Hierarchy Process) to obtain the third psychological risk factor weight value of each psychological risk factor further includes:

[0136] Multiple experts evaluate whether the psychological risk factors obtained through the psychological risk system should be retained, and the evaluation criterion is the correlation with the occurrence of accidents;

[0137] Construct a comparison matrix to conduct pairwise comparisons of the psychological risk factors;

[0138] Calculate the consistency ratio of each expert and the average consistency ratio of all experts; for the consistency ratio CR of each expert, if CR is less than 0.1, it is considered that the scoring of this expert is consistent. The calculation formula for the CR value is as follows: Among them, CI is the consistency index, which examines the logical consistency of the expert when constructing the judgment matrix, aiming to ensure that the expert's judgment has no contradictions. CI = (λ max -n) / (n - 1), where λmax is the maximum eigenvalue of the judgment matrix; n is the order of the judgment matrix; RI is the random consistency index, which is a constant related to the order of the matrix and is obtained by looking up the table;

[0139] . Allocate weights to the experts passing the consistency test. The weight calculation formula for each expert is as follows:

[0140]

[0141] Among them, wi is the weight of the i-th expert; CI i is the consistency index of the i-th expert; m is the total number of experts;

[0142] The average consistency ratio GCR of all experts is the mean value of the CR of each expert. If GCR is less than 0.1, it is considered that the overall scoring is consistent;

[0143] Use the eigenvalue method to calculate the psychological risk factor weight of each expert and calculate the third psychological risk factor weight value Among them, Wj is the comprehensive weight of the psychological risk factor j, wi is the weight of the i-th expert, Wij is the weight of the psychological risk factor j given by the i-th expert, and m is the total number of experts.

[0144] A psychological risk prediction system according to this embodiment has the same implementation process, method, and effects as the psychological risk prediction method described in Embodiment 1, and will not be elaborated herein.

[0145] Embodiment 3

[0146] As Figure 3 shown, this embodiment relates to an electronic device including at least one processor and a memory communicatively connected to the at least one processor. Among them, the memory stores a computer program that can be run by the at least one processor. The computer program is executed by the at least one processor so that the at least one processor can execute a psychological risk prediction method of Embodiment 1 and achieve the corresponding beneficial effects of a psychological risk prediction method, which will not be elaborated herein. The electronic device provided in this embodiment can be a personal computer, such as a desktop computer, an all-in-one computer, a laptop computer, a tablet computer, etc., and can also be a terminal device such as a mobile phone, a wearable device, a handheld computer, etc. In this embodiment, the electronic device is a flight control computer. The electronic device is only an example and should not impose any limitations on the functions and usage scope of the embodiments of the present invention.

[0147] The components of the electronic device 3 may include, but are not limited to: the above-mentioned at least one processor 4, the above-mentioned at least one memory 5, and a bus 6 connecting different system components (including the memory 5 and the processor 4).

[0148] The bus 6 includes a data bus, an address bus, and a control bus.

[0149] The memory 5 may include volatile memory, such as a random access memory (RAM) 51 and / or a cache memory 52, and may further include a read-only memory (ROM) 53.

[0150] The memory 5 may further include a program / utility 55 having a set (at least one) of program modules 54. Such program modules 54 include, but are not limited to: an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include the implementation of a network environment.

[0151] The processor 4 executes various functional applications and data processing by running the computer program stored in the memory 5, such as the above-mentioned psychological risk prediction method.

[0152] The electronic device 3 can also communicate with one or more external devices 7 (such as a keyboard, a pointing device, etc.). Such communication can be performed through an input / output (I / O) interface 8. And, the electronic device 3 can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 9. As Figure 3As shown, network adapter 9 communicates with other modules of electronic device 3 via bus 6. It should be understood that although Figure 3 not shown in Figure 3 , other hardware and / or software modules can be used in conjunction with electronic device 3, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (redundant array of independent disks) systems, tape drives, and data backup storage systems, etc.

[0153] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present invention, the features and functions of two or more of the above-described units / modules can be embodied in one unit / modules. Conversely, the features and functions of one unit / modules described above can be further divided and embodied by multiple units / modules.

