A risk importance ranking method for flight dispatchers based on hesitant fuzzy

Abstract

The present invention discloses a method for ranking the importance of flight dispatcher risks based on hesitant fuzzy. Based on the risk assessment index system, a hesitant fuzzy matrix is ​​obtained. The hesitant fuzzy matrix represents the severity and possibility of the risk assessment index. By approximating the ideal solution sorting method, based on the hesitant fuzzy matrix, a risk importance ranking is obtained. A dispatcher artificial risk assessment index system is constructed. Secondly, the hesitant fuzzy number is introduced to characterize the uncertainty of the assessment information, the assessment attribute weight is objectively determined, and the risk index ranking result is determined.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a flight dispatcher risk importance ranking method based on hesitant fuzzy. Background Art

[0002] As the core of airline operational risk control, dispatchers are an important part of ensuring aviation safety. They play an irreplaceable and important role in complex flight systems. The human factors of dispatchers directly determine the operational safety status of civil aviation flights. In the actual problem of evaluating the human factors of dispatchers, there are many subjective evaluation factors, and experts often hesitate between several evaluation values. Therefore, it is an important issue to accurately determine the importance of risk assessment indicators for operational safety status. Summary of the invention

[0003] The purpose of the present invention is to provide a flight dispatcher risk importance ranking method based on hesitant fuzzy to solve the above problems existing in the prior art.

[0004] The present application provides a flight dispatcher risk importance ranking method based on hesitant fuzziness, including:

[0005] Acquire data of the dispatcher when executing the flight process to obtain a risk assessment index; the risk assessment index represents an index that may cause flight unsafety;

[0006] Through human factor analysis and classification system model, the risk assessment indicators are divided into four levels: unsafe behavior, premise of unsafe behavior, unsafe supervision and organizational impact, and a risk assessment indicator system is obtained;

[0007] Based on the risk assessment indicator system, a hesitant fuzzy matrix is ​​obtained; the hesitant fuzzy matrix represents the severity and possibility of the risk assessment indicator;

[0008] By approximating the ideal solution ranking method, based on the hesitant fuzzy matrix, the risk importance ranking is obtained.

[0009] Optionally, the risk importance ranking is obtained based on the hesitant fuzzy matrix by approximating the ideal solution ranking method, including:

[0010] Based on multiple risk assessment information entropies, the possibility and severity of the risk assessment indicators are judged to obtain the indicator weight; the indicator weight represents the weight of the risk caused by the risk assessment indicator to the flight among all the risk assessment indicators;

[0011] Based on the indicator weights and the hesitant fuzzy matrix, a hesitant fuzzy positive ideal solution and a hesitant fuzzy negative ideal solution are obtained; the elements in the hesitant fuzzy positive ideal solution represent the maximum evaluation value of each risk assessment indicator; the elements in the hesitant fuzzy negative ideal solution represent the minimum evaluation value of each risk assessment indicator;

[0012] By using the hesitant fuzzy Euclidean distance measure, the distance between each risk assessment indicator and the hesitant fuzzy positive ideal solution and the hesitant fuzzy negative ideal solution is calculated to obtain multiple relative closeness degrees; one relative closeness degree corresponds to one risk assessment indicator;

[0013] According to multiple relative closenesses, the risk assessment indicators are sorted from large to small to obtain the risk importance ranking.

[0014] Optionally, the hesitant fuzzy matrix is ​​obtained based on the risk assessment indicator system, including:

[0015] According to the risk assessment indicator system, the risk of the risk assessment indicator is assessed to obtain a plurality of hesitant fuzzy numbers; one hesitant fuzzy number represents a set of multiple assessment values ​​of the plurality of risk assessment indicators; an element value of the hesitant fuzzy number is a value representing the severity and possibility of the risk of a risk assessment indicator;

[0016] By means of the hierarchical analysis method, the weights of the values ​​in the hesitant fuzzy number are evaluated to obtain an evaluation weight; one evaluation weight corresponds to one hesitant fuzzy number; the evaluation weight represents the accuracy of the evaluation of the risk of the risk assessment indicator;

[0017] A hesitant fuzzy matrix is ​​obtained based on the evaluation weights and a plurality of hesitant fuzzy numbers through a hesitant fuzzy weighted average operator; the hesitant fuzzy matrix is ​​used to normalize the evaluation value.

[0018] Optionally, judging the possibility and severity of risk assessment indicators based on multiple risk assessment information entropies to obtain indicator weights includes:

[0019] Calculate the proportion of each evaluation value of each risk evaluation indicator in the risk evaluation indicator to obtain a plurality of first indicator proportions; one first indicator proportion corresponds to one evaluation value;

[0020] According to the proportions of the multiple first indicators, the proportion of information provided by the risk assessment indicators is detected to obtain risk assessment information entropy; multiple risk assessment information entropies are obtained corresponding to the multiple risk assessment indicators;

[0021] According to the information entropy of multiple risk assessments, the possibility and severity of risk assessment indicators are judged, and the indicator weights are obtained.

