Improved FMEA method and device based on risk mechanism analysis

Through the improved FMEA method based on risk mechanism analysis, using technical means such as interval intuitive fuzzy numbers and expert weight vectors, the limitations of risk assessment in traditional FMEA methods are solved, the accurate risk ranking of failure mode is achieved, and the accuracy of assessment is improved. degree and reliability.

CN118965963APending Publication Date: 2024-11-15TONGJI UNIV
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410956199.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-17
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The traditional FMEA method has problems that the importance of not being considered when evaluating the risk of failure modes, different combinations may lead to the same RPN value, difficulty in accurately assessing the evaluation factors, and unconsideration of interdependence, resulting in insufficient accuracy of the risk level.

Method used

Using an improved FMEA method based on risk mechanism analysis, the comprehensive risk assessment value is calculated and sorted through interval intuitive fuzzy number matrix, expert weight vector, team risk assessment matrix, center degree and weight calculation, combined with the DEMATEL method and risk factor direct impact matrix. , obtain accurate risk ranking results.

Benefits of technology

Accurate risk ranking of failure modes is achieved, limitations in traditional methods are overcome, and the accuracy and reliability of risk assessment are improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118965963A_ABST
    Figure CN118965963A_ABST
Patent Text Reader

Abstract

The invention provides an improved FMEA method and device based on risk mechanism analysis, and the method is characterized in that the method comprises the following steps: S1, obtaining an interval intuitionistic fuzzy number matrix according to all scores; s2, obtaining a one-dimensional risk assessment matrix according to the risk function and the interval intuitionistic fuzzy number; s3, processing the one-dimensional risk assessment matrix to obtain an expert weight vector; s4, obtaining a team risk assessment matrix according to all the one-dimensional risk assessment matrixes and the expert weight vectors; s5, obtaining the centrality of each risk factor according to the risk factor direct influence matrix; s6, coefficient weights and index weights are obtained according to all the centralities; s7, obtaining a comprehensive risk assessment value according to the coefficient weight, the index weight and the team risk assessment matrix; and S8, obtaining a risk ranking result according to the comprehensive risk assessment value. In a word, the method can obtain an accurate risk ranking result of each failure mode.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of reliability analysis, and in particular to an improved FMEA method and device based on risk mechanism analysis. Background Art

[0002] Failure Mode and Effects Analysis (FMEA) is a systematic, forward-looking risk management technique. It evaluates systems, designs, processes, or services to determine where and how failures might occur, assess the impact of different failure modes, and ultimately identify the most significant failure modes and take actions to eliminate or mitigate their effects. The method was first applied to the design phase of aerospace products in the 1890s. It has since been applied to risk assessments in other fields.

[0003] During the FMEA process, the failure effects, failure causes, and detection measures for each potential failure mode are first listed based on expert knowledge and relevant experience. Then, the FMEA team must determine the assessment of the risk factors Occurrence (Occurrence), Severity (S) and Detection (Detection) (D) for the corresponding failure mode. In the traditional method, FMEA team members estimate the risk level of each failure mode under each risk factor by using a 10-point numerical ranking method. After that, the risk priority number RiskPriorityNumber, or RPN, is calculated for all identified failure modes. It is obtained by multiplying the risk factors O, S, and D. Finally, the failure modes are ranked according to the RPN value. The larger the RPN value, the higher the risk level of the failure mode. However, in practice, traditional FMEA has great limitations. For example, (1) the relative importance of O, S, and D is not considered; (2) different combinations of O, S, and D may have the same RPN value; (3) these three risk factors are difficult to evaluate accurately; (4) the interdependence between various failure modes and effects is not considered; (5) only three risk factors are taken into account, etc.

[0004] In summary, the existing FMEA method has great limitations, and the calculated failure mode risk level is not accurate enough. Summary of the Invention

[0005] The present invention is made to solve the above problems, and its purpose is to provide an improved FMEA method and device based on risk mechanism analysis.

[0006] The present invention provides an improved FMEA method based on risk mechanism analysis, which is used to obtain the risk ranking results of each failure mode based on the scoring data corresponding to each risk factor of multiple failure modes by multiple evaluation experts and the direct influence matrix of the risk factors. The method has the following characteristics, including the following steps: step S1, converting all scoring scores into corresponding interval intuitionistic fuzzy numbers, and obtaining the interval intuitionistic fuzzy number matrix corresponding to each failure mode by each evaluation expert; step S2, for each interval intuitionistic fuzzy number matrix, defuzzifying each interval intuitionistic fuzzy number according to the risk function, and obtaining a one-dimensional risk assessment matrix containing the corresponding one-dimensional risk value; step S3, processing all one-dimensional risk assessment matrices according to the minimization of inconsistency method, and obtaining To the expert weight vector corresponding to each evaluation expert; step S4, according to all one-dimensional risk assessment matrices and expert weight vectors, obtain the team risk assessment matrix; step S5, according to the decision experiment and evaluation experiment method (DEMATEL) method and the risk factor direct impact matrix, obtain the centrality of each risk factor; step S6, according to all centralities, calculate the coefficient weight and index weight; step S7, according to the coefficient weight, index weight and team risk assessment matrix, obtain the comprehensive risk assessment value of each failure mode; step S8, according to the comprehensive risk assessment value from large to small, sort all failure modes to obtain the risk ranking result, wherein, in the risk ranking result, the smaller the serial number of the failure mode, the greater the risk of the failure mode.

[0007] The improved FMEA method based on risk mechanism analysis provided by the present invention may also have the following features: wherein the scoring data is obtained by the evaluation experts scoring the failure mode according to the q-granularity language term set, and in step S1, the interval intuitionistic fuzzy number corresponding to the scoring score i is The calculation expression is: Where sgn(i) is the sign function.

