A flight safety assessment method based on human reliability

Abstract

The present invention discloses a flight safety assessment method based on human reliability, wherein the method comprises: clarifying the flight mission and mission scenario, decomposing the selected mission into a series of subtasks, and determining the subtask execution plan; judging the task category to which the subtask belongs in combination with the subtask characteristics; determining the influencing factors that cause human errors in the subtasks, and determining the weight of each human error influencing factor through a group decision method; establishing a human reliability model based on the mapping relationship between human error influencing factors and human errors, and calculating the human error probability of the subtasks to evaluate flight safety. The present invention comprehensively considers the correlation and difference between human error influencing factors, solves the problem of erroneous estimation caused by insufficient data and over-reliance on expert judgment, ensures the objectivity and correctness of the calculation results, and can provide countermeasures for reducing human errors and improving flight safety.

Description

Technical Field

[0001] The invention belongs to the field of aviation safety assessment, and in particular relates to a flight safety assessment method based on human reliability. Background Art

[0002] With the continuous emergence of advanced avionics technology and the improvement of the reliability of airborne equipment, the incidence of aviation accidents related to the electromechanical systems of civil aircraft has dropped sharply in recent decades. However, the incidence of aviation unsafe events / accidents related to human errors has not been effectively controlled. The aviation accident statistics reported by the International Civil Aviation Organization in 2019 show that the main cause of aviation accidents or catastrophic accidents is human error of pilots, and more than 70% of accidents are directly or indirectly related to human error of pilots. In addition, since air transport volume has been growing steadily and is expected to double in the next few decades, and the increase in air transport volume will inevitably lead to an increase in air traffic density, if reasonable improvement measures are not actively taken, the probability of human error of pilots and flight accidents will also increase. Therefore, the analysis of the human reliability of pilots during mission execution is particularly important.

[0003] In the field of aviation safety assessment, retrospective analysis and assessment of flight safety is usually carried out based on accident statistics. There is no effective method for preventing and controlling human errors that affect flight safety. Therefore, there is an urgent need for a method to quantify the human reliability of aircraft pilots in order to meet the needs of forward-looking analysis and assessment of flight safety, thereby providing important support for reducing human errors and improving flight safety. Summary of the invention

[0004] The present invention aims to provide a flight safety assessment method based on human reliability to solve the problem of insufficient data and over-reliance on expert judgment leading to erroneous estimates, thereby ensuring the objectivity and correctness of the calculation results, providing a forward-looking quantitative analysis method for flight safety assessment, and providing countermeasures for reducing human errors and improving flight safety.

[0005] The present invention provides a flight safety assessment method based on human reliability, comprising the following steps:

[0006] (1) Define the flight mission and mission scenario, conduct detailed analysis and decomposition of the selected mission, obtain a series of subtasks, and determine the subtask execution plan;

[0007] (2) For each subtask, determine the task category to which it belongs and give the corresponding nominal human error probability;

[0008] (3) Select the factors that affect human error on the subtask, determine the effect strength of each factor, and calculate the weight of each factor;

[0009] (4) Clarify the relationship between the factors affecting human errors and human errors in the task execution process, establish a task-oriented aircraft pilot reliability model, calculate the probability of human errors for each subtask, and accurately evaluate flight safety.

[0010] Furthermore, in step (1), task analysis and decomposition are specifically implemented as follows:

[0011] (1.1) Define the flight mission to be analyzed and give the specific mission scenario, including the mission, environment, and cockpit system;

[0012] (1.2) Collect detailed data related to the task being analyzed, including the task steps involved, task sequence, personnel involved, etc., to provide guidance for task decomposition.

[0013] (1.3) The given task is analyzed and decomposed through the tabular task analysis method, and a series of subtasks and their execution plans are obtained in tabular form.

