Prediction and identification method of dispatcher's mental fatigue based on physiological and reaction time indicators

Abstract

The present invention relates to a method for predicting and distinguishing the psychological fatigue of dispatchers based on physiological and reaction time indicators. First, a psychological fatigue state indicator is constructed; by designing a test program, a psychological fatigue state test webpage system is constructed; test data of objective reaction time indicators are collected and data processing is performed; the influence probability of subjective and objective influencing factors under different psychological fatigue states is calculated, and a reaction time prediction model is constructed; based on the psychological fatigue state indicator threshold, a model for distinguishing the psychological fatigue state of dispatchers is constructed; combining the test data with the prediction results, the psychological fatigue state of dispatchers is distinguished according to three states: no fatigue, moderate fatigue, and severe fatigue. This method realizes the quantitative evaluation of the psychological fatigue state of dispatchers; the system is simple and easy to deploy: it overcomes the problems of large size, high cost, and cumbersome operation of traditional equipment; it meets the autonomous guarantee requirements of high-reliability operation of civil aviation; and it has broad application prospects in aviation operation safety management and monitoring.

Description

Technical Field

[0001] The present invention relates to a method for predicting and identifying the mental fatigue state of civil aviation dispatchers, and in particular to a method for predicting and identifying mental fatigue of dispatchers based on physiological and reaction time indicators, belonging to the field of aviation operation human factor engineering and intelligent safety assurance technology. Background Art

[0002] Dispatchers, a core position in civil aviation operations control, are responsible for monitoring flight dynamics, assessing operational risks, and coordinating operational resources. During flight operations, complex weather conditions, airspace restrictions, and emergencies often impact flight schedules, forcing dispatchers to rapidly adjust flight plans and make safety assessments. This requires a high degree of sustained attention and reaction time.

[0003] The working characteristics such as long-term high-intensity duty and alternating day and night shifts make dispatchers prone to mental fatigue, which manifests as decreased alertness, prolonged reaction time, and increased operational errors. In severe cases, it may pose a hidden danger to aviation operation safety.

[0004] Existing research has largely focused on fatigue mechanisms and the changing trends of single physiological parameters, lacking systematic task design and quantitative standards. Testing tools often rely on expensive and cumbersome laboratory equipment, making them difficult to implement in the field. Existing methods primarily rely on static fatigue state identification, lacking the ability to predict fatigue trends, making timely warnings and interventions difficult, limiting their practical application in operational support scenarios. Furthermore, fatigue state classification methods are highly subjective and lack precise identification based on data driven by multiple indicators and influencing factors. Summary of the Invention

[0005] This invention aims to provide a method for predicting and identifying dispatcher mental fatigue based on physiological and reaction-time indicators, enabling dynamic prediction and graded identification of fatigue states applicable on-site. This method integrates EEG signals, eye movement parameters, and behavioral response data to construct a structured mental fatigue identification system.

[0006] First, a psychological fatigue state index constructed from physiological indicators and objective reaction time indicators was determined; eight fatigue-sensitive discrimination indicators were screened, including EEG, eye movement and multiple reaction time parameters; a psychological fatigue test task system was constructed to collect EEG, eye movement and individual influencing factor data and automatically record reaction time information; data preprocessing was completed through steps such as filtering and anomaly removal; a Bayesian network model was constructed to infer the reaction time growth rate based on individual influencing factors, and recursively generate hourly prediction values ​​based on the first test data; Gaussian kernel density estimation, sliding window and clustering methods were used to determine the discrimination thresholds of each indicator between different fatigue states; combining the manually set thresholds with the Gaussian mixture model clustering results, a multi-weight fusion method was used to realize the discrimination of no fatigue, moderate fatigue and severe fatigue in psychological fatigue states.

[0007] The technical solution adopted by the present invention is: a method for predicting and distinguishing psychological fatigue of dispatchers based on physiological and reaction time indicators includes the following steps:

[0008] Step 1: Determine the mental fatigue state index constructed by physiological indicators and objective reaction time indicators; the physiological indicators are α wave power spectral density, θ wave power spectral density, gaze speed, and blink frequency; the objective reaction time indicators are psychomotor vigilance reaction time, sustained attention reaction time, subtle attention reaction time, and visual fatigue reaction time.