[0008] The improved FMEA method based on risk mechanism analysis provided by the present invention may also have the following characteristics: wherein, in step S2, the calculation expression of the risk function based on the risk interval intuitionistic fuzzy number mechanism analysis is: Where <[a,b],[c,d]> is the risk interval intuitionistic fuzzy number, and F(<[a,b],[c,d]>) is the corresponding one-dimensional risk value.

[0009] The improved FMEA method based on risk mechanism analysis provided by the present invention may also have the following features: wherein, in step S3, the expert weight vector is obtained by solving the nonlinear programming model, and the expression of the nonlinear programming model is: Where l is the total number of evaluation experts, λ * =(λ1,λ2,...,λ l ) is the solution result, including the weight vectors of each expert, is the value corresponding to the i-th row and j-th column in the one-dimensional risk assessment matrix of the k-th evaluation expert. The dimension of the one-dimensional risk assessment matrix is ​​m×n.

[0010] The improved FMEA method based on risk mechanism analysis provided by the present invention may also have the following characteristics: wherein, in step S5, the calculation expression of the centrality is: Where M i is the centrality of the ith risk factor, D i is the influence of the i-th risk factor, C i is the influence degree of the i-th risk factor, t ij is the value corresponding to the i-th row and j-th column in the comprehensive influence matrix T, t ji is the value corresponding to the jth row and ith column in the comprehensive impact matrix T, N is the normalized matrix, O is the direct impact matrix of the risk factor, and o ij The risk factor directly affects the value corresponding to the i-th row and j-th column in the matrix O.

[0011] The improved FMEA method based on risk mechanism analysis provided by the present invention may also have the following characteristics: wherein, in step S6, the calculation expressions of the coefficient weight and the exponential weight are: Where w1 is the index weight, w j (j=2,3,…,n) is the coefficient weight, M j is the centrality of the j-th risk factor.

[0012] The improved FMEA method based on risk mechanism analysis provided by the present invention may also have the following characteristics: wherein, in step S7, the calculation expression of the comprehensive risk assessment value based on the risk factor mechanism analysis is: Where z i is the comprehensive risk assessment value of the i-th failure mode, x ij is the value corresponding to the i-th failure mode and the j-th risk factor in the team risk assessment matrix.

[0013] In the improved FMEA method based on risk mechanism analysis provided by the present invention, it can also have the following characteristics: wherein, the risk factor hierarchical framework includes first-level risk factors and second-level risk factors, and the risk factors are second-level risk factors, including risk factor RF1, risk factor RF2, risk factor RF3, risk factor RF4, risk factor RF5, risk factor RF6, risk factor RF7 and risk factor RF8. The first-level risk factors include severity S, occurrence O, detection D, repair R and correlation C. Risk factor RF1 corresponds to occurrence O, risk factor RF2, risk factor RF3 and risk factor RF4 all correspond to severity S, risk factor RF5 corresponds to detection D, risk factor RF6 and risk factor RF7 all correspond to repair R, and risk factor RF8 corresponds to correlation C.

[0014] The improved FMEA method based on risk mechanism analysis provided by the present invention may also have the following characteristics: wherein, risk factor RF1 indicates the possibility of occurrence of a failure mode, risk factor RF2 indicates the severity of the impact of the failure mode on the production planning and preparation process, risk factor RF3 indicates the severity of the impact of the failure mode on the system task execution process, risk factor RF4 indicates the severity of the impact of the failure mode on the system basic support process, risk factor RF5 indicates the difficulty of detecting the failure mode after it occurs, risk factor RF6 indicates the difficulty of repairing the failure mode due to resource consumption after it is detected, risk factor RF7 indicates the difficulty of repairing the failure mode due to technical requirements after it is detected, and risk factor RF8 indicates the extent to which the failure mode will cause other failure modes.

[0015] The present invention provides an improved FMEA device based on risk mechanism analysis, which is used to obtain the risk ranking results of each failure mode based on the scoring data corresponding to each risk factor of multiple failure modes by multiple evaluation experts and the direct influence matrix of the risk factors. The device has the following characteristics, including: an interval intuitionistic fuzzy number matrix generation module, which is used to convert all scoring scores into corresponding interval intuitionistic fuzzy numbers, and obtain the interval intuitionistic fuzzy number matrix corresponding to each failure mode by each evaluation expert; a one-dimensional risk assessment matrix generation module, which is used to defuzzify each interval intuitionistic fuzzy number matrix according to a risk function, and obtain a one-dimensional risk assessment matrix containing the corresponding one-dimensional risk value; an expert weight vector generation module, which is used to process all one-dimensional risk assessment matrices according to a method of minimizing inconsistency, and obtain to the expert weight vector corresponding to each evaluation expert; a team risk assessment matrix generation module is used to obtain the team risk assessment matrix based on all one-dimensional risk assessment matrices and expert weight vectors; a centrality calculation module is used to obtain the centrality of each risk factor based on the DEMATEL method and the risk factor direct impact matrix; a weight calculation module is used to calculate the coefficient weight and index weight based on all centralities; a comprehensive risk assessment value calculation module is used to obtain the comprehensive risk assessment value of each failure mode based on the coefficient weight, index weight and team risk assessment matrix; a risk ranking result generation module is used to sort all failure modes from large to small according to the comprehensive risk assessment value to obtain the risk ranking result, wherein, in the risk ranking result, the smaller the serial number of the failure mode, the greater the risk of the failure mode.