[0022] Optionally, obtaining a hesitant fuzzy positive ideal solution and a hesitant fuzzy negative ideal solution based on the indicator weight and the hesitant fuzzy matrix includes:

[0023] Each row of the hesitant fuzzy matrix is ​​segmented to obtain a plurality of risk assessment indicator sets; the risk assessment indicator set includes evaluation values ​​of a plurality of samples corresponding to a risk assessment indicator; and one risk assessment indicator corresponds to one risk assessment indicator set;

[0024] Find the maximum value in the risk assessment index set as the element in the hesitant fuzzy positive ideal solution;

[0025] Multiple risk assessment index sets correspond to multiple elements in the hesitant fuzzy positive ideal solution;

[0026] Find the minimum value in the risk assessment index set as the element in the hesitant fuzzy negative ideal solution;

[0027] Multiple risk assessment indicator sets correspond to obtaining multiple elements in the hesitant fuzzy negative ideal solution.

[0028] Optionally, the distance between each risk assessment indicator and the hesitant fuzzy positive ideal solution and the hesitant fuzzy negative ideal solution is calculated by the hesitant fuzzy Euclidean distance measurement to obtain multiple relative closeness, including:

[0029] Repeatedly add the elements in the hesitant fuzzy number, balance the hesitant fuzzy number with the elements in the hesitant fuzzy positive ideal solution, and obtain the first hesitant fuzzy number;

[0030] By using the hesitant fuzzy Euclidean distance measure, the distance between the hesitant fuzzy positive ideal solution and the first hesitant fuzzy number is determined to obtain a first hesitant fuzzy distance; and a plurality of hesitant fuzzy numbers correspond to a plurality of first hesitant fuzzy distances;

[0031] Finding a first hesitant fuzzy distance smaller than other first hesitant fuzzy distances among a plurality of first hesitant fuzzy distances as an optimal first hesitant fuzzy distance;

[0032] Based on the hesitant fuzzy number, obtaining an optimal second hesitant fuzzy distance;

[0033] The value obtained by dividing the second hesitant fuzzy distance by the optimal second hesitant fuzzy distance minus the value obtained by dividing the first hesitant fuzzy distance by the optimal first hesitant fuzzy distance is taken as the relative closeness;

[0034] Multiple hesitant fuzzy numbers correspond to multiple relative closeness.

[0035] Optionally, the hesitant fuzzy matrix is ​​obtained based on the evaluation weights and a plurality of hesitant fuzzy numbers by using a hesitant fuzzy weighted average operator, including:

[0036] Multiplying the hesitant fuzzy number with the corresponding evaluation weight to obtain a plurality of multiplication values;

[0037] Taking a weighted average of the multiple multiplication values ​​to obtain a plurality of normalized hesitant fuzzy numbers;

[0038] The hesitant fuzzy decision matrix is ​​obtained by superimposing the multiple normalized hesitant fuzzy numbers.

[0039] Optionally, the repeatedly adding elements in the hesitant fuzzy number and balancing the hesitant fuzzy number with the elements in the hesitant fuzzy positive ideal solution to obtain the first hesitant fuzzy number includes:

[0040] Subtract the number of the first hesitant fuzzy elements from the number of elements in the hesitant fuzzy positive ideal solution to obtain the number of replications;

[0041] taking the maximum value of the elements in the hesitant fuzzy number as the first hesitant fuzzy element value;

[0042] Adding a duplicate number of the first hesitant fuzzy element values ​​to the first hesitant fuzzy number.

[0043] Optionally, obtaining the optimal second hesitant fuzzy distance based on the hesitant fuzzy number includes:

[0044] Repeatedly add the elements in the hesitant fuzzy number, balance the hesitant fuzzy number with the elements in the hesitant fuzzy negative ideal solution, and obtain the second hesitant fuzzy number;

[0045] By using the hesitant fuzzy Euclidean distance measure, the distance between the hesitant fuzzy negative ideal solution and the second hesitant fuzzy number is detected to obtain a second hesitant fuzzy distance; a plurality of second hesitant fuzzy distances are obtained corresponding to a plurality of hesitant fuzzy numbers;

[0046] A second hesitant fuzzy distance having a maximum value among the plurality of second hesitant fuzzy distances is taken as the optimal second hesitant fuzzy distance.

