Manager risk preference-based mental load threshold determination method and system

Abstract

The embodiment of the invention provides a mental load threshold determination method and system based on manager risk preference, and relates to the technical field of human factor engineering, system design and human resource management, and the method comprises the steps: collecting not less than 30 task samples in the same task scene in a target field, obtaining mental load intensity data and a corresponding task error rate, wherein mental load intensity is calculated based on a multi-resource theoretical model; establishing a monotonically increasing mental load intensity-error rate linear regression model in a set interval by adopting a least square method, and verifying the significance of the model through F test or t test; the acceptable maximum error rate input by the manager is substituted into the mental load intensity-error rate linear regression model, and an analytical method is used for inverse solution to obtain a mental load threshold value; and outputting a mental load threshold value for personnel post compilation, task process design or man-machine function distribution. And a scientific and quantitative decision basis is provided for man-machine function distribution, post compilation and process optimization in a system design stage.

Description

Technical Field

[0001] This invention relates to the fields of human factors engineering, system design and human resource management, and in particular to a method and system for determining mental workload thresholds based on managers' risk preferences. Background Technology

[0002] In the design and operation of large and complex human-machine systems (such as ship control, air traffic control, and nuclear power plant operation), rational allocation of human-machine functions, determination of job positions, and optimization of operating procedures are crucial to ensuring system safety and efficiency. The mental workload of operators is one of the core considerations. Mental workload refers to the intensity of psychological effort an individual expends per unit of time to process a specific task. Too low a workload level may lead to decreased operator alertness and idle resources, while too high a workload level can cause distraction, decreased cognitive ability, and consequently, a sharp increase in human error rates, even leading to major safety accidents.

[0003] Currently, there are various methods for assessing and managing mental workload thresholds in existing technologies, mainly including the following two categories: The first category is post-event assessment methods based on subjective scales or physiological measurements. For example, the NASA-TLX (NASA Task Load Index) scale, subjective mental workload assessment techniques, etc., are used to allow operators to subjectively rate their workload after task execution, or physiological indicators such as electroencephalograms (EEGs) and heart rate variability are collected to indirectly reflect the workload level. Existing technologies in this category include Chinese patent application CN117679032A, which discloses a mental workload measurement method based on multiple parameters and subjective scales, but does not involve risk preference quantification and model inverse kinematics; and Chinese patent application CN114305441A, which discloses a threshold inverse kinematics method, but lacks a closed-loop verification mechanism. These methods are mainly used for verification after system development or for monitoring during operation, and cannot provide forward-looking quantitative guidance for function allocation and process design in the early stages of system design.

[0004] The second category is predictive modeling methods based on task analysis. These methods, such as the VCP (Visual, Auditory, Cognitive, Psychomotor) multi-resource theory model, calculate a theoretical mental workload intensity before task execution by decomposing the resource requirements of different cognitive channels for a task. However, these methods have limitations: First, they typically set a static, universal mental workload threshold (e.g., considering an VCP score exceeding a certain fixed value as overload). This method ignores the differentiated requirements for safety and reliability in different task scenarios. For example, in aircraft carrier landings and routine patrol missions, even if the mental workload level is the same, the acceptable risk level is drastically different, making the use of the same threshold unreasonable. Second, existing technologies have failed to establish a quantitative and predictable relationship between mental workload and specific performance risks (such as task error rate). This makes it difficult for system designers and managers to clearly answer key engineering decision questions such as, "To keep the error rate of core tasks below one in a thousand, at what level should the operator's mental workload be designed?"

[0005] Therefore, a long-standing technical challenge faced by those skilled in the art is the lack of a method and system that can be used in the early stages of complex system design to quantify managers' risk preferences into engineering parameters and establish a quantitative predictive model between mental workload and error rate. This leads to human-machine function allocation and job staffing often relying on experience-based judgment, lacking scientific and quantitative decision-making basis, which may create safety hazards and waste human resources. Summary of the Invention

[0006] In view of the above-mentioned problems in the prior art, the present invention provides a method and system for determining the mental workload threshold based on the risk preference of managers, thereby providing a scientific and quantitative basis for human-machine function allocation, job allocation and process optimization in the early design stage of complex systems.