[0016] Functions and effects of the invention

[0017] According to the improved FMEA method and device based on risk mechanism analysis involved in the present invention, because, first, the score of the failure mode is converted into an interval intuitionistic fuzzy number through the interval intuitionistic fuzzy number matrix generation module; secondly, the interval intuitionistic fuzzy number is defuzzified according to the risk function through the one-dimensional risk assessment matrix generation module to obtain a one-dimensional risk assessment matrix; thirdly, the expert weight vector is obtained according to the minimization of inconsistency method through the expert weight vector generation module; then, the team risk assessment matrix is ​​generated through the team risk assessment matrix generation module; then, the coefficient weight and exponential weight are obtained through the center degree calculation module and the weight calculation module; finally, the comprehensive risk assessment value is calculated and sorted through the comprehensive risk assessment value calculation module and the risk ranking result generation module, thereby obtaining the risk ranking result. Therefore, the improved FMEA method and device based on risk mechanism analysis of the present invention can obtain accurate risk ranking results for each failure mode. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a block diagram of an improved FMEA apparatus in an embodiment of the present invention;

[0019] Figure 2 Schematic diagram of interval-valued intuitionistic fuzzy numbers in a two-dimensional coordinate system in an embodiment of the present invention;

[0020] Figure 3 is a diagram illustrating the mechanism of risk factors in an embodiment of the present invention;

[0021] Figure 4 4 is a flow chart of failure mode risk ranking in an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to make the technical means, creative features, objectives and effects achieved by the present invention easier to understand, the following embodiments and accompanying drawings specifically illustrate the improved FMEA method and device based on risk mechanism analysis of the present invention.

[0023] This embodiment provides an improved FMEA device based on risk mechanism analysis, which is used to obtain risk ranking results of various failure modes based on the scoring data corresponding to various risk factors of multiple failure modes by multiple evaluation experts and the risk factor direct impact matrix.

[0024] The scoring data is obtained by evaluation experts scoring the failure modes according to a q-granularity language term set. In this embodiment, q=13.

[0025] In this embodiment, risk factors are obtained by constructing a risk factor hierarchical framework, which is shown in Table 1.

[0026] Table 1 Risk factor hierarchy framework

[0027]

[0028] The first column in Table 1 lists the first-tier risk factors, including severity S, occurrence O, detection D, repair R, and correlation C. The second column lists the second-tier risk factors, namely risk factors, including risk factor RF1, risk factor RF2, risk factor RF3, risk factor RF4, risk factor RF5, risk factor RF6, risk factor RF7, and risk factor RF8. The third column lists the meanings of each risk factor in the second tier. Among them, risk factor RF1 corresponds to occurrence O, risk factors RF2, risk factor RF3, and risk factor RF4 all correspond to severity S, risk factor RF5 corresponds to detection D, risk factors RF6 and risk factor RF7 both correspond to repair R, and risk factor RF8 corresponds to correlation C.

[0029] In this embodiment, severity S represents the impact of a failure mode on the system, modules, and customers. Occurrence O represents the likelihood of a failure mode occurring. Detectability D represents the ease of detecting a failure mode. Remediability R represents the ease of repairing a failure mode. Correlation C represents the degree to which a failure mode causes other failure modes.

[0030] In this example, the evaluation experts include Manufacturing Execution System (MES) implementation consultant A, information technology consulting expert B, information technology project senior manager C, and university scholar D. An analysis of the MES system yielded 23 failure modes, as shown in Table 2.

[0031] Table 2 Failure mode table

[0032] Serial number Failure Mode <![CDATA[FM1]]> Incorrect delivery of production tasks <![CDATA[FM2]]> Execution progress feedback is incorrect <![CDATA[FM3]]> Production line flexibility did not meet expectations <![CDATA[FM4]]> The workshop sets unreasonable target points <![CDATA[FM5]]> Failure to assign appropriate positions to employees based on their specialties <![CDATA[FM6]]> Unreasonable distribution of job positions and staff numbers <![CDATA[FM7]]> Information error in connection with ERP <![CDATA[FM8]]> The BOM is inaccurate <![CDATA[FM9]]> Data transmission interruption <![CDATA[FM 10 ]]> Inaccurate data collection <![CDATA[FM 11 ]]> Feedback of alarm information is not timely <![CDATA[FM 12 ]]> Returned products and scrap products are not processed in a timely manner <![CDATA[FM 13 ]]> Inefficient or incorrect product quality inspections <![CDATA[FM 14 ]]> Defective products not recovered <![CDATA[FM 15 ]]> Equipment maintenance is not timely <![CDATA[FM 16 ]]> Equipment utilization rate does not meet the standard <![CDATA[FM 17 ]]> Failure of work-in-process tracking based on barcode technology <![CDATA[FM 18 ]]> Materials cannot be delivered accurately <![CDATA[FM 19 ]]> Unreasonable allocation of production resources <![CDATA[FM 20 ]]> Confusing file format <![CDATA[FM 21 ]]> Abnormal file loss <![CDATA[FM 22 ]]> Frequent software errors <![CDATA[FM 23 ]]> Insufficient software functional modules

[0033] In Table 2, the first column is the serial number of each failure mode, and the second column is the failure mode.

[0034] Figure 1 4 is a block diagram of an improved FMEA device based on risk mechanism analysis in an embodiment of the present invention.

[0035] like Figure 1 As shown, the improved FMEA device 100 based on risk mechanism analysis includes an interval intuitionistic fuzzy number matrix generation module 10, a one-dimensional risk assessment matrix generation module 20, an expert weight vector generation module 30, a team risk assessment matrix generation module 40, a centrality calculation module 50, a weight calculation module 60, a comprehensive risk assessment value calculation module 70, a risk ranking result generation module 80 and a control module 90 for controlling the operation of the above modules.

[0036] The interval intuitionistic fuzzy number matrix generation module 10 is used to convert all scoring scores into corresponding interval intuitionistic fuzzy numbers, and obtain the interval intuitionistic fuzzy number matrix corresponding to each failure mode of each evaluation expert.

[0037] In this embodiment, a q-granularity language term set is defined.