[0154] Embodiment 4

[0155] The present invention relates to a computer-readable storage medium storing instructions that, when executed, perform a psychological risk prediction method according to Embodiment 1. The implementation process, method, and effects during its execution are the same as those of the psychological risk prediction method described in Embodiment 1 and will not be elaborated here.

[0156] It should be noted that in this document, the term "comprising", "including" or any other variation thereof is intended to cover a non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the phrase "comprising a..." does not exclude the presence of additional identical elements in the process, method, article or device comprising such element.

[0157] The above are only the preferred embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structural or equivalent process transformation made using the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A psychological risk prediction method, characterized in that: include: Obtain the behaviors of pilots that reflect psychological risks in the relevant investigation reports on human factors in aviation accidents as behavioral evidence; Extracting, summarizing and defining a plurality of psychological risk factors corresponding to the behavioral evidence of the aviation accident, and constructing a psychological risk system for the aviation accident, wherein the psychological risk system includes psychological risk factors, risk levels of psychological risk factors, and types of behavioral evidence corresponding to the risk levels of the psychological risk factors and their occurrence frequencies; Determining the danger level standard of the pilots in the analysis sample under the psychological risk factor based on the number and frequency of the behavioral evidences shown by the pilots in the aviation accident under each of the psychological risk factors; Selecting aviation accident cases that meet specific conditions as quantitative analysis samples, scoring the danger level of the psychological risk factors of the quantitative analysis samples according to the danger level standard of the psychological risk factors, and obtaining accident psychological risk level data; The weight of the first psychological risk factor is extracted based on partial least squares discriminant analysis, and the model fitting effect is evaluated using R2X, R2Y and Q2Y, and the VIP value is used as the weight of the first psychological risk factor of each of the psychological risk factors; the second psychological risk factor weight of each of the psychological risk factors is output according to the constructed Lasso regularized logistic regression model; the AHP hierarchical analysis method is used to obtain the third psychological risk factor weight of each of the psychological risk factors; The first psychological risk factor weight, the second psychological risk factor weight and the third psychological risk factor weight of the psychological risk factor are normalized, and the final weight of the psychological risk factor is obtained by using the range maximization method.

2. A psychological risk prediction method according to claim 1, characterized in that: The screening of aviation accident cases that meet specific conditions as quantitative analysis samples specifically includes: More than or equal to 50 investigation reports on the human factors that meet the first screening criteria are selected as quantitative analysis samples, wherein the first screening criteria is: determining that the factors causing the aviation accident include human factors and that the investigation reports on the aviation accident include pilot behavior descriptions.

3. A psychological risk prediction method according to claim 2, characterized in that: The step of scoring the danger level of the psychological risk factors of the quantitative analysis sample according to the danger level standard of the psychological risk factors to obtain accident psychological risk level data specifically includes: The severity level of the psychological risk factors is scored according to the danger level standard of the psychological risk factors to obtain the accident psychological risk level data, wherein 0 points corresponds to no abnormality, 1 point corresponds to mild, 2 points corresponds to moderate and 3 points corresponds to severe.

4. A psychological risk prediction method according to claim 1, characterized in that: The extracting the behavior in the accident that can map the psychological risk of the pilot as the behavioral evidence specifically comprises: comparing the behavior of the pilot in the analysis sample with the standardized operation to extract the behavior in the accident that can map the psychological risk of the pilot as the behavioral evidence; and / or The using the VIP value as the corresponding weight of the weighted psychological risk factor specifically includes: screening and determining the weighted psychological risk factors with a VIP value greater than 1, and using the VIP value as the corresponding weight of the weighted psychological risk factor.

5. A psychological risk prediction method according to claim 1, characterized in that: After the final weight of the psychological risk factor is obtained by the range maximization method, the method further includes: predicting the overall psychological risk level of the pilot in aviation flight based on the obtained dangerousness level of the psychological risk factor of the aviation accident and the final weight of the psychological risk factor.