[0009] Step 2: Determine that the dispatcher's psychological fatigue test consists of three parts: physiological index test, objective reaction time index test, and subjective and objective influencing factors of psychological fatigue. By designing a test program, build a psychological fatigue status test web page system. The psychological fatigue status test web page system connects Java and the backend database through PHP.

[0010] Step 3: Collect test data of objective reaction time indicators through the psychological fatigue test web system, and collect physiological indicators through EEG equipment and eye movement equipment respectively, and then process the collected test data.

[0011] Step 4: Based on the processed test data, a Bayesian network of dispatchers is constructed to calculate the probability of the influence of subjective and objective influencing factors on sustained attention reaction time, subtle attention reaction time, and visual fatigue reaction time under different psychological fatigue states, and then a prediction model for the dispatcher's sustained attention reaction time, subtle attention reaction time, and visual fatigue reaction time is constructed.

[0012] Step 5. Based on the dispatcher's sustained attention reaction time, subtle attention reaction time, and visual fatigue reaction time prediction models, the reaction time data of each item in the first test are used to calculate the corresponding reaction time prediction value for each hour thereafter, which is the prediction result.

[0013] Step 6: Determine the threshold value of each mental fatigue status indicator by analyzing each test data.

[0014] Step 7. Based on the threshold of the psychological fatigue status indicator, a model for distinguishing the psychological fatigue status of the dispatcher is constructed. Based on the model, the psychological fatigue status of the dispatcher is distinguished according to the test data and the prediction results according to the three states of no fatigue, moderate fatigue, and severe fatigue.

[0015] The design principle of this invention is to construct a psychological fatigue prediction and discrimination model centered on reaction time, based on the high-intensity mental workload inherent in the civil aviation dispatcher's job. Dispatchers are required to continuously receive and process large amounts of dynamic information, such as flight information, ground-to-air communications, and flight operations data. The reception, discrimination, judgment, and feedback of this information are highly dependent on the individual's ability to maintain attention and react quickly. Reaction time, the time elapsed from external stimulus to response, serves as a key indicator of neuromuscular coordination and work agility, reflecting changes in psychological fatigue.

[0016] In combination with the dispatcher's task requirements, the present invention constructs a reaction time prediction system from three aspects: sustained attention, subtle discrimination ability and visual fatigue. Among them, sustained attention affects the efficiency of monitoring external information, subtle discrimination ability is related to the judgment of complex or small changes, and visual fatigue limits the accuracy and response speed under long-term high-load conditions. The system collects multimodal physiological indicators such as EEG and eye movements, extracts the discrimination threshold of each indicator based on a data-driven method, and divides the sample space in combination with a clustering algorithm to achieve the discrimination of the current state of mental fatigue. Finally, a prediction model based on the reaction time growth trend is constructed, and a mental fatigue state discrimination model is constructed in combination with physiological and objective reaction time indicators, corresponding to the two major functional modules of mental fatigue development prediction and state discrimination.

[0017] The effects of the present invention and the beneficial effects produced are:

[0018] 1. Numerical identification and trend prediction: This invention realizes the quantitative evaluation of psychological fatigue status, predicts the development of dispatchers' psychological fatigue status based on the changing trend of reaction time, and completes the identification of psychological fatigue status in combination with physiological data, assisting in job risk identification and response strategy formulation.

[0019] 2. The system is simple and easy to deploy: The test platform has a compact structure and a convenient collection process, and can be deployed and applied in actual working environments. It overcomes the problems of traditional equipment such as large size, high cost, and cumbersome operation, and avoids the defects of questionnaire evaluation that are easily affected by subjective bias.

[0020] 3. Strong technological autonomy: Break the dependence on foreign key technologies, build a full-process controllable psychological fatigue testing and prediction method system, and meet the independent guarantee needs of high-reliability operation of civil aviation.