[0038] Define the intuitionistic fuzzy set A = { <x,μ A (x),v A (x)>|x∈X}, where X is a non-empty set, <x,μ A (x),v A (x)> is an intuitionistic fuzzy set, μ A (x),v A (x) are the membership and non-membership of element x in X to A: μ A (x):X→[0,1],V A(x): X→[0,1], and satisfy the condition: 0≤μ A (x)+v A (x)≤1,x∈X,π A (x) = 1 - μ A (x)-v A (x) is called the hesitation degree of element x in X with respect to A. Intuitionistic fuzzy numbers are ordered pairs consisting of the membership and non-membership of element x in X with respect to A, which can be abbreviated as <μ,v>, satisfying

[0039] Define interval-valued intuitionistic fuzzy sets Where X is a non-empty set, is an interval-valued intuitionistic fuzzy set, They are the membership interval and non-membership interval of element x in X to A respectively: and they satisfy the conditions: In the formula It is called the hesitation interval of element x in X with respect to A. The interval intuitionistic fuzzy number is an ordered interval pair consisting of the membership interval and non-membership interval of element x in X with respect to A, abbreviated as <[a,b],[c,d]>, satisfying

[0040] Then, the interval intuitionistic fuzzy number corresponding to the score i is The calculation expression is:

[0041]

[0042] Where sgn(i) is the sign function.

[0043] In this embodiment, the interval intuitionistic fuzzy number matrices of the first five failure modes obtained by MES implementation consultant A are shown in Tables 3 and 4.

[0044] Table 3 Interval intuitive fuzzy number matrix of the first five failure modes on risk factors RF1 to RF4 by MES implementation consultant A

[0045]

[0046] The first column in Table 3 shows failure modes FM1 to FM5, and the second to fifth columns respectively show the interval intuitionistic fuzzy numbers of each failure mode on risk factors RF1 to RF4 given by MES implementation consultant A. For example, the cell in the second row and second column indicates that the interval intuitionistic fuzzy number of failure mode FM1 given by MES implementation consultant A on risk factor RF1 is ([0.011, 0.017], [0.975, 0.983]).

[0047] Table 4 Interval intuitive fuzzy number matrix of the first five failure modes on risk factors RF5 to RF8 by MES implementation consultant A

[0048]

[0049]

[0050] The first column in Table 4 shows failure modes FM1 to FM5, and the second to fifth columns respectively show the interval intuitionistic fuzzy numbers of each failure mode on risk factors RF5 to RF8 given by MES implementation consultant A. For example, the cell in the second row and second column indicates that the interval intuitionistic fuzzy number of failure mode FM1 given by MES implementation consultant A on risk factor RF5 is ([0.837, 0.852], [0.133, 0.148]).

[0051] The one-dimensional risk assessment matrix generation module 20 is used to defuzzify each interval intuitionistic fuzzy number matrix according to the risk function to obtain a one-dimensional risk assessment matrix containing the corresponding one-dimensional risk value.

[0052] Figure 2 Schematic diagram of interval-valued intuitionistic fuzzy numbers in a two-dimensional coordinate system in an embodiment of the present invention.

[0053] like Figure 2 As shown, the abscissa of this two-dimensional coordinate system represents membership, and the ordinate represents non-membership. For an intuitionistic fuzzy number <μ,v>, where μ represents membership and v represents non-membership, and μ+v≤1, the intuitionistic fuzzy number <μ,v> can be represented by a point (μ,v) in the two-dimensional coordinate system. For an interval intuitionistic fuzzy number <[a,b],[c,d]>, [a,b] represents the range of membership, and [c,d] represents the range of non-membership, and b+d≤1, the interval intuitionistic fuzzy number <[a,b],[c,d]> can be represented by a rectangular region that is necessarily contained in the triangle enclosed by the origin O, (0,1), and (1,0).

[0054] In FMEA problems, the intuitive fuzzy number <0,1> represents a membership of 0 and a non-membership of 1, indicating minimal risk. Therefore, the point (0,1) is the point with the lowest risk in the coordinate system. The intuitive fuzzy number <1,0> represents a membership of 1 and a non-membership of 0, indicating maximum risk. Therefore, the point (1,0) is the point with the highest risk in the coordinate system. The closer the point (μ,v) corresponding to the intuitive fuzzy number <μ,v> is to (0,1) and the farther it is from (1,0), the lower the risk. The farther the point (μ,v) corresponding to the intuitive fuzzy number <μ,v> is from (0,1) and the closer it is to (1,0), the greater the risk.

[0055] Then, the risk size corresponding to the intuitive fuzzy number <[a,b[,[c,d]> of the risk interval is The calculation expression is:

[0056]

[0057] In the formula Represents the average distance from the point in the rectangular area to (0,1) and (1,0) respectively.

[0058] Due to the size of the risk The analytical form of is too complicated, but considering that the commonly used interval intuitionistic fuzzy numbers satisfy ba<<1 and dc<<1, We can approximate the distance between the point ((a+b) / 2, (c+d) / 2) and (0,1) and (1,0), and we have:

[0059]

[0060] Then, the calculation expression of the risk function is:

[0061]

[0062] Where <[a,b],[c,d]> is the risk interval intuitionistic fuzzy number, and F(<[a,b],[c,d]>) is the corresponding one-dimensional risk value.

[0063] In this embodiment, the calculation results of the one-dimensional risk values ​​of the first five failure modes by MES implementation consultant A are shown in Table 5.

[0064] Table 5 Calculation results of the one-dimensional risk values ​​of the first five failure modes by MES implementation consultant A

[0065] <![CDATA[RF1]]> <![CDATA[RF2]]> <![CDATA[RF3]]> <![CDATA[RF4]]> <![CDATA[RF5]]> <![CDATA[RF6]]> <![CDATA[RF7]]> <![CDATA[RF8]]> <![CDATA[FM1]]> 0.018 0.852 0.027 0.018 0.064 0.055 0.432 0.840 <![CDATA[FM2]]> 0.140 0.840 0.027 0.018 0.064 0.064 0.064 0.791 <![CDATA[FM3]]> 0.791 0.140 0.018 0.009 0.064 0.840 0.840 0.500 <![CDATA[FM4]]> 0.500 0.018 0.140 0.009 0.064 0.009 0.009 0.064 <![CDATA[FM5]]> 0.140 0.500 0.791 0.064 0.064 0.432 0.055 0.064

[0066] The first column in Table 5 lists the failure modes, and the second through ninth columns show the calculated one-dimensional risk values ​​for each failure mode with respect to risk factors RF1 through RF8. For example, the cell in the second row, second column shows that MES Implementation Consultant A's one-dimensional risk value for failure mode FM1 and risk factor RF1 is 0.018.