[0047] Optionally, the risk assessment index is divided into four levels: unsafe behavior, premise of unsafe behavior, unsafe supervision and organizational impact through human factor analysis and classification system model, and a risk assessment index system is obtained, including:

[0048] The risk assessment indicator system includes the first layer, the second layer and the third layer;

[0049] The first layer includes the risk assessment indicators divided into four levels: unsafe behavior, premise of unsafe behavior, unsafe supervision and organizational impact;

[0050] The second layer includes skill errors, decision errors, cognitive errors, habitual violations and occasional violations corresponding to unsafe behaviors, dispatcher status, environmental factors and dispatch resource management corresponding to the premise of unsafe behaviors, insufficient supervision, inappropriate operation plan, uncorrected problems and supervision violations corresponding to unsafe supervision, and organizational atmosphere, organizational procedures, organizational management and resource management corresponding to organizational influences;

[0051] The third layer includes multiple risk assessment indicators.

[0052] Compared with the prior art, this application achieves the following beneficial effects:

[0053] Because complexity and diversity are the characteristics of civil aviation dispatchers' work, it is far from enough to simply analyze the limitations of dispatchers' own abilities when identifying human risks. Construct hesitant fuzzy numbers to solve risk assessment indicators of uncertainty. Construct a risk assessment indicator system and sort the risk importance, identify explicit errors and invisible errors, and dig out deep organizational reasons from surface behaviors. Determine the causal relationship between risk assessment indicators and the status of each risk assessment indicator in the risk assessment indicator system by detecting the relative relationship between risk assessment indicators. Thereby achieving a more accurate technical effect of sorting the importance of risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 The present invention provides a flowchart of a method for ranking flight dispatcher risk importance based on hesitant fuzzy.

[0055] Figure 2 It is a schematic diagram of obtaining risk assessment indicators in a hesitant fuzzy-based flight dispatcher risk importance ranking method provided in an embodiment of the present invention.

[0056] Figure 3 It is a structural schematic diagram of a risk assessment index system in a hesitant fuzzy-based flight dispatcher risk importance ranking method provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The present invention will be described in detail below in conjunction with the accompanying drawings.

[0058] Example:

[0059] like Figure 1 As shown, an embodiment of the present invention provides a flight dispatcher risk importance ranking method based on hesitant fuzzy, the method comprising:

[0060] S101: Obtain data of the dispatcher when executing the flight process to obtain a risk assessment index. The risk assessment index represents an index that may cause flight unsafety.

[0061] Among them, the risk assessment indicators are as follows: Figure 2 The “Flight Dispatcher Human Risk Assessment Indicators” shown in the figure, such as “Dispatcher’s Insufficient Professional Knowledge Level” is a risk assessment indicator, marked as .

[0062] The risk assessment indicators include habitual and occasional violations due to the status of the dispatcher during shift handover, skill errors, decision errors and cognitive errors caused by environmental factors or inappropriate operation plans when making flight plans, and skill errors, decision errors, cognitive errors, habitual and occasional violations caused by environmental factors or dispatch resource management during dispatch release. Inadequate supervision, uncorrected problems and supervision violations detected under monitoring. Organizational procedures are tested according to manuals and procedures.

[0063] S102: Through human factor analysis and classification system model, the risk assessment indicators are divided into four levels: unsafe behavior, premise of unsafe behavior, unsafe supervision and organizational impact, to obtain a risk assessment indicator system.

[0064] Among them, the human factors analysis and classification system (HFACS) model.

[0065] Among them, the risk assessment indicator system is as follows: Figure 3 shown.

[0066] S103: Based on the risk assessment indicator system, a hesitant fuzzy matrix is ​​obtained; the hesitant fuzzy matrix represents the severity and possibility of the risk assessment indicators.

[0067] S105: Obtain risk importance ranking based on the hesitant fuzzy matrix by approximating the ideal solution ranking method.

[0068] Optionally, the risk importance ranking is obtained based on the hesitant fuzzy matrix by approximating the ideal solution ranking method, including:

[0069] Based on multiple risk assessment information entropies, the possibility and severity of the risk assessment indicators are judged to obtain the indicator weight. The indicator weight represents the weight of the risk caused by the risk assessment indicator to the flight among all the risk assessment indicators.

[0070] The probability indicates the estimated value of the risk, and the severity indicates the severity of the risk. Both the probability and the severity are expressed in numerical values. For example, if the severity is 0.8, it means that the severity of the risk is high.

[0071] Among them, the maximum value of the indicator weight represents the weight corresponding to the risk assessment indicator with the highest risk, and the minimum value of the indicator weight represents the weight corresponding to the risk assessment indicator with the lowest risk.

[0072] Among them, the weight of the indicator representing the highest risk is greater than the weights of other indicators, and the weight of the indicator representing the lowest risk is less than the weights of other indicators.