[0007] This invention provides a method for determining a mental workload threshold based on a manager's risk preference, comprising the following steps: S1: Collect no less than 30 task samples in the same task scenario within the target domain, and obtain mental workload intensity data and corresponding task error rates. The mental workload intensity is calculated based on a multi-resource theoretical model. S2: A monotonically increasing linear regression model of mental workload intensity-error rate is established within a set interval using the least squares method, and the significance of the linear regression model of mental workload intensity-error rate is verified by setting a test method. S3: Receive the maximum acceptable error rate input by the administrator; S4: Substitute the acceptable maximum error rate into the mental workload intensity-error rate linear regression model, and use the analytical method to inversely solve for the mental workload threshold. S5: Outputs mental workload threshold, used for personnel job allocation, task flow design, or human-machine function allocation.

[0008] In some embodiments of the present invention, the set interval is [Lmin, Lmax]; The multi-resource theory model performs calculations in four dimensions: visual, auditory, cognitive, and psychomotor. The acceptable maximum error rate is denoted by ε, and ε ∈ [0.1%, 5%].

[0009] In some embodiments of the present invention, the mental workload intensity is specifically the average mental workload intensity during task execution, calculated using a multi-resource theoretical model.

[0010] In some embodiments of the present invention, the mental workload threshold is represented by T, and T = (ε-b) / k; The mental workload intensity-error rate regression model is expressed as follows: Error rate = k × mental workload + b Where k and b are regression coefficients obtained through data fitting. k is the regression coefficient, which means that for every unit increase in mental workload, the error rate changes by an average of k units. In the model, k>0 indicates that mental workload is positively correlated with the error rate. b is the intercept, which represents the theoretical error rate when the mental workload is zero. The linear regression model is fitted within the mental workload range [Lmin, Lmax] of the task samples. The test method is set as either F-test or t-test. The F-test is used to test the significance of the linear regression model of mental workload intensity-error rate, that is, to determine whether mental workload intensity has a significant impact on the error rate as a whole. The t-test is used to test the significance of a single regression coefficient. In validating the mental workload intensity-error rate linear regression model, p < 0.05 and R² ≥ 0.8 were used. Here, p is the probability value in hypothesis testing, representing the probability of observing the current data or more extreme data given the null hypothesis. p < 0.05 indicates a significance level of 5%, meaning the model or coefficient is statistically significant, and the null hypothesis is rejected. R² is the coefficient of determination, representing the degree to which the mental workload intensity-error rate linear regression model explains data variation. R² values ​​are between 0 and 1. R² ≥ 0.8 indicates that the model explains at least 80% of the error rate variation, suggesting a good model fit.

[0011] In some embodiments of the present invention, after step S5, the method further includes a verification step: S6: Calculate the actual mental workload intensity for one or more tasks to be evaluated within the target domain; S7: Compare the actual mental workload intensity with the mental workload threshold. If the threshold is exceeded, generate an early warning message or optimization plan. S8: By optimizing the system design or adjusting the manager's risk preference, iterate until all actual mental workload intensity is less than the threshold, forming a closed-loop optimization.

[0012] In the above embodiments, the determined mental workload threshold can be used for team workload prediction for multi-person collaborative tasks. When performing team workload prediction for multi-person collaborative tasks, the method may include the following steps: Step 1: Using two parallel role experiment platforms, simultaneously collect the identity judgment performance data of the first operator on one role experiment platform and the device allocation performance data of the second operator on the other role experiment platform. With a set sampling rate and a set time alignment error range, simultaneously collect multi-channel physiological signals of the two operators. The multi-channel physiological signals include electroencephalogram (EEG) signals, electrocardiogram (ECG) signals, and eye movement (EMG) signals. The EEG signals are acquired using an 18-lead electrode, the ECG signals are acquired using a chest strap sensor, and the EMG signals are acquired using an eye tracker. The synchronous acquisition is achieved through either hardware or software synchronization. Hardware synchronization is based on the IEEE-1588 PTP protocol with a time alignment error of <3ms; software synchronization is based on heartbeat frame alignment with a time alignment error of <5ms. The set sampling rate is 512 Hz.