[0014] Furthermore, in step (2), the task category to which the subtask belongs is determined, and the nominal human error probability of the subtask is determined, which is specifically implemented as follows:

[0015] According to the task characteristics, h general task categories are defined, and each task category has a corresponding nominal human error probability (NHEP), denoted as NHEP h , the nominal human error probability refers to the probability of human error under ideal conditions without any external interference; for each subtask after decomposition, the task category matching each subtask is selected from the above h task categories to determine the nominal human error probability of each subtask.

[0016] Furthermore, in step (3), determining the effect intensity of the human error influencing factor and calculating the weight include the following steps:

[0017] (3.1) Define the factors that lead to human error in tasks, and each factor has a corresponding effect strength, S p (·) represents the effect strength of the factor numbered p among all N factors. For a specific subtask, combined with the task scenario, select n (n≤N) factors that cause human errors in the subtask from N human error influencing factors, then the effect strength of the i-th factor among all n human error influencing factors is S p (i).

[0018] (3.2) Construct the judgment matrix A. The specific construction process is as follows:

[0019]

[0020]

[0021]

[0022]

[0023] In the formula, the matrix A (k) Represents the judgment matrix given by the kth expert, the matrix elements represents the relative importance of the influencing factors of the i-th human error compared to the influencing factors of the j-th human error, T(k) represents the credibility of the k-th expert; all experts involved in the construction of the judgment matrix are divided into two groups based on their experience and knowledge, ψ k represents the number of experts in the group to which the kth expert belongs; It represents the sum of the number of experts in the group from the first expert to the group from the hth expert; the matrix A represents the unified judgment matrix that integrates the credibility of different experts.

[0024] (3.3) The judgment matrix A is converted into the fuzzy consistency matrix R. The calculation process is as follows:

[0025]

[0026]

[0027]

[0028] In the formula, r ij is the element of the fuzzy consistent matrix R, r i represents the sum of the elements in the i-th row of the judgment matrix A, r j Represents the sum of the elements in the j-th row of the judgment matrix A.

[0029] (3.4) The Newton square root method is used to calculate the relative weight of each factor affecting human error. The calculation formula is as follows:

[0030]

[0031] In the formula, w i (i=1,2,…,n) represents the weight of the i-th factor among all the influencing factors that lead to human errors in the subtask.

[0032] Furthermore, the aircraft pilot human reliability model expression in step (4) is:

[0033]

[0034] Where HEP is the probability of human error in the subtask, NHEP ish represents the nominal human error probability of the subtask belonging to task category h; S p (i) represents the effect strength of the i-th factor (factor number is p) among all the influencing factors that lead to human errors in the subtask; w i Represents the weight of the i-th factor among all the influencing factors that lead to human errors in the subtask.

[0035] The beneficial effects of the present invention are as follows: the present invention comprehensively considers the correlation and difference between the factors affecting human errors, solves the problem of insufficient data and over-reliance on expert judgment leading to erroneous estimates, ensures the objectivity and correctness of the calculation of the probability of human errors, and further provides a quantitative analysis method for flight safety assessment, and provides countermeasures for reducing human errors and improving flight safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a schematic diagram of the structure of the flight safety assessment method of the present invention;

[0037] Figure 2 A flow chart of a method for calculating weights of human error influencing factors provided by an example of the present invention. DETAILED DESCRIPTION

[0038] Figure 1 This is a structural schematic diagram of a flight safety assessment method based on human reliability proposed in the present invention, including: task analysis and decomposition, judgment of the task category to which subtasks belong, determination of human error influencing factors and weight calculation, human error probability calculation, and flight safety assessment.

[0039] Step 1: Task analysis and decomposition, the specific implementation is as follows:

[0040] S1. Define the flight mission to be analyzed and give the specific mission scenario, including the mission, environment, and cockpit system;

[0041] S2. Collect detailed data related to the analyzed task, including the task steps involved, task sequence, personnel involved, etc., to provide guidance for task decomposition;

[0042] S3. Analyze and decompose the given task through the tabular task analysis method, and display the obtained series of subtasks and their execution plans in tabular form.