[0073] Based on the indicator weights and the hesitant fuzzy matrix, a hesitant fuzzy positive ideal solution and a hesitant fuzzy negative ideal solution are obtained. The elements in the hesitant fuzzy positive ideal solution represent the maximum evaluation value of each risk assessment indicator. The elements in the hesitant fuzzy negative ideal solution represent the minimum evaluation value of each risk assessment indicator.

[0074] The maximum evaluation value of each risk assessment indicator indicates a value greater than other evaluation values ​​among multiple evaluation values ​​corresponding to a risk assessment indicator. The minimum evaluation value of each risk assessment indicator indicates a value less than other evaluation values ​​among multiple evaluation values ​​corresponding to a risk assessment indicator.

[0075] By using the hesitant fuzzy Euclidean distance measure, the distance between each hesitant ideal number and the hesitant fuzzy positive ideal solution and the hesitant fuzzy negative ideal solution is calculated to obtain a relative closeness. One relative closeness corresponds to one risk assessment indicator.

[0076] According to multiple relative closenesses, the risk assessment indicators are sorted from large to small to obtain the risk importance ranking.

[0077] Optionally, the hesitant fuzzy matrix is ​​obtained based on the risk assessment indicator system, including:

[0078] According to the risk assessment indicator system, the risk of the risk assessment indicator is assessed to obtain multiple hesitant fuzzy numbers; one hesitant fuzzy number represents a set of multiple evaluation values ​​of the multiple risk assessment indicators; an element value of the hesitant fuzzy number is a value representing the severity and possibility of the risk of a risk assessment indicator.

[0079] Among them, the hesitant fuzzy number describes the degree to which an element belongs to a set with multiple values, rather than the traditional single value description. This provides greater flexibility in dealing with uncertainty and ambiguity, because it can express the hesitation or uncertainty of decision makers or experts in evaluating the degree of membership of elements. In the hesitant fuzzy number, the degree of membership of element t is represented by a set of values ​​in the interval [0,1], rather than a single value. For example, if the degree of membership of element t can be described by {0.2, 0.5, 0.8}, it means that the decision maker holds three different opinions on the degree of membership of t.

[0080] Among them, element t represents the risk assessment index, and the elements in the hesitant fuzzy number are arranged in increasing order.

[0081] In this embodiment, the severity and possibility of the risk are divided into "very low", "low", "medium-low", "medium", "medium-high", "high", and "very high", and the evaluation values ​​corresponding to the severity and possibility of the risk are obtained.

[0082] The weights of the values ​​in the hesitant fuzzy number are evaluated by the hierarchical analysis method to obtain the evaluation weights; one evaluation weight corresponds to one hesitant fuzzy number; the evaluation weights represent the accuracy of the evaluation of the risk of the risk assessment indicator.

[0083] Among them, according to the experience, education level and working tenure of the experts who evaluate the risk assessment indicators, the corresponding evaluation values ​​are obtained to evaluate the evaluation weights.

[0084] Among them, the Analytic Hierarchy Process (AHP) is used. The target layer is the evaluation weight, and the criterion layer includes the expert's experience, education level and job tenure. The judgment matrix of the criterion layer is constructed, and then the evaluation weight is obtained according to the arithmetic average method.

[0085] A hesitant fuzzy matrix is ​​obtained based on the evaluation weights and a plurality of hesitant fuzzy numbers through a hesitant fuzzy weighted average operator; the hesitant fuzzy matrix is ​​used to normalize the evaluation value.

[0086] The hesitant fuzzy matrix represents the normalized expert evaluation opinions.

[0087] Wherein, a column of the hesitant fuzzy matrix represents a hesitant fuzzy number corresponding to a sample, and a row of the hesitant fuzzy number represents a risk assessment index corresponding to the hesitant fuzzy number.

[0088] Optionally, judging the possibility and severity of risk assessment indicators based on multiple risk assessment information entropies to obtain indicator weights includes:

[0089] The proportion of each evaluation value of each risk evaluation indicator in the risk evaluation indicator is calculated to obtain a plurality of first indicator proportions; one first indicator proportion corresponds to one evaluation value.

[0090] Among them, a risk assessment indicator system of a dispatcher during a flight is a sample, and the value of the risk assessment indicator in the risk assessment indicator system corresponding to the sample is the assessment value. Figure 3 In The value of .

[0091] The evaluation value indicates the probability of risk existence according to expert evaluation of the risk assessment indicator.

[0092] In this embodiment, the number of evaluation values ​​corresponding to one risk evaluation indicator is assumed to be m.

[0093] Among them, the first indicator proportion represents the proportion of samples with risks to all samples.