6. A psychological risk prediction method according to claim 1, characterized in that: The final weight of the psychological risk factor is obtained by using the range maximization method, which specifically includes: Normalizing the first psychological risk factor weight, the second psychological risk factor weight, and the third psychological risk factor weight of the psychological risk factors to obtain a weight matrix; Calculate the range of each of the psychological risk factors, and for each of the psychological risk factors j (j = 1, 2, ..., n), calculate the maximum weight Max j and weight minimum Min j : in, is the weight of the first psychological risk factor, is the weight of the second psychological risk factor, is the weight of the third psychological risk factor; Calculate the range between the maximum weight value and the minimum weight value: [Tidy j =Max j -Min j ] Calculate the final weights of the stated psychological risk factors: Among them, (k j )for: is the weight representing the weight source of the psychological risk factor.

7. A psychological risk prediction method according to claim 3, characterized in that: The AHP hierarchical analysis method is used to obtain the third psychological risk factor weight of each of the psychological risk factors, and further includes: Multiple experts assess whether the psychological risk factors derived through the psychological risk system should be retained, and the assessment criteria are the relevance to the occurrence of accidents; A comparison matrix was constructed to conduct pairwise comparisons of psychological risk factors; Calculate the consistency ratio of each expert and the average consistency ratio of all experts; the consistency ratio CR of each expert, if CR is less than 0.1, it is considered that the expert's score is consistent. The CR value calculation formula is as follows: Where CI = (λ max -n) / (n-1), where λmax is the maximum eigenvalue of the judgment matrix; n is the order of the judgment matrix; RI is the random consistency index, which is a constant related to the matrix order and is obtained by looking up the table; The experts who passed the consistency test were assigned weights. The weight calculation formula for each expert is as follows: Among them, wi is the weight of the i-th expert; CI i is the consistency index of the ith expert; m is the total number of experts; The average consistency ratio of all experts, GCR, is the mean of the CR of each expert. If the GCR is less than 0.1, the overall score is considered to be consistent; The eigenvalue method was used to calculate the weight of each expert's psychological risk factor and the weight of the third psychological risk factor. Among them, Wj is the comprehensive weight of psychological risk factor j, wi is the weight of the i-th expert, Wij is the weight of psychological risk factor j given by the i-th expert, and m is the total number of experts.

8. A psychological risk prediction system, characterized in that: include: Behavior evidence acquisition module, used to obtain the behaviors of pilots mapping the psychological risks in the relevant investigation reports on human factors in aviation accidents as behavioral evidence; a risk system construction module, for extracting, summarizing and defining a plurality of psychological risk factors corresponding to behavioral evidence of the aviation accident, and constructing a psychological risk system for the aviation accident, wherein the psychological risk system includes psychological risk factors, risk levels of psychological risk factors, and types of behavioral evidence corresponding to the risk levels of the psychological risk factors and their occurrence frequencies; A danger level determination module, for determining the danger level standard of the psychological risk factor for the pilots in the analysis sample based on the number and frequency of the behavioral evidences shown by the pilots in the aviation accident under each of the psychological risk factors; A quantitative analysis scoring module is used to select aviation accident cases that meet specific conditions as quantitative analysis samples, and score the danger level of the psychological risk factors of the quantitative analysis samples according to the danger level standard of the psychological risk factors to obtain accident psychological risk level data; The psychological risk factor weight acquisition module is used to extract the first psychological risk factor weight based on partial least squares discriminant analysis, use R2X, R2Y and Q2Y to evaluate the model fitting effect, and use the VIP value as the first psychological risk factor weight of each of the psychological risk factors; divide the data into a training set and a test set to build a Lasso regularized logistic regression model, and output the second psychological risk factor weight of each of the psychological risk factors according to the logistic regression model; use the AHP hierarchical analysis method to obtain the third psychological risk factor weight of each of the psychological risk factors; The final weight acquisition module is used to normalize the first psychological risk factor weight, the second psychological risk factor weight and the third psychological risk factor weight of the psychological risk factor, and obtain the final weight of the psychological risk factor using the range maximization method.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, a psychological risk prediction method as described in any one of claims 1-7 is implemented.

10. A computer-readable storage medium, characterized in that: The storage medium stores instructions, which, when executed, execute a psychological risk prediction method as described in any one of claims 1-7.

Images