[0013] In this embodiment, the identity verification performance data includes reaction time, identity verification time, sending verification time, confirmation feedback time, false negative rate, false positive rate, completion rate, accuracy rate, and feedback click rate; the device allocation performance data includes reaction time, device allocation time, feedback verification time, false negative rate, allocation accuracy rate, completion rate, and accuracy rate.

[0014] Step 2: The identity judgment performance data, the device allocation performance data, and the multi-channel physiological signals are subjected to principal component analysis to reduce dimensionality and generate an 8-dimensional feature vector with a cumulative variance contribution rate not less than a set contribution rate; Step 3: Input the 8-dimensional feature vector into the BP neural network, and refresh the weights online after each trial. The backpropagation neural network has an 8-3-4 network structure, with the hidden layer activation function being logsig, the output layer activation function being purelin, and a learning rate of 0.68.

[0015] The weight refresh of the backpropagation neural network adopts a sliding window adaptive learning rate strategy with a window length of 100 trials. If the verification error increases for 10 consecutive trials, the learning rate is multiplied by 0.9, and the minimum learning rate is no less than 0.1.

[0016] Step 4: The backpropagation neural network outputs the task load, team load, task-team balanced load and total evaluation score in real time, with the average absolute percentage error not exceeding a set threshold, wherein the set threshold is not less than 16.7%.

[0017] When outputting and visualizing the above information, the specific method is as follows: the visualization is displayed in a color output manner, such as outputting a real-time bar chart or pie chart. The determined mental load threshold can be used as the load value and compared with the set color threshold, as follows: when the load value is less than 40, it is displayed in green; when the load value is greater than or equal to 40 and less than or equal to 70, it is displayed in yellow; and when the load value is greater than 70, it is displayed in red.

[0018] The interface data refresh cycle can be set to 1 second, the rendering frame rate is stable at 58fps, and the latency is <200 ms.

[0019] Furthermore, the mental workload threshold determined above can also be used for predicting the mental workload of a single ship operator. When predicting the mental workload of a single ship operator, the method may include the following steps: Acquire ship attitude data, and perform motion interference compensation on the collected electrocardiogram signal based on the attitude data to obtain the low-frequency power index after sway compensation normalization. Simultaneously collect eye movement parameters (blink duration) and EEG parameters, input the low-frequency power index, the eye movement parameter (blink duration), and the EEG parameters into the GMDH nonlinear mapping model, and output a continuous load score of 0-100; Receive the timestamp of the target object occurrence event from the electronic nautical chart and synchronize the timestamp with the physiological data stream; Based on the timestamp, determine the peak load score within the time window; Based on the peak load score, a corresponding warning instruction is generated according to the preset warning conditions, and a data log is recorded to realize closed-loop intervention for mental load exceeding the limit. The warning conditions include a condition threshold determined based on the mental load threshold determined above.

[0020] In this embodiment, the step of generating corresponding early warning instructions based on the peak load score and preset early warning conditions specifically includes: generating a level one early warning instruction when the peak load score continuously exceeds a first threshold for a first predetermined time length; generating a level two early warning instruction when the peak load score continuously exceeds a second threshold higher than the first threshold for a second predetermined time length; and generating a level three early warning instruction when the peak load score continuously exceeds a third threshold higher than the second threshold for a third predetermined time length. The early warning instructions include at least one of sending a task adjustment plan to the task management system, triggering tactile and visual alarms, and automatically controlling ship equipment to reduce operational load. Automatically controlling ship equipment to reduce operational load involves adjusting the radar pulse repetition frequency or autopilot parameters. The first threshold, the second threshold, and the third threshold are all determined based on the aforementioned determined mental load threshold, specifically by increasing or decreasing the value range of the threshold based on the aforementioned determined mental load threshold.

