Comprehensive detection method for workload of civil aviation pilot

Abstract

The invention relates to the technical field of pilot load detection, in particular to a civil aviation pilot workload comprehensive detection method. A workload index system including six first-level indexes including working conditions, role pressure, interpersonal relationship factors, occupational development, organization structures and family factor influences is constructed. The method is combined with a polymorphic simulation evaluation method to obtain the measurement data of each index of the workload of the tested pilot, calculates the evaluation score of the workload of the civil aviation pilot through an algorithm, carries out classified prediction and grading on different pressure states, can achieve the quantitative evaluation of the daily workload of the pilot, and improves the evaluation accuracy of the workload of the pilot. And on the basis of real-time evaluation of physiological indexes, data support is provided for civil aviation pilot selection and continuous monitoring of the psychological competency of in-service pilots, and a basis is provided for airline companies and bureau parties to carry out pilot psychological competency targeted training.

Description

Technical Field

[0001] The present invention relates to the technical field of flight crew workload detection, and specifically to a comprehensive detection method for the workload of civil aviation pilots. Background Art

[0002] Civil aviation safety is of great significance in protecting the safety of people's lives, maintaining the sustainable development of the aviation industry, strengthening international cooperation and regulatory compliance, etc. Human factors, as one of the main causes of civil aviation safety accidents, include the behavior, attitude, sense of responsibility, pressure, etc. of flight operation personnel. In particular, pilots of civil aviation small and medium-sized aircraft are mostly responsible for all aspects of flight tasks by themselves, including planning, preparation, execution and monitoring of flights, and also need to cope with long flight tasks, frequent takeoffs and landings and tight schedules. In addition, they need to comply with aviation regulations and procedures, maintain accurate navigation and communication, etc. The flight pressure they face cannot be ignored. Therefore, the detection of the psychological pressure of civil aviation small and medium-sized aircraft pilots has become an indispensable task.

[0003] However, in actual use of the existing technologies, the existing workload assessment methods mostly focus on single indicators or simple syntheses, lacking comprehensiveness and accuracy. Summary of the Invention

[0004] The purpose of the present invention is to provide a comprehensive detection method for the workload of civil aviation pilots to solve the problems raised in the above background art.

[0005] To achieve the above purpose, the present invention provides the following technical solution: A comprehensive detection method for the workload of civil aviation pilots, including the following steps:

[0006] S1. Construct a pilot workload index system: Based on psychological principles and the professional characteristics of pilots, construct a workload index system including six first-level indicators of working conditions, role pressure, interpersonal relationship factors, career development, organizational structure and family factor influence. Each first-level indicator contains multiple second-level indicators;

[0007] S2. Determine the index weights: Use the analytic hierarchy process to construct a judgment matrix through the importance degree among various load factors, and calculate the weights of each index;

[0008] S3. Data collection and analysis: Through a multi-state simulation evaluation method, obtain the measurement data of each index of the tested pilots under different workloads, use an algorithm to calculate the measured score value of the workload of civil aviation pilots, and conduct classification prediction and grade division for different pressure states;

[0009] S4. Result Output and Application: Calculate the evaluation score value of the civil aviation pilot's workload through the algorithm, display the evaluation result in a visual form, classify and predict the different stress states of the pilot, and divide the levels.

[0010] Preferably, the secondary indicators corresponding to each primary indicator in the step S1 are as follows:

[0011] Working conditions: Low attendance efficiency of pilots, too long station stay time during the mission period, and overly heavy training courses;

[0012] Role stress: Crew cooperation stress, responsibility stress of flight tasks, and stress of regular assessment and evaluation;

[0013] Interpersonal relationship factors: The dispatcher does not respect the pilot, adapting to flights with different crews and the atmosphere within the crew is not very harmonious;

[0014] Career development: Too slow promotion of flight levels, worry about stability and development prospects, and the impact of unsafe events on the career;

[0015] Organizational structure: Unreasonable company reward and punishment system, too low autonomy in scheduling, and lack of guarantee for pilots' work;

[0016] Influence of family factors: Influence of irregular working hours, frequent conflicts with family members, and lack of understanding and support from family members.