[0067] The expert weight vector generation module 30 is used to process all one-dimensional risk assessment matrices according to the minimization inconsistency method to obtain the expert weight vector corresponding to each evaluation expert.

[0068] Among them, the expert weight vector is obtained by solving the nonlinear programming model. The expression of the nonlinear programming model is:

[0069]

[0070] Where l is the total number of evaluation experts, λ *=(λ1,λ2,...,λ l ) is the solution result, including the weight vectors of each expert, is the value corresponding to the i-th row and j-th column in the one-dimensional risk assessment matrix of the k-th evaluation expert. The dimension of the one-dimensional risk assessment matrix is ​​m×n.

[0071] In this embodiment, the expert weight vectors of MES implementation consultant A, information consulting expert B, information project senior manager C and university scholar D are calculated by the expert weight vector generation module 30 as λ * =(0.359,0.340,0.133,0.168).

[0072] The team risk assessment matrix generation module 40 is used to obtain a team risk assessment matrix according to all one-dimensional risk assessment matrices and expert weight vectors.

[0073] In this embodiment, the team risk assessment matrix [x ij ] m×n middle:

[0074]

[0075] Where m is the total number of failure modes, n is the total number of risk factors, l is the total number of evaluation experts, and λ k is the expert weight corresponding to the kth evaluation expert in the expert weight vector, is the one-dimensional risk value of the kth evaluation expert for the i-th failure mode and the j-th risk factor, x ij is the value corresponding to the i-th failure mode and the j-th risk factor in the team risk assessment matrix.

[0076] The team risk assessment matrix in this embodiment is shown in Table 6.

[0077] Table 6 Team Risk Assessment Matrix

[0078] <![CDATA[x ij ]]> <![CDATA[RF1]]> <![CDATA[RF2]]> <![CDATA[RF3]]> <![CDATA[RF4]]> <![CDATA[RF5]]> <![CDATA[RF6]]> <![CDATA[RF7]]> <![CDATA[RF8]]> <![CDATA[FM1]]> 0.180 0.602 0.284 0.364 0.183 0.156 0.199 0.763 <![CDATA[FM2]]> 0.366 0.593 0.263 0.374 0.076 0.159 0.069 0.748 <![CDATA[FM3]]> 0.442 0.612 0.318 0.322 0.349 0.656 0.608 0.547 <![CDATA[FM4]]> 0.438 0.478 0.506 0.411 0.453 0.238 0.129 0.393 <![CDATA[FM5]]> 0.199 0.697 0.775 0.558 0.208 0.645 0.309 0.526 <![CDATA[FM6]]> 0.383 0.597 0.487 0.444 0.156 0.752 0.414 0.467 <![CDATA[FM7]]> 0.381 0.709 0.394 0.654 0.527 0.715 0.265 0.531 <![CDATA[FM8]]> 0.256 0.559 0.571 0.328 0.139 0.288 0.035 0.330 <![CDATA[FM9]]> 0.324 0.434 0.315 0.351 0.224 0.545 0.373 0.528 <![CDATA[FM 10 ]]> 0.213 0.761 0.562 0.844 0.560 0.739 0.607 0.412 <![CDATA[FM 11 ]]> 0.589 0.788 0.612 0.579 0.359 0.688 0.078 0.375 <![CDATA[FM 12 ]]> 0.445 0.461 0.104 0.431 0.562 0.465 0.289 0.283 <![CDATA[FM 13 ]]> 0.346 0.381 0.453 0.350 0.583 0.398 0.220 0.303 <![CDATA[FM 14 ]]> 0.410 0.583 0.417 0.642 0.300 0.418 0.277 0.430 <![CDATA[FM 15 ]]> 0.490 0.643 0.541 0.385 0.566 0.414 0.348 0.521 <![CDATA[FM 16 ]]> 0.458 0.058 0.053 0.006 0.565 0.317 0.200 0.009 <![CDATA[FM 17 ]]> 0.062 0.331 0.679 0.193 0.384 0.207 0.372 0.420 <![CDATA[FM 18 ]]> 0.252 0.703 0.604 0.180 0.114 0.311 0.283 0.668 <![CDATA[FM 19 ]]> 0.710 0.491 0.352 0.018 0.438 0.651 0.393 0.230 <![CDATA[FM 20 ]]> 0.160 0.030 0.288 0.007 0.032 0.025 0.004 0.030 <![CDATA[FM 21 ]]> 0.150 0.361 0.544 0.151 0.156 0.026 0.060 0.480 <![CDATA[FM 22 ]]> 0.413 0.582 0.449 0.077 0.153 0.273 0.084 0.459 <![CDATA[FM 23 ]]> 0.582 0.560 0.454 0.176 0.158 0.613 0.223 0.204

[0079] The first column in Table 6 lists the failure modes, and the second through ninth columns list the corresponding values ​​for each failure mode and risk factors RF1 through RF8 in the team risk assessment matrix. For example, the cell in the second row and second column indicates that the value corresponding to failure mode FM1 and risk factor RF1 in the team risk assessment matrix is ​​0.180.

[0080] The centrality calculation module 50 is used to obtain the centrality of each risk factor according to the DEMATEL method and the risk factor direct impact matrix.