[0021] This invention also provides a system for determining a mental workload threshold based on manager risk preference for implementing the method described in the above embodiments, comprising: The data acquisition module is used to acquire mental workload intensity data and corresponding task error rates for no less than 30 task samples in the same task scenario within the target domain. The mental workload intensity is calculated based on a multi-resource theoretical model. The model building module is used to construct a monotonically increasing linear regression model of mental workload intensity-error rate within a set interval using the least squares method, and to verify the significance of the linear regression model of mental workload intensity-error rate by setting a test method. The input module is used to manage the maximum acceptable error rate for input; The calculation module is used to substitute the acceptable maximum error rate into the mental workload intensity-error rate linear regression model and use the analytical method to solve the mental workload threshold. The output and application module is used to output the mental workload threshold, which is used for personnel job allocation, task process design, or human-machine function allocation.

[0022] In some embodiments of the present invention, the computing module is also used to calculate the actual mental workload intensity of one or more tasks to be evaluated within the target domain.

[0023] In some embodiments of the present invention, the system further includes: The verification module is used to compare the actual mental workload intensity with the mental workload threshold. If the threshold is exceeded, an early warning message or optimization plan is generated.

[0024] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the method for determining mental workload threshold based on manager risk preference as described in the above embodiments.

[0025] This invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. When the processor executes the program, it implements the method for determining the mental workload threshold based on the manager's risk preference as described in the above embodiments.

[0026] Compared with the prior art, the beneficial effects of the method and system for determining mental workload threshold based on manager risk preference provided in this invention include: Quantifying risk appetite: For the first time, the abstract concept of "risk appetite" is transformed into a concrete and calculable "acceptable error rate," making safety management decisions quantifiable and traceable.

[0027] Proactive design tools: Transform mental workload assessment from "post-event measurement" to "pre-event prediction", avoiding the risk of potential overload in the early stages of system design.

[0028] Versatility and Flexibility: By establishing predictive models for different fields, this method can be widely applied to multiple high-risk areas such as shipping, aviation, healthcare, and nuclear power. Furthermore, the threshold can be dynamically adjusted according to the safety requirements of different tasks and periods.

[0029] Improving system safety and efficiency: By scientifically controlling mental workload, the rate of human error can be effectively reduced, the operational safety and efficiency of the entire complex system can be improved, and resource waste caused by overstaffing can be avoided. Attached Figure Description

[0030] Figure 1 A flowchart of a method for determining mental workload threshold based on manager risk preference provided in an embodiment of the present invention; Figure 2 A schematic diagram of the linear regression model of mental workload intensity-error rate in the method for determining mental workload threshold based on manager risk preference provided in the embodiments of the present invention; Figure 3 A block diagram of a system for determining mental workload thresholds based on manager risk preferences, provided in an embodiment of the present invention. Detailed Implementation

[0031] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0032] Various embodiments and features of this application are described herein with reference to the accompanying drawings.

[0033] These and other features of this application will become apparent from the following description of preferred forms of embodiments given as non-limiting examples, with reference to the accompanying drawings.

[0034] It should also be understood that although this application has been described with reference to some specific examples, those skilled in the art can certainly implement many other equivalent forms of this application, which have the features described in the claims and are therefore all within the scope of protection defined herein.

[0035] The above and other aspects, features and advantages of this application will become more apparent when taken in conjunction with the accompanying drawings and in view of the following detailed description.

[0036] Specific embodiments of this application are described below with reference to the accompanying drawings; however, it should be understood that the claimed embodiments are merely examples of this application, which can be implemented in various ways. Well-known and / or repeated functions and structures are not described in detail to ascertain the true intent based on the user's historical operations, and to avoid unnecessary or redundant details that would obscure this application. Therefore, the specific structural and functional details claimed herein are not intended to be limiting, but merely serve as the basis and representative basis for the claims to teach those skilled in the art to use this application in various ways with substantially any suitable detailed structure.