[0017] Preferably, the calculation method of the weight in the step S2 is as follows:

[0018] a. Construct a judgment matrix, and the elements in the matrix are the comparison results of the relative importance between indicators;

[0019] b. Conduct a consistency test on the judgment matrix, including matrix normalization, summing by row, obtaining the weight vector, calculating the maximum eigenvalue and the consistency index to ensure the consistency of the judgment matrix;

[0020] c. According to the consistency test results, adjust the judgment matrix until the consistency requirement is met, so as to determine the weights of each indicator.

[0021] Preferably, the evaluation method of polymorphic simulation in the step S3 includes simulating flight tasks, simulating flight environments, and simulating flight stress to comprehensively reflect the workload state of pilots in real flights.

[0022] Preferably, the evaluation score value can be used to evaluate the psychological stress level of pilots, predict their operation behaviors, and provide a scientific basis for the safety and reliability of flight operations.

[0023] Compared with the prior art, the beneficial effects of the present invention are:

[0024] 1. The evaluation method of the present invention combines polymorphic simulation to obtain measurement data of various indicators of the workload of the tested pilots, calculates the evaluation score of the workload of civil aviation pilots through an algorithm, classifies and predicts different stress states, and divides grades, which can realize the quantitative evaluation of the daily workload of pilots and real-time evaluation based on physiological indicators, provide data support for the selection of civil aviation pilots and the continuous monitoring of the psychological competence of in-service pilots, and provide a basis for airlines and the aviation authority to carry out targeted training on the psychological competence of pilots. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 FIG. is the overall flowchart of the comprehensive detection method for the workload of civil aviation pilots of the present invention;

[0026] Figure 2 FIG. is the structural diagram of the workload index system of civil aviation pilots of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0027] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0028] Today, flight technology and equipment have been highly developed. Through the risk matrix evaluation method, the assignment of the possibility and consequence severity of civil aviation safety accidents shows that civil aviation safety accidents belong to high-risk accidents. According to statistics, among the flight accidents since 2010, 57% were caused by pilot errors. As the person on the plane who most urgently needs safety and whose duty is to ensure safety, it is undoubtedly the need of the times, civil aviation, and safety to detect the pressure of pilots, identify and prevent potential safety hazards caused by pressure before flight, and reduce the possibility of safety accidents.

[0029] As a reflection of an individual's psychology, pressure itself is not quantifiable, and we cannot accurately perceive pressure by ourselves. According to existing research, the pressure level of pilots has an impact on various key factors such as negative emotions, safety compliance behaviors, unsafe behaviors, and safety performance. Therefore, detecting the psychological pressure of civil aviation pilots can quantify the degree of pilot pressure, more rationally evaluate and screen pilots with qualified psychological conditions, and reduce the occurrence of aviation safety accidents.

[0030] The existing academic research and achievements on the psychological stress detection of civil aviation pilots of small and medium-sized aircraft still have deficiencies. For example, there is a lack of standardized assessment tools, a lack of large-sample and long-term follow-up studies, and a lack of research on refined psychological stress types for different objects. Therefore, the present invention proposes a comprehensive detection method for the workload of civil aviation pilots.