[0021] 4. Wide range of applications: The system can be extended to psychological fatigue assessment tasks for dispatchers, air traffic controllers, safety inspectors and other positions, and has broad application prospects in aviation operation safety management and human factors monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a flow chart of a method for predicting and distinguishing dispatcher mental fatigue based on physiological and reaction time indicators in an embodiment of the present invention;

[0023] Figure 2 This is an example diagram of a sustained attention reaction time data test interface in an embodiment of the present invention;

[0024] Figure 3 This is an example diagram of the subtle attention reaction time data test interface in an embodiment of the present invention;

[0025] Figure 4 This is an example diagram of the visual fatigue reaction time data test interface in an embodiment of the present invention;

[0026] Figure 5 Flowchart of the data testing procedure for psychological fatigue influencing factors in an embodiment of the present invention;

[0027] Figure 6 A flowchart of a program for constructing a prediction model for sustained attention, subtle attention, and visual fatigue reaction time in an embodiment of the present invention;

[0028] Figure 7 This is a flow chart of a procedure for determining a threshold value of a mental fatigue state indicator according to an embodiment of the present invention;

[0029] Figure 8 Flowchart of the mental fatigue state determination program in an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The present invention will be further described below with reference to the accompanying drawings and examples.

[0031] like Figure 1 As shown, a method for predicting and distinguishing dispatcher mental fatigue based on physiological and reaction time indicators includes the following steps:

[0032] Step 1: Determine the mental fatigue state index constructed by physiological indicators and objective reaction time indicators; the physiological indicators are α wave power spectral density, θ wave power spectral density, gaze speed, and blink frequency; the objective reaction time indicators are psychomotor vigilance reaction time, sustained attention reaction time, subtle attention reaction time, and visual fatigue reaction time.

[0033] The preliminary test indicators include electrocardiogram (ECG) signals, electrodermal signals, electromyography (EMG) signals, eye movement indicators, psychomotor alertness test indicators, sustained attention test indicators, fine-grained attention test indicators, and visual fatigue test indicators. EEG signal indicators include delta wave power spectral density, theta wave power spectral density, alpha wave power spectral density, beta wave power spectral density, and gamma wave power spectral density. Eye movement indicators include gaze velocity, blink frequency, gaze duration, number of fixations, total gaze time, saccade amplitude, saccade velocity, blink duration, and percentage of eyes closed.

[0034] After actual measurements and based on the sensitivity, specificity and repeatability of the psychological fatigue indicators of the dispatchers being tested, further screening was carried out among the above-mentioned preliminary indicators, and eight items including physiological indicators and objective reaction time indicators were finally determined as psychological fatigue state indicators. The physiological indicators were α-wave power spectral density, θ-wave power spectral density, gaze speed and blinking frequency. The objective reaction time indicators were psychomotor vigilance reaction time, sustained attention reaction time, subtle attention reaction time and visual fatigue reaction time.

[0035] Step 2: Determine that the dispatcher's psychological fatigue test consists of three parts: physiological index test, objective reaction time index test, and subjective and objective influencing factors of psychological fatigue. By designing a test program, build a psychological fatigue status test web page system. The psychological fatigue status test web page system connects Java and the backend database through PHP.

[0036] The dispatcher's psychomotor alertness, sustained attention, fine attention, and visual fatigue reaction time are tested through the psychological fatigue test web system; the psychological fatigue test web system automatically records the start time, average reaction time, number of errors, and data on psychological fatigue influencing factors of each test, and generates a timestamp. The reaction time is measured in milliseconds.

[0037] When testing dispatchers, EEG signal data is obtained by wearing EEG equipment, and eye movement data is obtained by installing an eye tracker; data on factors affecting mental fatigue and objective reaction time data are obtained through the mental fatigue test web system and test program.

[0038] Based on the mental fatigue test web system, such as Figures 2 to 5 As shown, the objective reaction time index test is divided into four task tests: psychomotor vigilance, sustained attention, fine attention, and visual fatigue. The test procedures of the four task tests are as follows:

[0039] A1. Before the test begins, the center of the screen displays the test judgment, click method prompts, and the start test button.