[0037] This specification may use the phrases “in one embodiment,” “in another embodiment,” “in yet another embodiment,” or “in other embodiments,” all of which may refer to one or more of the same or different embodiments according to this application.

[0038] This invention provides a method for determining a mental workload threshold based on a manager's risk preference, comprising the following steps: S1: Collect no fewer than 30 task samples in the same task scenario within the target domain to obtain mental workload intensity data and the corresponding task error rate. The mental workload intensity is calculated based on a multi-resource theory model, which calculates in four dimensions: visual, auditory, cognitive, and psychomotor. That is, the multi-resource theory model is the VCP theory model. The mental workload intensity is specifically the average mental workload intensity during task execution, calculated using the multi-resource theory model.

[0039] That is, data foundation is built through step S1, and mental workload intensity (such as average intensity calculated based on the VCP model) and corresponding performance data (mainly task error rate) of various tasks in the target domain are collected through experimental or historical data.

[0040] S2: A monotonically increasing linear regression model of mental workload intensity-error rate is established within a set interval using the least squares method, and the significance of the linear regression model of mental workload intensity-error rate is verified by a set test method. The set interval is [Lmin, Lmax]. A general mental workload intensity-error rate prediction model in this field is established by using statistical methods (linear regression).

[0041] S3: Receive the maximum acceptable error rate input by the manager, the maximum acceptable error rate is represented by ε, and ε∈[0.1%, 5%]; by inputting risk preferences, the manager sets an acceptable maximum task error rate according to the importance and security requirements of the task, for example, it can be 0.1 or 1%.

[0042] S4: Substitute the acceptable maximum error rate into the mental workload intensity-error rate linear regression model, and use the analytical method to inversely solve for the mental workload threshold. The mental workload threshold is denoted by T, and T = (ε-b) / k; The mental workload intensity-error rate regression model is expressed as follows: Error rate = k × mental workload + b Where k and b are regression coefficients obtained through data fitting. k is the regression coefficient, which means that for every unit increase in mental workload, the error rate changes by an average of k units. In the model, k>0 indicates that mental workload is positively correlated with the error rate. b is the intercept, which represents the theoretical error rate when the mental workload is zero. The linear regression model is fitted within the mental workload range [Lmin, Lmax] of the task samples. The test method is set as either F-test or t-test. The F-test is used to test the significance of the linear regression model of mental workload intensity-error rate, that is, to determine whether mental workload intensity has a significant impact on the error rate as a whole. The t-test is used to test the significance of a single regression coefficient. In validating the mental workload intensity-error rate linear regression model, p < 0.05 and R² ≥ 0.8 were used. Here, p is the probability value in hypothesis testing, representing the probability of observing the current data or more extreme data given the null hypothesis. p < 0.05 indicates a significance level of 5%, meaning the model or coefficient is statistically significant, and the null hypothesis is rejected. R² is the coefficient of determination, representing the degree to which the mental workload intensity-error rate linear regression model explains data variation. R² values ​​are between 0 and 1. R² ≥ 0.8 indicates that the model explains at least 80% of the error rate variation, suggesting a good model fit.

[0043] S5: Outputs a mental workload threshold for personnel staffing, task flow design, or human-machine function allocation. Specifically, this threshold can be used as a design standard to assess whether newly designed task flows and job workloads are excessive. An optimization scheme can be provided by calculating the actual workload of the task to be evaluated and comparing it with the threshold.

[0044] S6: Calculate the actual mental workload intensity for one or more tasks to be evaluated within the target domain; S7: Compare the actual mental workload intensity with the mental workload threshold. If the threshold is exceeded, generate an early warning message or optimization plan. S8: By optimizing the system design or adjusting the manager's risk preference, iterate until all actual mental workload intensity is less than the threshold, forming a closed-loop optimization.