[0031] Please refer to Figure 1-2 , the present invention provides a technical solution: a comprehensive detection method for the workload of civil aviation pilots, a workload index for pilots proposed based on psychological principles and the professional characteristics of pilots. This index system includes 6 first-level indicators: working conditions, role stress, interpersonal relationship factors, career development, organizational structure, and family factor influence. Each first-level indicator contains multiple second-level indicators, and the number of second-level indicators is 18. The 18 second-level indicators include: low attendance efficiency of pilots, too long stationing time during the mission period, overly heavy training courses, corresponding to the first-level indicator of working conditions. Crew cooperation pressure, responsibility pressure of flight tasks, pressure of regular assessment and evaluation, corresponding to the first-level indicator of role stress. The dispatcher does not respect the pilot, adapting to flights of different crews, and the atmosphere within the crew is not very harmonious, corresponding to the first-level indicator of interpersonal relationship factors. Too slow promotion of flight levels, worrying about stability and development prospects, and the impact of unsafe events on the career, corresponding to the first-level indicator of career development. The company's reward and punishment system is unreasonable, too low autonomy in scheduling, and the work of pilots lacks guarantee, corresponding to the first-level indicator of organizational structure. Family members do not understand, do not support, the impact of irregular working hours, and often have conflicts with family members, corresponding to the first-level indicator of family factor influence. For each level of indicators, the analytic hierarchy process is used to select the weights of the evaluation indicators, compare the importance of each pair of indicators, and determine the judgment matrix through the nine-level scale method. Through the analysis of the relevant knowledge accumulation of professionals, the judgment matrix is constructed.

[0032] Through the analysis of the relevant knowledge accumulation of professionals to obtain the importance of various load factors to obtain the weights of each factor, the specific calculation steps are as follows.

[0033] (1) Construct the judgment matrix. Where a ij is the comparison result of element i and element j, and n is the number of indicators.

[0034] A=(a ij ) n×n i,j = 1,2,...,n(1)

[0035] Consistency test of the matrix.

[0036] a. Normalize each column of the matrix.

[0037]

[0038] b. Sum by row to obtain

[0039]

[0040] Normalize the sum vector to obtain the weight as

[0041]

[0042] c. Calculate the maximum eigenvalue.

[0043]

[0044] The consistency index is

[0045]

[0046] where n is the number of indicators.

[0047] Consistency index: I C = 0, the matrix is consistent; the larger I C , the greater the degree of inconsistency of the matrix. As long as I C / R C < 0.1, it is considered that the judgment result of the comparison matrix is acceptable. The average random consistency index R C values are shown in Table 1.

[0048] Table 1: R C values

[0049]

[0050] Normalize each column of the matrix, sum by row, then normalize the vector to obtain the weight, calculate the maximum eigenvalue. If the consistency test is passed, it is considered that the judgment result of the matrix is acceptable and the obtained weight value can be used.

[0051] The weights of the first-level indicators and each second-level indicator can be obtained through the analytic hierarchy process. The fuzzy comprehensive evaluation method is used to calculate the psychological index rating and score. Using the method of expert investigation, multiple experts score each workload factor based on typical civil aviation flight accidents, symptoms, and the research results and their own experience. The scoring indicators include 2 dimensions of load status and evaluation status. Among them, the load status represents the pressure level of the load, while the evaluation status is an implementation evaluation based on the current actual situation. These 2 scores jointly reflect the load degree of this factor. The evaluation score ranges from 0 to 100 points. Determine the comment grade domain = {extremely high pressure (81 - 100 points), relatively high pressure (61 - 80 points), medium pressure (41 - 60 points), relatively low pressure (21 - 40 points), no pressure (0 - 20 points)}.

[0052] The membership degree calculation adopts a smooth transition function based on predefined scoring levels to map the expert scores to membership degree values. The score of each influencing factor consists of the scores of two dimensions: the load status and the evaluation status. For a given score s, the membership function M(s) is defined as in Equation (7).

[0053]

[0054] where: l1, l2,..., l n are the predefined scoring levels; m is the midpoint between levels; k is a constant used to adjust the steepness of the curve.

[0055] For each factor, the membership degree vectors V 负荷 , V 评估 of the load status and the evaluation status are calculated respectively. Then, the weighted average method is used to synthesize these two statuses into a single membership degree vector V 综合 .

[0056] As shown in Equation (8):

[0057] V 综合 = W 危险 * V 危险 + W 评估 * V 评估 (8)

[0058] where: W 危险 , W 评估 represent the weights of the load status and the evaluation status respectively. The weight of 0.5 can be adjusted according to the actual situation to reflect the importance of different statuses.