[0040] A2. After the dispatcher clicks the Start Test button, several test questions appear in the center of the screen. The dispatcher makes decisions and clicks buttons based on the prompts before the test begins.

[0041] A3. After the dispatcher makes a judgment on each set of test questions presented on the screen and clicks a button, the mental fatigue test webpage system automatically jumps to the next set of test questions until several sets of test questions are completed.

[0042] A4. After the mental fatigue test web system completes all presented test questions, the test data is automatically saved to the background database.

[0043] A5. Based on the dispatcher's judgments and keystrokes on the test questions, count the number of correct answers and incorrect answers for each set of test questions.

[0044] A6. Calculate the reaction time R for completing each set of test questions. k :

[0045] (1);

[0046] Where k is the objective reaction time index, The timestamp displayed for the test question, The timestamp of the key operation.

[0047] A7. Store the test data of each dispatcher and calculate the average reaction time after the test. k :

[0048] (2);

[0049] Where R k When the reaction is k is the total number of operations for the kth objective reaction time indicator test.

[0050] The average reaction time of psychomotor vigilance was taken as the psychomotor vigilance reaction time data, the average reaction time of sustained attention was taken as the sustained attention reaction time data, the average reaction time of subtle attention was taken as the subtle attention reaction time data, and the average reaction time of visual fatigue was taken as the visual fatigue reaction time data.

[0051] like Figure 2 As shown, sustained attention reaction time data test: the average reaction time of judging the aircraft's nose direction is used as test data.

[0052] like Figure 3 As shown, the subtle attention reaction time data test: the average reaction time is judged by matching the flight number matrix as the test data.

[0053] like Figure 4As shown, visual fatigue reaction time data test: the average reaction time is selected as the test data using route tracking.

[0054] like Figure 5 As shown, data test of factors affecting mental fatigue: data of various factors affecting mental fatigue are collected as test data.

[0055] Step 3: Collect test data of objective reaction time indicators through the psychological fatigue test web system, and collect physiological indicators through EEG equipment (forehead-mounted EEG signal recorder) and eye movement equipment (eye movement signal tracking recorder), and then process the collected test data.

[0056] After completing the test data collection, the acquired EEG signal data was filtered and denoised. The α-wave power spectral density and the θ-wave power spectral density were extracted by fast Fourier transform and outliers were removed by quartile method. The mean of the α-wave and θ-wave power spectral density was calculated for each hour. The gaze speed and blink count were extracted from the acquired eye movement data and outliers were removed by quartile method. The mean of the gaze speed and blink frequency was calculated for each hour. The collected reaction time data was subjected to quartile method to remove outliers. The test data after processing of each indicator was directly time-aligned according to the recorded timestamp.

[0057] Step 4: Based on the processed test data, a Bayesian network of dispatchers is constructed to calculate the probability of the influence of subjective and objective influencing factors on sustained attention reaction time, subtle attention reaction time, and visual fatigue reaction time under different psychological fatigue states, and then a prediction model for the dispatcher's sustained attention reaction time, subtle attention reaction time, and visual fatigue reaction time is constructed.

[0058] like Figure 6 As shown in Figure 2, the steps to construct a prediction model for dispatcher sustained attention reaction time, subtle attention reaction time, and visual fatigue reaction time are as follows:

[0059] 1. Calculate the reaction time growth rate based on the test data, and use piecewise linear fitting to determine the growth rate stages and corresponding growth rates under different mental fatigue states.

[0060] The growth rate stages of various indicators and their corresponding growth rates are shown in Table 1:

[0061] Table 1 Growth rate stages of various indicators and corresponding growth rates

[0062]

[0063] 2. Construct a Bayesian network for flight dispatchers to analyze and calculate the probability of the impact of subjective and objective factors of mental fatigue on sustained attention, subtle attention, and visual fatigue.