[0081] The calculation expression of centrality is:

[0082] Mi =D i +C i ,

[0083]

[0084] Where M i is the centrality of the ith risk factor, D i is the influence of the i-th risk factor, C i is the influence degree of the i-th risk factor, t ij is the value corresponding to the i-th row and j-th column in the comprehensive influence matrix T, t ji is the value corresponding to the jth row and ith column in the comprehensive impact matrix T, N is the normalized matrix, O is the direct impact matrix of the risk factor, and o ij The risk factor directly affects the value corresponding to the i-th row and j-th column in the matrix O.

[0085] In this embodiment, the risk factor direct impact matrix is ​​described using a five-level language, 0 to 4, as shown in Table 7.

[0086] Table 7 Direct impact matrix of risk factors described using five-level language (0-4)

[0087]

[0088]

[0089] The value of each cell in Table 7 represents the degree of direct impact of the horizontal risk factor on the vertical risk factor, and the greater the impact, the larger the corresponding value.

[0090] The weight calculation module 60 is used to calculate the coefficient weight and the exponential weight according to all the centralities.

[0091] Among them, the calculation expressions of coefficient weight and exponential weight are:

[0092]

[0093] Where w1 is the index weight, w j (j=2,3,…,n) is the coefficient weight, M j is the centrality of the j-th risk factor.

[0094] In this embodiment, the calculation results of the centrality and weight of each risk factor, namely the coefficient weight and index weight, are shown in Table 8.

[0095] Table 8 Centrality and weight of each risk factor

[0096] <![CDATA[RF j ]]> <![CDATA[RF1]]> <![CDATA[RF2]]> <![CDATA[RF3]]> <![CDATA[RF4]]> <![CDATA[RF5]]> <![CDATA[RF6]]> <![CDATA[RF7]]> <![CDATA[RF8]]> Centrality 3.835 8.710 8.158 7.647 5.544 8.323 4.447 7.850 Weight 0.530 0.172 0.161 0.151 0.109 0.164 0.088 0.155

[0097] In Table 8, the first row contains the risk factors, the second row contains the corresponding centrality, and the third row contains the corresponding weight. For example, the cell in the second row and second column indicates that the centrality of risk factor RF1 is 3.835.

[0098] The comprehensive risk assessment value calculation module 70 is used to obtain the comprehensive risk assessment value of each failure mode according to the coefficient weight, the index weight and the team risk assessment matrix.

[0099] Figure 3 2 is a diagram illustrating the mechanism of risk factors in an embodiment of the present invention.

[0100] like Figure 3 As shown, the occurrence of a failure mode is a prerequisite or condition for the failure mode to trigger a risk. If the failure mode does not occur, no risk is triggered. The severity of a failure mode is a direct reflection of the failure mode risk. Severity 1, Severity 2, and Severity 3 correspond to risk factors RF2, RF3, and RF4, respectively. Detection and remediation of the failure mode ensure that the failure mode no longer triggers risks. Detection corresponds to risk factor RF5, and remediation levels 1 and 2 correspond to risk factors RF6 and RF7, respectively. After a failure mode occurs, it must be detected and remediated to ensure the proper functioning of the MES. The correlation of failure modes is an indirect reflection of the risk triggered by a failure mode. Correlation corresponds to risk factor RF8. After a failure mode occurs, its correlation may trigger the occurrence of other failure modes, which in turn trigger risks.

[0101] Therefore, through the mechanism analysis of the above risk factors, the calculation expression of the comprehensive risk assessment value is:

[0102]

[0103] Where z i is the comprehensive risk assessment value of the i-th failure mode, x ij is the value corresponding to the i-th failure mode and the j-th risk factor in the team risk assessment matrix.

[0104] In this embodiment, based on the above weights, the calculation expression for the comprehensive risk assessment value is:

[0105] CRAV i =x i1 0.530 ×(0.172x i2 +0.161x i3 +0.151x i4 +0.109x i5 +0.164x i6 +0.088xi7 +0.155x i8 ).

[0106] The risk ranking result generating module 80 is used to sort all failure modes from large to small according to the comprehensive risk assessment value to obtain a risk ranking result.

[0107] Among them, in the risk ranking results, the smaller the failure mode number, the greater the risk of the failure mode.

[0108] The control module 90 stores a control program for controlling the operation of each module.

[0109] The following describes the process of performing failure mode risk ranking using the improved FMEA device 100 based on risk mechanism analysis in conjunction with the accompanying drawings.

[0110] Figure 4 4 is a flow chart of failure mode risk ranking in an embodiment of the present invention.

[0111] like Figure 4 As shown, the improved FMEA device 100 based on risk mechanism analysis is used to perform failure mode risk ranking, including the following steps:

[0112] In step S1, the interval intuitionistic fuzzy number matrix generation module 10 is used to convert all the scoring scores into corresponding interval intuitionistic fuzzy numbers, and obtain the interval intuitionistic fuzzy number matrix corresponding to each failure mode by each evaluation expert.

[0113] In step S2, the one-dimensional risk assessment matrix generation module 20 is used to defuzzify each interval intuitionistic fuzzy number matrix according to the risk function to obtain a one-dimensional risk assessment matrix containing the corresponding one-dimensional risk value.

[0114] In step S3, the expert weight vector generation module 30 is used to process all one-dimensional risk assessment matrices according to the minimization inconsistency method to obtain the expert weight vector corresponding to each evaluation expert.

[0115] In step S4, the team risk assessment matrix generation module 40 is used to obtain a team risk assessment matrix based on all one-dimensional risk assessment matrices and expert weight vectors.

[0116] Step S5: Using the centrality calculation module 50 to obtain the centrality of each risk factor according to the DEMATEL method and the risk factor direct impact matrix.

[0117] Step S6: The weight calculation module 60 is used to calculate the coefficient weight and the index weight according to all the centralities.

[0118] In step S7 , the comprehensive risk assessment value calculation module 70 is used to obtain the comprehensive risk assessment value of each failure mode according to the coefficient weight, the index weight and the team risk assessment matrix.