[0059] Based on the load status obtained from the scores of each expert and the self-evaluation status, the membership degrees of each evaluation index are comprehensively analyzed and normalized to finally obtain six first-level index evaluation matrices R1, R2,..., R 6。

[0060] Single-factor fuzzy evaluation is carried out. By using the min-max operator, the first-level fuzzy evaluation matrix is obtained. The first-level fuzzy comprehensive evaluation is to comprehensively evaluate each level of a certain factor and evaluate its impact on the object to be evaluated. The first-level fuzzy evaluation B i is obtained by using M(∧, ∨), as shown in Equation (9):

[0061]

[0062] where: a i represents the weight of the i-th factor, and r ij represents the contribution or correlation degree of the i-th factor to the j-th evaluation level.

[0063] Equation (9) can also be expressed as shown in Equation (10):

[0064] B i = max[min(a1, r1), min(a2, r 2j ),..., min(a1, r nj )](10)

[0065] Normalization is used to establish the total evaluation matrix. The product of the first-level index weights and the total evaluation matrix gives the second-level fuzzy matrix, and normalization is performed again. The distribution of the second-level comprehensive fuzzy evaluation results is based on a 5-level division of the system load status. Based on the comment-level domain, the total score f of the system can be obtained, where the total score is the product of the scores of each level and the second-level comprehensive fuzzy evaluation results.

[0066] The workload is measured through multiple physiological indicators such as electroencephalogram, electrocardiogram, and skin conductance. In this step, multi-state simulation evaluations of the simulated flight mission, simulated flight environment, and simulated flight pressure are carried out, and different workload levels are simulated and distinguished in the actual mission. A total of three levels are divided, namely: low pressure, medium pressure, and high pressure. The simulated tasks are takeoff and landing.

[0067] Low pressure: Participants will perform task simulations in a standard flight simulator. The instructor provides simple task instructions and operation guides. Participants execute standard takeoff and landing procedures. The simulator provides suitable flight conditions, such as good weather, smooth airflow, etc. Participants can complete the tasks within a suitable time without excessive time pressure and complex operation requirements.

[0068] Medium pressure: Participants still perform task simulations in the flight simulator. The instructor provides slightly more complex task instructions and operation guides. Participants execute takeoff and landing procedures, but the simulator will introduce some challenges, such as strong crosswinds, reduced visibility, etc. Participants need to maintain flight accuracy and control the aircraft under more challenging conditions. The time limit is tighter than the previous level, and participants need to complete the tasks within the specified time.

[0069] High pressure: Participants continue to perform task simulations in the flight simulator. The instructor provides complex task instructions and operation guides and introduces factors of emergency situations. Participants execute takeoff and landing procedures, but the simulator will simulate emergency situations, such as engine failure, drastic changes in airflow, etc. Participants need to make quick decisions, take appropriate emergency measures, and maintain control of the flight state. The time limit is even more urgent, and participants need to respond to emergency situations and complete the tasks within a very limited time.

[0070] For the three physiological indices of C0 complexity, KC complexity, and L1 maximum value in EEG signal acquisition. For ECG signals, they are mainly SDNN, PNN50, and the time-domain and frequency-domain eigenvalue of P, Q, R, S, and T waves. 22 indices such as the mean of R wave amplitude, the variance of R wave amplitude, the difference between the maximum and minimum values of R wave, the mean of R-R interval, the variance of R-R interval, the mean of heart rate, the variance of heart rate, etc. For GSR signals, they are mainly 26 indices such as the standard deviation of GSR data, the mean of the absolute value of GSR data, the median frequency value of the integral of power spectral density, the mean / variance / maximum value / minimum value of GSR data, the minimum ratio of GSR data, the maximum ratio of GSR data, etc. A total of 51 measurable features can be obtained. Too many features will increase the complexity of the classifier, cause overfitting, and reduce the generalization ability of the classifier. Therefore, it is necessary to screen the features and extract the effective features. The support vector machine (SVM) model is selected to estimate the classification accuracy rate using only a single feature respectively, so as to measure the classification effect of a single feature, and the features are sorted according to the size of the classification accuracy rate. Then, the features with the most important classification contribution are combined, and SVM is used again to classify mental stress for the combined features, and the mental stress classification result based on physiological data is obtained. The kernel function is selected as the Gaussian kernel function, and the penalty coefficient is 1.