[0094] The calculation method of the first index proportion is as follows:

[0095]

[0096] Wherein, n represents the number of risk assessment indicators; m represents the number of samples; i represents the i-th risk assessment indicator in the risk assessment indicator system; j represents the j-th sample in the risk assessment indicator system;

[0097] represents the evaluation value corresponding to the jth sample under the i-th indicator; Indicates the proportion of the evaluation value corresponding to the jth sample under the i-th risk assessment indicator to the risk assessment indicator

[0098] According to the proportions of the multiple first indicators, the proportion of information provided by the risk assessment indicators is detected to obtain the risk assessment information entropy; and multiple risk assessment information entropies are obtained corresponding to the multiple risk assessment indicators.

[0099] The calculation method of the risk assessment information entropy is as follows:

[0100]

[0101] Wherein, n represents the number of risk assessment indicators; m represents the number of samples; i represents the i-th risk assessment indicator in the risk assessment indicator system; j represents the j-th sample in the risk assessment indicator system; Represents the risk assessment information entropy.

[0102] According to the information entropy of multiple risk assessments, the possibility and severity of risk assessment indicators are judged, and the indicator weights are obtained.

[0103]

[0104] Wherein, n represents the number of risk assessment indicators; i represents the i-th risk assessment indicator in the risk assessment indicator system; is the indicator weight; Represents the risk assessment information entropy.

[0105] Optionally, obtaining a hesitant fuzzy positive ideal solution and a hesitant fuzzy negative ideal solution based on the indicator weight and the hesitant fuzzy matrix includes:

[0106] Each row of the hesitant fuzzy matrix is ​​segmented to obtain multiple risk assessment indicator sets; the risk assessment indicator set includes evaluation values ​​of multiple samples corresponding to one risk assessment indicator; one risk assessment indicator corresponds to one risk assessment indicator set.

[0107] The columns of the hesitant fuzzy matrix represent risk assessment indicators; and the rows of the hesitant fuzzy matrix represent the hesitant fuzzy number corresponding to a risk assessment indicator.

[0108] Find the maximum evaluation value in the risk assessment index set as the element in the hesitant fuzzy positive ideal solution;

[0109] The hesitant fuzzy positive ideal solution is a set.

[0110] Among them, the largest evaluation value in the risk assessment index set is greater than the other evaluation values ​​in the hesitant fuzzy number.

[0111] Multiple risk assessment indicator sets correspond to obtaining multiple elements in the hesitant fuzzy positive ideal solution.

[0112] The number of elements in the hesitant fuzzy positive ideal solution is equal to the number of samples.

[0113] Find the minimum evaluation value in the risk assessment index set as the element in the hesitant fuzzy negative ideal solution.

[0114] The hesitant fuzzy negative ideal solution is a set.

[0115] Multiple risk assessment indicator sets correspond to obtaining multiple elements in the hesitant fuzzy negative ideal solution.

[0116] Optionally, the distance between each hesitant ideal number and the hesitant fuzzy positive ideal solution and the hesitant fuzzy negative ideal solution is calculated by the hesitant fuzzy Euclidean distance measure to obtain multiple relative closeness, including:

[0117] The elements in the hesitant fuzzy number are repeatedly added, and the hesitant fuzzy number is quantitatively balanced with the elements in the hesitant fuzzy positive ideal solution to obtain the first hesitant fuzzy number.

[0118] The distance between the hesitant fuzzy positive ideal solution and the first hesitant fuzzy number is determined by the hesitant fuzzy Euclidean distance measure to obtain a first hesitant fuzzy distance; and a plurality of hesitant fuzzy numbers correspond to a plurality of first hesitant fuzzy distances.

[0119] The first hesitant fuzzy distance is calculated by the following method:

[0120]

[0121] Among them, the represents the distance between the hesitant fuzzy number and the hesitant fuzzy positive ideal solution. Indicates hesitant fuzzy positive ideal solution, represents a hesitant fuzzy number. represents the indicator weight, represents the j-th evaluation value of the hesitant fuzzy number corresponding to the i-th row of the hesitant fuzzy matrix, represents the jth element in the hesitant fuzzy positive ideal solution. In this embodiment, represents the Euclidean distance algorithm, that is .

[0122] Finding a first hesitant fuzzy distance smaller than other first hesitant fuzzy distances among a plurality of first hesitant fuzzy distances as an optimal first hesitant fuzzy distance;

[0123] Based on the hesitant fuzzy number, an optimal second hesitant fuzzy distance is obtained.

[0124] The value obtained by dividing the second hesitant fuzzy distance by the optimal second hesitant fuzzy distance minus the value obtained by dividing the first hesitant fuzzy distance by the optimal first hesitant fuzzy distance is taken as the relative closeness.