[0064] 3. After inputting the test data, the growth rate of the sustained attention reaction time, subtle attention reaction time, and visual fatigue reaction time indicators corresponding to each hour is inferred based on the probability of the impact of the subjective and objective influencing factors of mental fatigue on sustained attention, subtle attention, and visual fatigue. Finally, taking the first test data of sustained attention reaction time, subtle attention reaction time, and visual fatigue reaction time as the starting point, the corresponding reaction time prediction value for each hour is recursively generated in combination with the growth rate:

[0065] (3);

[0066] Where, is the predicted value of the kth objective reaction time indicator after the tth hour, is the first test reaction time data of the kth objective reaction time indicator, n is the number of tests that need to be predicted, is the growth rate stage inferred from the i-th test of the k-th objective reaction time indicator, is the end time of the i-th test, is the start time of the i-th test.

[0067] Step 5. Based on the dispatcher's sustained attention reaction time, subtle attention reaction time, and visual fatigue reaction time prediction models, the reaction time data of each item in the first test are used to calculate the corresponding reaction time prediction value for each hour thereafter, which is the prediction result.

[0068] Step 6: Determine the threshold value of each mental fatigue status indicator by analyzing each test data.

[0069] like Figure 7 As shown in Figure 2, the threshold determination process for mental fatigue state indicators performs the following operations: based on the test data, the threshold of each mental fatigue state indicator is determined respectively. The threshold of the EEG indicator α wave power and θ wave power is determined by extracting the distribution valley value through Gaussian kernel density estimation; the eye movement indicator gaze speed and blink frequency are analyzed by sliding window analysis to extract the extreme value of the coefficient of variation as the reference boundary; the reaction time indicator performs K-means clustering and density-based spatial (DBSCN) clustering to divide the agile, slightly slow and severely slow response groups, and calculates the cluster boundary mean as the threshold; the discrimination threshold of the mental fatigue state discrimination indicator in the three states of no fatigue, moderate fatigue and severe fatigue is obtained, and the cluster boundary discrimination is obtained by calculating the ratio of the between-class variance to the within-class variance of each mental fatigue state discrimination indicator. The discrimination threshold of each mental fatigue state discrimination indicator in the three states of no fatigue, moderate fatigue and severe fatigue is shown in Table 2:

[0070] Table 2 Discrimination threshold

[0071]

[0072] The discrimination thresholds of the various mental fatigue state discrimination indicators given in Table 2 above under the three states of no fatigue, moderate fatigue, and severe fatigue are used as a reference in the process of constructing the discrimination model.

[0073] Step 7. Based on the threshold of the psychological fatigue status indicator, a model for distinguishing the psychological fatigue status of the dispatcher is constructed. Based on the model, the psychological fatigue status of the dispatcher is distinguished according to the test data and the prediction results according to the three states of no fatigue, moderate fatigue, and severe fatigue.

[0074] like Figure 8 As shown in the figure, the mental fatigue state discrimination process performs the following operations: Based on the discrimination thresholds of each mental fatigue state discrimination indicator in the three states of no fatigue, moderate fatigue, and severe fatigue, the test data is first clustered and preliminarily divided using a Gaussian mixture model to obtain pseudo-classification results. Based on the pseudo-classification results, the ratio of the between-class variance to the within-class variance of each mental fatigue state indicator is calculated to obtain the rule path discrimination of each mental fatigue state indicator. The rule path discrimination of each mental fatigue state indicator is normalized by standard deviation, and a fusion weight is set based on the cluster boundary discrimination and rule path discrimination of each mental fatigue state indicator as the threshold weight. This constitutes the dispatcher mental fatigue state discrimination model, which outputs the three states of no fatigue, moderate fatigue, and severe fatigue.

[0075] Taking the test data of a dispatcher after working for 7 hours as an example, the collected and processed test data are calculated by formula (1) and formula (2), and the fatigue state discrimination result is obtained according to the psychological fatigue state discrimination model. The test data and discrimination results of various indicators of the dispatcher are shown in Table 3:

[0076] Table 3 Test data and discrimination results

[0077]

[0078] According to formula (3), the predicted results of the three reaction times of sustained attention, subtle attention, and visual fatigue of the dispatchers after 2 hours were obtained. The prediction and discrimination results were further obtained based on the predicted data, as shown in Table 4:

[0079] Table 4 Prediction results and prediction discrimination results

[0080]

[0081] According to the test data shown in Table 2, formulas (1), (2), and (3) were used to predict the three reaction times of sustained attention, fine attention, and visual fatigue of the dispatchers after working for 2 hours. The fatigue state of the dispatchers in this test was judged by the psychological fatigue state discrimination model.