[0119] In step S8, the risk ranking result generating module 80 is used to sort all failure modes from large to small according to the comprehensive risk assessment value to obtain a risk ranking result.

[0120] In this embodiment, the failure mode risk ranking is performed according to the existing traditional FMEA method and the improved FMEA device 100 based on risk mechanism analysis, that is, the ranking results of 23 failure modes by the improved FMEA method are shown in Table 9.

[0121] Table 9 Ranking results of failure modes by improved FMEA method and traditional FMEA method

[0122]

[0123]

[0124] The first column in Table 9 shows the failure modes. The second to sixth columns show the occurrence O, severity S, detection D, RPN and ranking of each failure mode calculated under the traditional FMEA method. The seventh to eighth columns show the CRVA and ranking of the improved FMEA method.

[0125] It can be seen that the improved FMEA method and the traditional FMEA method give the same failure modes with the highest and lowest risk rankings, which are FM 19 and FM 21 For FM1, FM2, FM4, FM6, FM 10 ,FM 11 ,FM 13 ,FM 15 ,FM 19 and FM 21 The risk rankings derived from the improved FMEA method are very similar to those from the traditional FMEA method. The rankings for the same failure mode differ by no more than 2, accounting for 43% of the total failure modes. This demonstrates the effectiveness of the improved FMEA method. The rankings for the remaining failure modes vary significantly. The shortcomings of the traditional FMEA method can largely explain these differences in failure mode risk rankings.

[0126] For example, FM 10 and FM 17 With the same RPN value 96, FM5 and FM 23Have the same RPN value of 45. That is to say, in the traditional FMEA method, multiple failure modes with different combinations of O, S, and D have the same RPN value, while this problem will not be encountered when using the improved FMEA method. According to the improved FMEA method, FM 10 and FM 17 are successfully distinguished, FM 17 has less risk and a lower ranking, and FM 10 has greater risk and a higher ranking. FM5 and FM 23 are also similarly distinguished in the improved FMEA method.

[0127] For another example, for FM8 and FM9, the risk rankings given by the traditional FMEA method are 11 and 21, while the risk rankings given by the improved FMEA method are 16 and 15. In the traditional FMEA method, the evaluation values of the three risk factors O, S, and D for FM8 are 2, 9, and 5 respectively; the evaluation values of the three risk factors O, S, and D for FM9 are 2, 7, and 1 respectively. In the improved FMEA method, the one-dimensional risk evaluation values of the eight risk factors for FM8 are 0.073, 0.897, 0.879, 0.805, 0.434, 0.265, 0.049, 0.811 respectively; the one-dimensional risk evaluation values of the eight risk factors for FM9 are 0.071, 0.899, 0.552, 0.597, 0.024, 0.746, 0.639, 0.888 respectively. The first five risk factors correspond to the three risk factors of the traditional FMEA. It can be seen that if only the first five risk factors are considered, the relationship of risk magnitude should be the same: FM8 > FM9; but for the added risk factors, the risk magnitude of FM9 is greater than that of FM8, especially the 6th and 7th risk factors, that is, the two sub-factors of the repair degree, which results in the relationship of risk magnitude according to the improved FMEA method being: FM8 < FM9. Similarly, the relative change in the risk rankings between FM 12 and FM 18 , FM 14 and FM 16 can be explained according to the above idea. This shows that after adding risk factors, the improved FMEA method can fully consider the risk information of failure modes, thus giving a more reasonable and accurate risk ranking.

[0128] Functions and effects of the embodiments [[ID=2,6]]

[0129] According to the improved FMEA method and device based on risk mechanism analysis involved in this embodiment, first, the failure mode score is converted into an interval intuitionistic fuzzy number through the interval intuitionistic fuzzy number matrix generation module; second, the interval intuitionistic fuzzy number is defuzzified according to the risk function through the one-dimensional risk assessment matrix generation module to obtain a one-dimensional risk assessment matrix; third, the expert weight vector is obtained by the expert weight vector generation module according to the minimum inconsistency method; then, the team risk assessment matrix is ​​generated through the team risk assessment matrix generation module; then, the coefficient weight and exponential weight are obtained through the center calculation module and the weight calculation module; finally, the comprehensive risk assessment value is calculated and ranked through the comprehensive risk assessment value calculation module and the risk ranking result generation module to obtain the risk ranking result. In summary, this method can obtain accurate risk ranking results for each failure mode.

[0130] Those skilled in the art will appreciate that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An improved FMEA method based on risk mechanism analysis is used to obtain the risk ranking results of each failure mode according to the scoring data corresponding to each risk factor of multiple failure modes by multiple evaluation experts and the direct impact matrix of the risk factors, characterized in that: The following steps are involved: Step S1, converting all the scoring scores into corresponding interval intuitionistic fuzzy numbers to obtain the interval intuitionistic fuzzy number matrix corresponding to each failure mode by each evaluation expert; Step S2, for each of the interval intuitionistic fuzzy number matrices, defuzzifying each of the interval intuitionistic fuzzy numbers according to the risk function to obtain a one-dimensional risk assessment matrix containing a corresponding one-dimensional risk value; Step S3, processing all the one-dimensional risk assessment matrices according to the method of minimizing inconsistency, and obtaining the expert weight vector corresponding to each of the evaluation experts; Step S4, obtaining a team risk assessment matrix according to all the one-dimensional risk assessment matrices and the expert weight vectors; Step S5, obtaining the centrality of each risk factor according to the DEMATEL method and the risk factor direct impact matrix; Step S6, calculating the coefficient weight and the index weight according to all the centralities; Step S7, obtaining a comprehensive risk assessment value of each failure mode according to the coefficient weight, the index weight and the team risk assessment matrix; Step S8, sorting all the failure modes from large to small according to the comprehensive risk assessment value to obtain the risk ranking result, Among them, in the risk ranking result, the smaller the sequence number of the failure mode is, the greater the risk of the failure mode is.