[0045] This invention also provides a system for determining a mental workload threshold based on manager risk preference for implementing the method described in the above embodiments, comprising: The data acquisition module is used to acquire mental workload intensity data and corresponding task error rates for no less than 30 task samples in the same task scenario within the target domain. The mental workload intensity is calculated based on a multi-resource theoretical model. The model building module is used to construct a monotonically increasing linear regression model of mental workload intensity-error rate within a set interval using the least squares method, and to verify the significance of the linear regression model of mental workload intensity-error rate by setting a test method. The input module is used to manage the maximum acceptable error rate for input; The calculation module is used to substitute the acceptable maximum error rate into the mental workload intensity-error rate linear regression model and use the analytical method to solve the mental workload threshold. The output and application module is used to output the mental workload threshold, which is used for personnel job allocation, task process design, or human-machine function allocation.

[0046] In this embodiment, the computing module is also used to calculate the actual mental workload intensity of one or more tasks to be evaluated within the target domain.

[0047] In this embodiment, the system further includes a verification module, used to compare the actual mental workload intensity with the mental workload threshold, and if the threshold is exceeded, generate an early warning message or an optimization plan.

[0048] To facilitate understanding of the above technical solutions, the following detailed explanation, in conjunction with specific examples and accompanying drawings, illustrates their application in ship control system design. 1. Data Foundation Construction: Through combined task experiments, we obtained data on the mental workload (calculated based on task completion time) and corresponding total error rate of ship visual channel tasks (such as perception, localization, and inspection) under different combinations. Specifically, we collected n=100 task samples to ensure data representativeness.

[0049] 2. Model Establishment: Perform linear regression analysis on the above data to establish a model. For example, construct a model where error rate = k × mental workload intensity + b. Data fitting yields: Error rate = 0.03 × mental workload intensity - 0.09 (this is a schematic model; specific coefficients are determined by actual data). The model is only applicable to the intensity ≥ 3 range; goodness of fit R² = 0.85, sample size n = 100. When intensity < 3, the error rate is 0. The model's significance is verified by the F-test (p < 0.05).

[0050] 3. Risk Preference Input: Assume that ship managers require the error rate to be controlled within 1% (i.e. 0.01) for the key task of "avoiding dynamic objectives".

[0051] 4. Threshold calculation: Substitute the acceptable error rate of 0.01 into the model: 0.01 = 0.03 × mental workload intensity - 0.09.

[0052] The inverse solution yields: mental workload intensity = (0.01 + 0.09) / 0.03 ≈ 3.33.

[0053] Therefore, based on a 1% risk tolerance, the mental workload threshold for this task should be set at 3.33.

[0054] 5. Application and Verification: When designing new ship handling procedures, the VCP model is used to predict the load intensity of the pre-designed procedures for the "avoiding dynamic targets" task (see attached diagram). Figure 1 Step S105). If the predicted value is 4.5, which is higher than the threshold of 3.33 (corresponding to the attached...) Figure 1 Step S106), then the system will issue a warning (corresponding to the attached document). Figure 1 Step S107).

[0055] 6. Iterative feedback: Designers can use this feedback to optimize the system design, such as extending task completion time, simplifying operation steps, adding automation assistance, or adjusting risk preferences, until the predicted load intensity drops below the threshold.

[0056] The above embodiments are merely exemplary embodiments of the present invention and are not intended to limit the present invention. The scope of protection of the present invention is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to the present invention within its spirit and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of the present invention.

Claims

1. A method for determining mental workload thresholds based on managers' risk preferences, characterized in that, Includes the following steps: S1: Collect no less than 30 task samples in the same task scenario within the target domain, and obtain mental workload intensity data and corresponding task error rates. The mental workload intensity is calculated based on a multi-resource theoretical model. S2: A monotonically increasing linear regression model of mental workload intensity-error rate is established within a set interval using the least squares method, and the significance of the linear regression model of mental workload intensity-error rate is verified by setting a test method. S3: Receive the maximum acceptable error rate input by the administrator; S4: Substitute the acceptable maximum error rate into the mental workload intensity-error rate linear regression model, and use the analytical method to inversely solve for the mental workload threshold. S5: Outputs mental workload threshold, used for personnel job allocation, task flow design, or human-machine function allocation.