[0125] The relative closeness is calculated by the following method:

[0126]

[0127] Among them, the Relative closeness, represents the distance between the hesitant ideal number and the hesitant fuzzy positive ideal solution, represents the distance between the hesitant ideal number and the hesitant fuzzy negative ideal solution, It represents the maximum value of the distance between the hesitant ideal number and the hesitant fuzzy positive ideal solution in multiple samples. It represents the minimum value of the distance between the hesitant ideal number and the hesitant fuzzy positive ideal solution among multiple samples.

[0128] Multiple hesitant fuzzy numbers correspond to multiple relative closeness.

[0129] Optionally, the hesitant fuzzy matrix is ​​obtained based on the multiple evaluation weights and the multiple hesitant fuzzy numbers by using the hesitant fuzzy weighted average operator, including:

[0130] Multiplying the hesitant fuzzy number with the corresponding evaluation weight to obtain a plurality of multiplication values;

[0131] The multiple multiplication values ​​are weighted averaged to obtain multiple normalized hesitant fuzzy numbers.

[0132] The hesitant fuzzy decision matrix is ​​obtained by superimposing the multiple normalized hesitant fuzzy numbers.

[0133] Wherein, the hesitant fuzzy decision matrix is ​​shown in the following method:

[0134]

[0135] in, represents the hesitant fuzzy decision matrix, represents multiple hesitant fuzzy numbers, n represents the number of hesitant fuzzy numbers, represents the hesitant fuzzy number, Represents the evaluation weight, j is a natural number from 1 to n, and represents a subscript.

[0136] Among them, the The calculation method is in, , , represents the number of elements in the hesitant fuzzy number, represents a hesitant fuzzy number, Indicates except A hesitant fuzzy number of Indicates a parameter, with a value ranging from 1 to The real number, , n is a natural number greater than 0, express Contains elements, express Contains elements, represents the number of the element, and H represents a set named H. Add the first hesitant fuzzy element value of the number of copies to the first hesitant fuzzy number

[0137] Optionally, the repeatedly adding elements in the hesitant fuzzy number and balancing the hesitant fuzzy number with the elements in the hesitant fuzzy positive ideal solution to obtain the first hesitant fuzzy number includes:

[0138] Subtract the number of the first hesitant fuzzy elements from the number of elements in the hesitant fuzzy positive ideal solution to obtain the number of replications;

[0139] taking the maximum value of the elements in the hesitant fuzzy number as the first hesitant fuzzy element value;

[0140] Adding a duplicate number of the first hesitant fuzzy element values ​​to the first hesitant fuzzy number.

[0141] Through the above method, considering the conservatism of the dispatcher's risk assessment, the largest element in the hesitant fuzzy number is repeatedly added until the number of elements in the hesitant fuzzy number is the same as the number of elements in another hesitant fuzzy number, which can further increase the accuracy of risk assessment.

[0142] Optionally, obtaining the optimal second hesitant fuzzy distance based on the hesitant fuzzy number includes:

[0143] Repeatedly add the elements in the hesitant fuzzy number, balance the hesitant fuzzy number with the elements in the hesitant fuzzy negative ideal solution, and obtain the second hesitant fuzzy number;

[0144] The distance between the hesitant fuzzy negative ideal solution and the second hesitant fuzzy number is detected by the hesitant fuzzy Euclidean distance measure to obtain the second hesitant fuzzy distance; and multiple second hesitant fuzzy distances are obtained corresponding to multiple hesitant fuzzy numbers.

[0145] The second hesitant fuzzy distance is calculated by the following method:

[0146]

[0147] Among them, the represents the distance between the hesitant ideal number and the hesitant fuzzy negative ideal solution. represents the hesitant fuzzy negative ideal solution, represents the hesitant fuzzy number, represents the indicator weight, represents the j-th evaluation value of the hesitant fuzzy number corresponding to the i-th row of the hesitant fuzzy matrix, represents the jth element in the hesitant fuzzy negative ideal solution. represents the Euclidean distance algorithm, that is .

[0148] A second hesitant fuzzy distance having a maximum value among the plurality of second hesitant fuzzy distances is taken as the optimal second hesitant fuzzy distance.

[0149] The maximum second hesitant fuzzy distance is a second hesitant fuzzy distance that is greater than the other second hesitant fuzzy distances.