[0043] Step 2: Determine the task category to which each subtask belongs after decomposition. The specific implementation content is as follows:

[0044] According to the characteristics of flight missions, h = 8 general task categories are defined, including: completely unfamiliar tasks, tasks that transfer the system to a new state without program guidance, complex tasks that require high skill levels, simple tasks that can be completed quickly with little attention, fast tasks with relatively low skill levels, tasks that require verification to transfer the system to a new state, routine tasks that are completely familiar and highly trained, and tasks with auxiliary decision-making. Each task category has a corresponding nominal human error probability, which can be obtained by looking up the table of the Human Error Assessment and Reduction Technique (HEART) (Table 1). For each subtask after decomposition, the task category that matches each subtask is selected from the above 8 task categories to determine the nominal human error probability of each subtask.

[0045] Table 1 Task categories and corresponding nominal human error probabilities

[0046] No.(h) Task Category Nominal Human Error Probability (NHEP) 1 Completely unfamiliar tasks 0.55 2 The task of transforming a system to a new state without program guidance 0.26 3 Complex tasks requiring a high skill level 0.16 4 Simple tasks that require little attention and can be completed quickly 0.09 5 Quick missions for relatively low skill levels 0.02 6 Tasks that require verification to move the system to a new state 0.003 7 Completely familiar and highly trained routine tasks 0.0004 8 Assist decision-making tasks 0.00002

[0047] Step 3: Determine the effect intensity of human error influencing factors and calculate the weight of each factor. The specific implementation is as follows:

[0048] For a specific subtask, select all the influencing factors that cause human errors in the subtask, a total of n, then the effect strength of the i-th factor numbered p is S p (i) (i = 1, 2, ..., n), the specific value can be determined by looking up the table (Table 2) using the HEART method.

[0049] Table 2 Factors affecting human error and their corresponding effect strengths

[0050]

[0051]

[0052] Figure 2 The flowchart of the method for calculating the weights of human error influencing factors provided by the example of the present invention is for all human error influencing factors of subtasks. Considering the correlation and differences between these factors, and the fact that the judgments of different experts are different due to the complexity of human cognition, the present invention proposes a method for calculating the weights of human error influencing factors based on group decision-making. The implementation steps are as follows:

[0053] P1. Construct the judgment matrix A. The specific construction process is as follows:

[0054]

[0055]

[0056]

[0057]

[0058] In the formula, the matrix A (k) Represents the judgment matrix given by the kth expert, the matrix elements represents the relative importance of the factors affecting the i-th person's error compared to the factors affecting the j-th person's error, T(k) represents the credibility of the k-th expert; all experts involved in the construction of the judgment matrix are divided into two groups based on their experience and knowledge, ψ k represents the number of experts in the group to which the kth expert belongs; represents the sum of the number of experts in the group from the first expert to the group from the hth expert; matrix A represents a unified judgment matrix that integrates the credibility of different experts. In addition, the criteria for constructing the judgment matrix are shown in Table 3.

[0059] Table 3 Criteria for determining the relative importance of matrix elements

[0060]

[0061] P2. Convert the judgment matrix A into the fuzzy consistency matrix R. The calculation process is as follows:

[0062]

[0063]

[0064]

[0065] In the formula, r ij is the element of the fuzzy consistent matrix R, r i represents the sum of the elements in the i-th row of the judgment matrix A, r j Represents the sum of the elements in the j-th row of the judgment matrix A.

[0066] P3. Use Newton's square root method to calculate the relative weight of each factor affecting human error. The calculation formula is as follows:

[0067] In the formula, w i (i=1,2,…,n) represents the weight of the i-th factor among all the influencing factors that lead to human errors in the subtask.

[0068] Step 4: The aircraft pilot human reliability model expression is:

[0069]

[0070] Where HEP is the probability of human error in the subtask, NHEP ish represents the nominal human error probability of the subtask belonging to task category h; S p (i) represents the effect strength of the i-th factor (factor number is p) among all the influencing factors that lead to human errors in the subtask; w i Represents the weight of the i-th factor among all the influencing factors that lead to human errors in the subtask.