[0082] As shown in Table 2, after the test data of the dispatchers were collected, the EEG signal data were filtered and denoised. The power spectral density of α and θ waves were extracted by fast Fourier transform and outliers were removed by quartile method. The mean of the power spectral density of α and θ waves was calculated for each hour, and the mean of the power spectral density of θ was -0.47 and the mean of the power spectral density of α was -1.58. The gaze speed and blink counts were extracted from the eye movement data and outliers were removed by quartile method. The mean of the gaze speed and blink counts for each hour were calculated, and the gaze speed was 1.96 pixels / millisecond and the blink count was 29 times / minute. According to the mental fatigue test webpage system, the reaction time data collected were quartile method to remove outliers. The number of sustained attention errors was 5 and the number of subtle attention errors was 9. The number of attention errors is 2, and the number of visual fatigue errors is 1. After removing outliers from the collected reaction time data, according to formulas (1) and (2), the sustained attention reaction time is 0.59 seconds, the subtle attention reaction time is 5.36 seconds, and the visual fatigue reaction time is 0.29 seconds. Based on the sustained attention, subtle attention, and visual fatigue reaction time prediction model, according to formula (3), it is predicted that after 2 hours, the sustained attention will reach 0.62 seconds, the subtle attention reaction time will reach 5.49 seconds, and the visual fatigue reaction time will reach 0.38 seconds. The test data after processing each indicator are directly time-aligned according to the recorded timestamps. According to the mental fatigue state discrimination model, this group of data meets the characteristics of severe mental fatigue. Therefore, the test data of the tested dispatcher is judged to be in a severe fatigue state.

[0083] By integrating multimodal indicators such as EEG, eye movement, physiology and reaction time, and through the discrimination of various feature items and model prediction results, it can be concluded that the dispatcher under test has shown a relatively obvious state of severe mental fatigue.

[0084] According to the Manual on Oversight of Mental Fatigue Management Practices (Doc9966) issued by the International Aviation Organization, personnel with moderate or above mental fatigue status are not allowed to perform safety-related duties.

Claims

1. A method for predicting and distinguishing dispatcher mental fatigue based on physiological and reaction time indicators, characterized by: The method comprises the following steps: Step 1: Determine the mental fatigue state index constructed by physiological indicators and objective reaction time indicators; the physiological indicators are alpha wave power spectral density, theta wave power spectral density, gaze speed, and blink frequency; the objective reaction time indicators are psychomotor vigilance reaction time, sustained attention reaction time, subtle attention reaction time, and visual fatigue reaction time; Step 2: Determine the dispatcher's mental fatigue test to consist of three parts: physiological index test, objective reaction time index test, and test of subjective and objective factors affecting mental fatigue. Design a test program and build a mental fatigue status test webpage system. The mental fatigue status test webpage system connects to Java and the backend database via PHP. Step 3: Collect objective reaction time index test data through the mental fatigue test web system, and collect physiological indicators through EEG equipment and eye movement equipment, and then process the collected test data; Step 4: Based on the processed test data, a Bayesian network of dispatchers is constructed to calculate the probability of the influence of subjective and objective factors on sustained attention reaction time, subtle attention reaction time, and visual fatigue reaction time under different mental fatigue states, and then a prediction model for the dispatcher's sustained attention reaction time, subtle attention reaction time, and visual fatigue reaction time is constructed; Step 5: Based on the dispatcher's sustained attention reaction time, subtle attention reaction time, and visual fatigue reaction time prediction models, calculate the corresponding reaction time prediction value for each hour after the initial test using the reaction time data, which is the prediction result; Step 6: Determine the threshold of each mental fatigue state indicator by analyzing each test data; Step 7. Based on the threshold of the psychological fatigue status indicator, a model for distinguishing the psychological fatigue status of the dispatcher is constructed. Based on the model, the psychological fatigue status of the dispatcher is distinguished according to the test data and the prediction results according to the three states of no fatigue, moderate fatigue, and severe fatigue.