[0071] After each experiment, a subjective workload assessment scale is used for scoring, and the workload is evaluated from 6 dimensions: mental demand, physical demand, time demand, self-performance, effort level, and frustration level. Each item is represented by a 20-equal division straight line, representing 0-100 points respectively. From 0 to 100 points represent the workload from "low" to "high". Among them, for the item of "self-performance", from left to right is "perfect" to "failure", that is, the lower the score, the more perfect the self-performance and the lower the task workload; the higher the score, the more failed the self-performance and the higher the task workload. The arithmetic average method is adopted, and the scores of the 6 items are added up to obtain the arithmetic average score, which is the total workload score of the surveyed object. And it is judged whether the scale passes the reliability and validity test.

[0072] The fuzzy comprehensive evaluation result obtained based on the workload index is the workload result in a recent period of time. While the data based on physiological data and the measurement data of the subjective workload assessment scale are real-time data. If two or more of the three-dimensional data exceed the medium stress level, it can be considered that the stress is relatively large.

[0073] It should be noted that, in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variation thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or device.

[0074] Although the embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. Comprehensive detection method for civil aviation pilot workload, characterized in that: It includes the following steps: S1. Construct a pilot workload index system: Based on psychological principles and the characteristics of the pilot profession, construct a workload index system including six first-level indicators, namely working conditions, role stress, interpersonal relationship factors, career development, organizational structure, and family factor influence. Each first-level indicator contains multiple second-level indicators; S2. Determine the index weights: Use the analytic hierarchy process to construct a judgment matrix through the importance degree among various load factors, and calculate the weights of each index; S3. Data collection and analysis: Through a multi-state simulation assessment method, obtain the measurement data of each index of the test pilot under different workloads, use an algorithm to calculate the measured score value of the civil aviation pilot's workload, and conduct classification prediction and grading for different stress states; S4. Result output and application: Calculate the measured score value of the civil aviation pilot's workload through an algorithm, display the evaluation results in a visual form, conduct classification prediction for different stress states of the pilot, and divide grades.

2. The comprehensive detection method for the workload of civil aviation pilots according to claim 1, wherein: The second-level indicators corresponding to each first-level indicator in step S1 are respectively: Working conditions: Low attendance efficiency of pilots, too long station time during the mission period, and overly heavy training courses; Role stress: Crew collaboration stress, responsibility stress of flight tasks, and stress of regular assessment and evaluation; Interpersonal relationship factors: The dispatcher does not respect the pilot, adapting to flights with different crews, and the atmosphere within the crew is not very harmonious; Career development: Too slow upgrade of flight levels, worry about stability and development prospects, and the impact of unsafe events on the career; Organizational structure: Unreasonable company reward and punishment systems, too low autonomy in scheduling, and lack of guarantee for pilots' work; Family factor influence: Influence of irregular working hours, frequent conflicts with family members, and lack of understanding and support from family members.

3. The comprehensive detection method for civil aviation pilot workload according to claim 1, wherein: The calculation method of the weight in step S2 is: a. Construct a judgment matrix, and the elements in the matrix are the comparison results of the relative importance among various indicators; b. Conduct a consistency test on the judgment matrix, including matrix normalization, row summation, obtaining the weight vector, calculating the maximum eigenvalue and the consistency index to ensure the consistency of the judgment matrix; c. According to the consistency test results, adjust the judgment matrix until the consistency requirement is met, so as to determine the weights of each index.

4. The comprehensive detection method for civil aviation pilot workload according to claim 1, characterized in that: The multi-state simulation assessment method in step S3 includes simulating flight tasks, flight environments, and flight stresses to comprehensively reflect the workload state of pilots in real flights.

5. The comprehensive detection method for civil aviation pilot workload according to claim 1, characterized in that: The measured score value can be used to evaluate the psychological stress level of pilots, predict their operation behaviors, and provide a scientific basis for the safety and reliability of flight operations.

Images