[0071] Step 5: Evaluate flight safety based on the calculated human error probabilities of all subtasks, and prevent and control human errors with high probability of occurrence during mission execution to improve flight safety.

[0072] By applying the embodiments of the present invention, a quantitative analysis tool can be provided for flight safety assessment in typical flight mission scenarios, and technical support can be provided for preventing human errors.

[0073] The above is only a preferred example of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A flight safety assessment method based on human reliability, characterized in that: include: (1) Define the flight mission and mission scenario, analyze and decompose the selected mission in detail, obtain a series of subtasks, and determine the subtask execution plan; the mission analysis and decomposition process is as follows: (1.1) Define the flight mission to be analyzed and give the specific mission scenario, including the mission, environment, and cockpit system; (1.2) Collect detailed data related to the analyzed task, including the task steps involved, task sequence, and personnel involved, to provide guidance for task decomposition; (1.3) Analyze and decompose the given task through the tabular task analysis method, and display the obtained series of subtasks and their execution plans in tabular form; (2) For each subtask, determine the task category to which it belongs and give the corresponding nominal human error probability; determine the task category to which the subtask belongs and determine the nominal human error probability of the subtask. The specific implementation is as follows: Based on the task characteristics, define h general task categories. Each task category has a corresponding nominal human error probability, denoted as NHEP h ; For each decomposed subtask, according to the subtask characteristics, select the task category matching each subtask from the above h task categories, so as to determine the nominal human error probability of each subtask; (3) Determine the influencing factors that lead to human errors in subtasks, give the effect strength of each factor, and calculate the weight of each factor respectively; the determination of the effect strength of human error influencing factors and the calculation of the weight of each factor involve the following process: (3.1) Define the factors that lead to human error in tasks, and each factor has a corresponding effect strength, S p (·) represents the effect strength of the factor numbered p among all N factors; For a specific subtask, combined with the task scenario, select n, n≤N factors that cause human errors in the subtask from N human error influencing factors. Then the effect strength of the i-th factor among all n human error influencing factors is S p (i); (3.2) Construct the judgment matrix A. The specific construction process is as follows: In the formula, the matrix A (k) Represents the judgment matrix given by the kth expert, the matrix elements represents the relative importance of the influencing factors of the i-th human error compared to the influencing factors of the j-th human error, T(k) represents the credibility of the k-th expert; all experts involved in the construction of the judgment matrix are divided into two groups based on their experience and knowledge, ψ k represents the number of experts in the group to which the kth expert belongs; represents the sum of the number of experts from the first expert's group to the number of experts from the hth expert's group; the matrix A represents a unified judgment matrix that integrates the credibility of different experts; (3.3) The judgment matrix A is converted into the fuzzy consistency matrix R. The calculation process is as follows: In the formula, r ij is the element of the fuzzy consistent matrix R, r i represents the sum of the elements in the i-th row of the judgment matrix A, r j represents the sum of the elements in the jth row of the judgment matrix A; (3.4) The Newton square root method is used to calculate the relative weight of each factor affecting human error. The calculation formula is as follows: In the formula, w i ,i=1,2,…,n represents the weight of the i-th factor among all the influencing factors that lead to human errors in subtasks; (4) Based on the mapping relationship between human error influencing factors and human errors, a task-oriented aircraft driving human reliability model is established to calculate the probability of human error for each subtask in order to accurately evaluate flight safety. The aircraft driving human reliability model is expressed as: Where HEP is the probability of human error in the subtask, NHEP is h represents the nominal human error probability of the subtask belonging to task category h; S p (i) represents the effect strength of the i-th factor, i.e., factor numbered p, among all the influencing factors that lead to human errors in the subtask; w i Represents the weight of the i-th factor among all the influencing factors that lead to human errors in the subtask.

Images