2. The improved FMEA method based on risk mechanism analysis according to claim 1, characterized in that: in, The scoring data is obtained by the evaluation expert scoring the failure mode according to the q-granularity language term set, In step S1, the interval intuitionistic fuzzy number corresponding to the score i The calculation expression is: Where sgn(i) is the sign function.

3. The improved FMEA method based on risk mechanism analysis according to claim 1 is characterized in that: in, In step S2, the calculation expression of the risk function based on the risk interval intuitive fuzzy number mechanism analysis is: Where <[a,b],[c,d]> is the intuitive fuzzy number of the risk interval, and F(<[a,b],[c,d]>) is the corresponding one-dimensional risk value.

4. The improved FMEA method based on risk mechanism analysis according to claim 1, characterized in that: in, In step S3, the expert weight vector is obtained by solving a nonlinear programming model. The expression of the nonlinear programming model is: Where l is the total number of evaluation experts, λ * =(λ1,λ2,...,λ l ) is the solution result, including the weight vectors of each expert, is the value corresponding to the i-th row and j-th column in the one-dimensional risk assessment matrix of the k-th evaluation expert. The dimension of the one-dimensional risk assessment matrix is ​​m×n.

5. The improved FMEA method based on risk mechanism analysis according to claim 1 is characterized in that: in, In step S5, the calculation expression of the centrality is: M i =D i +C i , Where M i is the centrality of the ith risk factor, D i is the influence of the ith risk factor, C i is the influence degree of the i-th risk factor, t ij is the value corresponding to the i-th row and j-th column in the comprehensive influence matrix T, t ji is the value corresponding to the jth row and the ith column in the comprehensive impact matrix T, N is the normalized matrix, O is the direct impact matrix of the risk factor, o ij is the value corresponding to the i-th row and j-th column in the risk factor directly affecting the matrix O.

6. The improved FMEA method based on risk mechanism analysis according to claim 1 is characterized in that: in, In step S6, the calculation expressions of the coefficient weight and the exponential weight are: Where w1 is the index weight, w j (j=2,3,…,8) is the coefficient weight, M j is the centrality of the j-th risk factor.

7. The improved FMEA method based on risk mechanism analysis according to claim 6 is characterized in that: in, In step S7, the calculation expression of the comprehensive risk assessment value based on the risk factor mechanism analysis is: Where z i is the comprehensive risk assessment value of the ith failure mode, x ij is the value corresponding to the i-th failure mode and the j-th risk factor in the team risk assessment matrix.

8. The improved FMEA method based on risk mechanism analysis according to claim 1, characterized in that: in, The risk factor hierarchy framework includes the first-level risk factors and the second-level risk factors. The risk factors are the second risk factors, including risk factor RF1, risk factor RF2, risk factor RF3, risk factor RF4, risk factor RF5, risk factor RF6, risk factor RF7 and risk factor RF8, The first-level risk factors include severity S, occurrence O, detection D, repair R and correlation C. The risk factor RF1 corresponds to the occurrence degree O, The risk factor RF2, the risk factor RF3 and the risk factor RF4 all correspond to the severity S, The risk factor RF5 corresponds to the detection degree D, The risk factor RF6 and the risk factor RF7 both correspond to the repair degree R, The risk factor RF8 corresponds to the correlation degree C.

9. The improved FMEA method based on risk mechanism analysis according to claim 8, characterized in that: in, The risk factor RF1 represents the possibility of the failure mode occurring. The risk factor RF2 indicates the severity of the impact of the failure mode on the production planning and preparation process. The risk factor RF3 indicates the severity of the failure mode affecting the execution process of the system task. The risk factor RF4 indicates the severity of the failure mode that the system foundation support process is affected by. The risk factor RF5 indicates the difficulty of detecting the failure mode after it occurs. The risk factor RF6 represents the difficulty of repair due to resource consumption after the failure mode is detected. The risk factor RF7 indicates the difficulty of repair due to technical requirements after the failure mode is detected. The risk factor RF8 indicates the extent to which a failure mode may cause other failure modes.

10. An improved FMEA device based on risk mechanism analysis, used to obtain the risk ranking results of each failure mode according to the scoring data corresponding to each risk factor of multiple failure modes by multiple evaluation experts and the direct impact matrix of the risk factors, characterized in that: include: An interval intuitionistic fuzzy number matrix generation module is used to convert all the scoring scores into corresponding interval intuitionistic fuzzy numbers to obtain the interval intuitionistic fuzzy number matrix corresponding to each failure mode by each evaluation expert; A one-dimensional risk assessment matrix generation module is used to defuzzify each of the interval intuitionistic fuzzy number matrices according to the risk function to obtain a one-dimensional risk assessment matrix containing the corresponding one-dimensional risk value; An expert weight vector generation module is used to process all the one-dimensional risk assessment matrices according to the method of minimizing inconsistency to obtain the expert weight vector corresponding to each of the evaluation experts; A team risk assessment matrix generation module, used to obtain a team risk assessment matrix according to all the one-dimensional risk assessment matrices and the expert weight vectors; A centrality calculation module, used for obtaining the centrality of each risk factor according to the DEMATEL method and the direct impact matrix of the risk factor; A weight calculation module, used to calculate the coefficient weight and the exponential weight according to all the centralities; A comprehensive risk assessment value calculation module, used to obtain a comprehensive risk assessment value of each failure mode according to the coefficient weight, the index weight and the team risk assessment matrix; A risk ranking result generating module is used to sort all the failure modes from large to small according to the comprehensive risk assessment value to obtain the risk ranking result. Among them, in the risk ranking result, the smaller the sequence number of the failure mode is, the greater the risk of the failure mode is.

Citation Information

Cited By

  • Improved FMEA device considering uncertainty and failure mode correlation

    CN119989859A