2. The method for predicting and distinguishing dispatcher mental fatigue based on physiological and reaction time indicators according to claim 1, characterized in that: When testing dispatchers, EEG signal data is obtained by wearing EEG equipment, and eye movement data is obtained by installing an eye tracker. Data on factors affecting mental fatigue and objective reaction time are obtained through the mental fatigue test web system and test program. After completing the test data collection, the acquired EEG signal data will be filtered and denoised, the α wave power spectral density and the θ wave power spectral density will be extracted and outliers will be removed, the mean of the α wave and the θ wave power spectral density for each hour will be calculated respectively, the acquired eye movement data will be used to extract the gaze speed and blink count, and outliers will be removed, the mean of the gaze speed and blink frequency for each hour will be calculated respectively, the outliers will be removed from the collected reaction time data, and the test data after processing of various indicators will be directly time-aligned according to the recorded timestamps.

3. The method for predicting and distinguishing dispatcher mental fatigue based on physiological and reaction time indicators according to claim 2, characterized in that: The dispatcher's psychomotor alertness, sustained attention, fine attention, and visual fatigue reaction time are tested through the psychological fatigue test web system; the psychological fatigue test web system automatically records the start time, average reaction time, number of errors, and psychological fatigue influencing factors of each test, and generates a timestamp.

4. The method for predicting and distinguishing dispatcher mental fatigue based on physiological and reaction time indicators according to claim 3, characterized in that: According to the mental fatigue test webpage system, the objective reaction time index test is divided into four task tests: psychomotor vigilance, sustained attention, fine attention, and visual fatigue. The test procedures for the four task tests are as follows: A1. Before the test begins, the center of the screen displays the test judgment and click method prompts and the start test button; A2. After the dispatcher clicks the Start Test button, several test questions will appear in the center of the screen. The dispatcher will make decisions and click buttons based on the prompts before the test begins. A3. After the dispatcher makes a judgment on each set of test questions presented on the screen and clicks a button, the mental fatigue test webpage system automatically jumps to the next set of test questions until several sets of test questions are completed; A4. After the mental fatigue test web system completes all presented test questions, the test data will be automatically saved to the backend database; A5. Based on the dispatcher's judgment and keystrokes on the test questions, count the number of correct answers and incorrect answers for each test question. A6. Calculate the reaction time to complete each set of test questions R k : （1）； Where, k As an objective reaction time indicator, The timestamp displayed for the test question, The timestamp of the key operation; A7. Store the test data of each dispatcher and calculate the average reaction time after the test. A k : （2）； Where, R k For reaction, G k For the k The total number of operations tested by the objective reaction time index.

5. The method for predicting and distinguishing dispatcher mental fatigue based on physiological and reaction time indicators according to claim 4, characterized in that: The steps to construct a prediction model for dispatcher sustained attention response time, subtle attention response time, and visual fatigue response time are as follows: The reaction time growth rate was calculated based on the test data, and piecewise linear fitting was used to determine the growth rate stages and corresponding growth rates under different mental fatigue states. A Bayesian network for flight dispatchers was constructed to analyze and calculate the probability of the impact of the subjective and objective influencing factors of mental fatigue on sustained attention, subtle attention, and visual fatigue. After inputting the test data, the growth rate stages of the sustained attention reaction time index, subtle attention reaction time index, and visual fatigue reaction time index corresponding to each hour were inferred through the probability of the impact of the subjective and objective influencing factors of mental fatigue on sustained attention, subtle attention, and visual fatigue. Finally, taking the first test data of sustained attention reaction time, subtle attention reaction time, and visual fatigue reaction time as the starting point, the predicted reaction time values ​​corresponding to each hour were recursively generated in combination with the growth rate: （3）； Where, For the k The objective reaction time index t The predicted reaction time after hours, For the k The first test of the objective reaction time index reaction time data, n is the number of tests that need to be predicted, For the k The growth rate stage inferred from the i-th test of the objective reaction time indicator, For the i The end time of the test, is the start time of the i-th test.

Images