[0150] Optionally, the risk assessment index is divided into four levels: unsafe behavior, premise of unsafe behavior, unsafe supervision and organizational impact through human factor analysis and classification system model, and a risk assessment index system is obtained, including:

[0151] The risk assessment indicator system includes the first layer, the second layer and the third layer;

[0152] The first layer includes the risk assessment indicators divided into four levels: unsafe behavior, premise of unsafe behavior, unsafe supervision and organizational impact;

[0153] The second layer includes skill errors, decision errors, cognitive errors, habitual violations and occasional violations corresponding to unsafe behaviors, dispatcher status, environmental factors and dispatch resource management corresponding to the premise of unsafe behaviors, insufficient supervision, inappropriate operation plan, uncorrected problems and supervision violations corresponding to unsafe supervision, and organizational atmosphere, organizational procedures, organizational management and resource management corresponding to organizational influences;

[0154] The third layer includes multiple risk assessment indicators.

[0155] Among them, the risk assessment indicator system is as follows: Figure 3 shown.

Claims

1. A flight dispatcher risk importance ranking method based on hesitant fuzzy, characterized in that: include: Acquire data of the dispatcher when executing the flight process to obtain a risk assessment index; the risk assessment index represents an index that may cause flight unsafety; Through human factor analysis and classification system model, the risk assessment indicators are divided into four levels: unsafe behavior, premise of unsafe behavior, unsafe supervision and organizational impact, and a risk assessment indicator system is obtained; Based on the risk assessment index system, a hesitant fuzzy matrix is ​​obtained; The hesitant fuzzy matrix represents the severity and possibility of risk assessment indicators; By approximating the ideal solution ranking method, based on the hesitant fuzzy matrix, the risk importance ranking is obtained; The method of approximating the ideal solution sorting method, based on the hesitant fuzzy matrix, obtains the risk importance sorting, including: Based on multiple risk assessment information entropies, the possibility and severity of the risk assessment indicators are judged to obtain the indicator weight; the indicator weight represents the weight of the risk caused by the risk assessment indicator to the flight among all the risk assessment indicators; Based on the indicator weights and the hesitant fuzzy matrix, a hesitant fuzzy positive ideal solution and a hesitant fuzzy negative ideal solution are obtained; the elements in the hesitant fuzzy positive ideal solution represent the maximum evaluation value of each risk assessment indicator; the elements in the hesitant fuzzy negative ideal solution represent the minimum evaluation value of each risk assessment indicator; By using the hesitant fuzzy Euclidean distance measure, the distance between each risk assessment indicator and the hesitant fuzzy positive ideal solution and the hesitant fuzzy negative ideal solution is calculated to obtain multiple relative closeness degrees; one relative closeness degree corresponds to one risk assessment indicator; According to multiple relative closeness, the risk assessment indicators are sorted from large to small to obtain the risk importance ranking; Based on the risk assessment index system, a hesitant fuzzy matrix is ​​obtained, including: According to the risk assessment indicator system, the risk of the risk assessment indicator is assessed to obtain a plurality of hesitant fuzzy numbers; one hesitant fuzzy number represents a set of multiple assessment values ​​of the plurality of risk assessment indicators; an element value of the hesitant fuzzy number is a value representing the severity and possibility of the risk of a risk assessment indicator; By means of the hierarchical analysis method, the weights of the values ​​in the hesitant fuzzy number are evaluated to obtain an evaluation weight; one evaluation weight corresponds to one hesitant fuzzy number; the evaluation weight represents the accuracy of the evaluation of the risk of the risk assessment indicator; A hesitant fuzzy matrix is ​​obtained based on the evaluation weights and a plurality of hesitant fuzzy numbers by a hesitant fuzzy weighted average operator; the hesitant fuzzy matrix is ​​used to normalize the evaluation value; The method of judging the possibility and severity of risk assessment indicators based on multiple risk assessment information entropies and obtaining indicator weights includes: Calculate the proportion of each evaluation value of each risk evaluation indicator in the risk evaluation indicator to obtain a plurality of first indicator proportions; one first indicator proportion corresponds to one evaluation value; According to the proportions of the multiple first indicators, the proportion of information provided by the risk assessment indicators is detected to obtain risk assessment information entropy; multiple risk assessment information entropies are obtained corresponding to the multiple risk assessment indicators; According to multiple risk assessment information entropies, the possibility and severity of risk assessment indicators are judged to obtain indicator weights; The hesitant fuzzy Euclidean distance measurement is used to calculate the distance between each risk assessment indicator and the hesitant fuzzy positive ideal solution and the hesitant fuzzy negative ideal solution, and multiple relative closeness degrees are obtained, including: Repeatedly add the elements in the hesitant fuzzy number, balance the hesitant fuzzy number with the elements in the hesitant fuzzy positive ideal solution, and obtain the first hesitant fuzzy number; By using the hesitant fuzzy Euclidean distance measure, the distance between the hesitant fuzzy positive ideal solution and the first hesitant fuzzy number is determined to obtain a first hesitant fuzzy distance; and a plurality of hesitant fuzzy numbers correspond to a plurality of first hesitant fuzzy distances; Finding a first hesitant fuzzy distance smaller than other first hesitant fuzzy distances among a plurality of first hesitant fuzzy distances as an optimal first hesitant fuzzy distance; Based on the hesitant fuzzy number, obtaining an optimal second hesitant fuzzy distance; The value obtained by dividing the second hesitant fuzzy distance by the optimal second hesitant fuzzy distance minus the value obtained by dividing the first hesitant fuzzy distance by the optimal first hesitant fuzzy distance is taken as the relative closeness; Multiple hesitant fuzzy numbers correspond to multiple relative closeness.