2. The method for determining the mental workload threshold based on manager risk preference according to claim 1, characterized in that, The set interval is [Lmin, Lmax]; The multi-resource theory model performs calculations in four dimensions: visual, auditory, cognitive, and psychomotor. The acceptable maximum error rate is denoted by ε, and ε ∈ [0.1%, 5%].

3. The method for determining the mental workload threshold based on manager risk preference according to claim 1, characterized in that, The mental workload intensity is specifically the average mental workload intensity during task execution, which is calculated using a multi-resource theory model.

4. The method for determining the mental workload threshold based on manager risk preference according to claim 3, characterized in that, The mental workload threshold is denoted by T, and T = (ε - b) / k; The mental workload intensity-error rate regression model is expressed as follows: Error rate = k × mental workload + b Where k and b are regression coefficients obtained through data fitting. k is the regression coefficient, which means that for every unit increase in mental workload, the error rate changes by an average of k units. In the model, k>0 indicates that mental workload is positively correlated with the error rate. b is the intercept, which represents the theoretical error rate when the mental workload is zero. The linear regression model is fitted within the mental workload range [Lmin, Lmax] of the task samples. The test method is set as either F-test or t-test. The F-test is used to test the significance of the linear regression model of mental workload intensity-error rate, that is, to determine whether mental workload intensity has a significant impact on the error rate as a whole. The t-test is used to test the significance of a single regression coefficient. In validating the mental workload intensity-error rate linear regression model, p < 0.05 and R² ≥ 0.8 were used. Here, p is the probability value in hypothesis testing, representing the probability of observing the current data or more extreme data given the null hypothesis. p < 0.05 indicates a significance level of 5%, meaning the model or coefficient is statistically significant, and the null hypothesis is rejected. R² is the coefficient of determination, representing the degree to which the mental workload intensity-error rate linear regression model explains data variation. R² values ​​are between 0 and 1. R² ≥ 0.8 indicates that the model explains at least 80% of the error rate variation, suggesting a good model fit.

5. The method for determining the mental workload threshold based on manager risk preference according to claim 4, characterized in that, Following step S5, the method further includes a verification step: S6: Calculate the actual mental workload intensity for one or more tasks to be evaluated within the target domain; S7: Compare the actual mental workload intensity with the mental workload threshold. If the threshold is exceeded, generate an early warning message or optimization plan. S8: By optimizing the system design or adjusting the manager's risk preference, iterate until all actual mental workload intensity is less than the threshold, forming a closed-loop optimization.

6. A system for determining a mental workload threshold based on manager risk preference for implementing the method as described in any one of claims 1-5, characterized in that, include: The data acquisition module is used to acquire mental workload intensity data and corresponding task error rates for no less than 30 task samples in the same task scenario within the target domain. The mental workload intensity is calculated based on a multi-resource theoretical model. The model building module is used to construct a monotonically increasing linear regression model of mental workload intensity-error rate within a set interval using the least squares method, and to verify the significance of the linear regression model of mental workload intensity-error rate by setting a test method. The input module is used to manage the maximum acceptable error rate for input; The calculation module is used to substitute the acceptable maximum error rate into the mental workload intensity-error rate linear regression model and use the analytical method to solve the mental workload threshold. The output and application module is used to output the mental workload threshold, which is used for personnel job allocation, task process design, or human-machine function allocation.

7. The mental workload threshold determination system based on manager risk preference according to claim 6, characterized in that, The computing module is also used to calculate the actual mental workload intensity for one or more tasks to be evaluated within the target domain.

8. The mental workload threshold determination system based on manager risk preference according to claim 7, characterized in that, The system also includes: The verification module is used to compare the actual mental workload intensity with the mental workload threshold. If the threshold is exceeded, an early warning message or optimization plan is generated.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for determining the mental workload threshold based on the manager's risk preference as described in any one of claims 1 to 5.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that, When the processor executes the program, it implements the method for determining the mental workload threshold based on the manager's risk preference as described in any one of claims 1 to 5.

Images