2. The hesitant fuzzy-based risk importance ranking method for flight dispatchers according to claim 1, characterized in that: The hesitant fuzzy positive ideal solution and the hesitant fuzzy negative ideal solution are obtained based on the indicator weights and the hesitant fuzzy matrix, including: Each row of the hesitant fuzzy matrix is ​​segmented to obtain a plurality of risk assessment indicator sets; the risk assessment indicator set includes evaluation values ​​of a plurality of samples corresponding to a risk assessment indicator; and one risk assessment indicator corresponds to one risk assessment indicator set; Find the maximum value in the risk assessment index set as the element in the hesitant fuzzy positive ideal solution; Multiple risk assessment index sets correspond to multiple elements in the hesitant fuzzy positive ideal solution; Find the minimum value in the risk assessment index set as the element in the hesitant fuzzy negative ideal solution; Multiple risk assessment indicator sets correspond to obtaining multiple elements in the hesitant fuzzy negative ideal solution.

3. The hesitant fuzzy-based risk importance ranking method for flight dispatchers according to claim 1, characterized in that: The hesitant fuzzy weighted average operator is used to obtain a hesitant fuzzy matrix based on the evaluation weights and a plurality of hesitant fuzzy numbers, including: Multiplying the hesitant fuzzy number with the corresponding evaluation weight to obtain a plurality of multiplication values; Taking a weighted average of the multiple multiplication values ​​to obtain a plurality of normalized hesitant fuzzy numbers; The hesitant fuzzy decision matrix is ​​obtained by superimposing the multiple normalized hesitant fuzzy numbers.

4. The hesitant fuzzy-based risk importance ranking method for flight dispatchers according to claim 1, characterized in that: The repeatedly adding elements in the hesitant fuzzy number and balancing the hesitant fuzzy number with the elements in the hesitant fuzzy positive ideal solution to obtain the first hesitant fuzzy number includes: Subtract the number of the first hesitant fuzzy elements from the number of elements in the hesitant fuzzy positive ideal solution to obtain the number of replications; taking the maximum value of the elements in the hesitant fuzzy number as the first hesitant fuzzy element value; Adding a duplicate number of the first hesitant fuzzy element values ​​to the first hesitant fuzzy number.

5. The hesitant fuzzy-based risk importance ranking method for flight dispatchers according to claim 1, characterized in that: The obtaining of the optimal second hesitant fuzzy distance based on the hesitant fuzzy number comprises: Repeatedly add the elements in the hesitant fuzzy number, balance the hesitant fuzzy number with the elements in the hesitant fuzzy negative ideal solution, and obtain the second hesitant fuzzy number; By using the hesitant fuzzy Euclidean distance measure, the distance between the hesitant fuzzy negative ideal solution and the second hesitant fuzzy number is detected to obtain a second hesitant fuzzy distance; a plurality of second hesitant fuzzy distances are obtained corresponding to a plurality of hesitant fuzzy numbers; A second hesitant fuzzy distance having a maximum value among the plurality of second hesitant fuzzy distances is taken as the optimal second hesitant fuzzy distance.

6. The hesitant fuzzy-based risk importance ranking method for flight dispatchers according to claim 1, characterized in that: Through the human factor analysis and classification system model, the risk assessment indicators are divided into four levels: unsafe behavior, premise of unsafe behavior, unsafe supervision and organizational impact, and a risk assessment indicator system is obtained, including: The risk assessment indicator system includes the first layer, the second layer and the third layer; The first layer includes the risk assessment indicators divided into four levels: unsafe behavior, premise of unsafe behavior, unsafe supervision and organizational impact; The second layer includes skill errors, decision errors, cognitive errors, habitual violations and occasional violations corresponding to unsafe behaviors, dispatcher status, environmental factors and dispatch resource management corresponding to the premise of unsafe behaviors, insufficient supervision, inappropriate operation plan, uncorrected problems and supervision violations corresponding to unsafe supervision, and organizational atmosphere, organizational procedures, organizational management and resource management corresponding to organizational influences; The third layer includes multiple risk assessment indicators.

Images