Failure mode risk level measure priority design method

Abstract

The invention discloses a failure mode risk level measure priority design method, which comprises the following steps: comprehensively describing the risk of a failure mode by adopting three risk factors, namely a severity factor S, a frequency factor O and a detectivity factor D, delimiting the risk level, constructing a risk matrix of the failure mode and defining a risk index, and determining the risk level of the failure mode, designing the corresponding AP according to the risk level, designing the AP in the failure mode by applying the design method and the process provided by the invention, and revealing the design rule contained in the design result. According to the method, the FMEA technology is better applied to carry out risk assessment, so that the method can be used as a theoretical basis for AP design and improvement and can also help a user to know the design process of the AP more clearly, and thus a better application effect is obtained.

Description

Technical Field

[0001] The present invention relates to a measure priority (AP) design method for failure mode risk level, which is applicable to the process of carrying out failure mode risk assessment using a failure mode and effects analysis (FMEA) tool, and in particular to a measure priority design method for failure mode risk level. Background Art

[0002] As an effective reliability analysis technology, FMEA has been widely used in various industries to identify and evaluate potential failure modes in the design and manufacturing process. Through the risk assessment of failure modes, key characteristics that require additional measures to control are determined. The premise of risk assessment is to have a proper setting and understanding of the risk level of the failure mode. Therefore, this study proposes an AP design method based on the risk level of failure mode to improve the design theory of AP. This is not only necessary for the use of FMEA technology to carry out risk assessment, but also can help users understand the design process of AP more clearly to obtain better application results.

[0003] At present, when enterprises use FMEA tools to assess failure mode risks, most of them still adopt the traditional risk priority number (RPN) method, that is, the RPN value is determined by the product of the integer values ​​of S, O and D from 1 to 10.

[0004] The AIAG-VDA Failure Mode and Effects Analysis (FMEA) Handbook (Southfield: Automotive Industry Action Group) has proposed to replace RPN with AP, using the logical combination of S, O and D to correspond to the three categories of high, medium and low AP. However, since the handbook does not provide the theory or rules of AP design, it is unable to reasonably explain or answer the scientific correspondence between the 1,000 possible combinations of S, O, D and the high, medium and low AP levels, which confuses users and limits the application of AP. In order to promote AP, an efficient and concise risk assessment method, to completely replace RPN and be widely accepted and used, it is necessary to study and propose a scientific AP design method for failure mode risk levels. Summary of the invention

[0005] In view of the above-mentioned problems existing in risk assessment using FMEA, the present invention provides a measure priority design method for failure mode risk level, which can not only serve as a theoretical basis for AP design and improvement, but also help users to more clearly understand the AP design process to obtain better application effects.

[0006] The purpose of the present invention is achieved by adopting the following technical solutions: A method for designing the priority of measures based on the risk level of a failure mode, the steps of which are as follows:

[0007] 1) The risk of failure mode is comprehensively described by three risk factors: severity factor S, frequency factor O and detection factor D, and the risk level is defined:

[0008] 1.1S, O and D were quantitatively evaluated using a 1-10 point system, with 1 and 10 representing the lowest and highest risks, respectively;

[0009] 1.2 The risk levels of S, O and D are divided into five risk levels: very high, high, medium, low and very low on a 10-point scale based on their quantitative evaluation results;

[0010] 1.3 Based on the logical combination of S, O and D risk levels, the risk level of failure mode FM is divided into three risk levels: low R1, medium R2 and high R3, which correspond to low AP, medium AP and high AP respectively;

[0011] 2) Construct a risk matrix of failure modes and define risk index:

[0012] 2.1 Constructing a three-dimensional risk matrix A including S, O and D SOD ;

[0013] 2.2 Taking the most concerned risk factor S among the three risk factors as the benchmark, dimensionality reduction processing is performed to form a two-dimensional risk matrix A consisting of risk factor O and risk level D. OD (S);

[0014] 2.3A OD (S) is a 5×5 risk matrix, containing 25 matrix elements;

[0015] 2.4 Define the risk index x, which is an integer in the range [1,25] and is used to measure the matrix A OD (S) is the relative risk level of a group of risk events contained in each element. The larger the risk index value, the greater the risk;

[0016] 2.5 The risk index allocation rule is as follows:

[0017]

[0018] Where: Q S , Q O and Q D are the weight coefficients of risk factors respectively. The quantitative evaluation scores of risk factors are regarded as the coordinate values ​​of points. Then each risk event is represented by a set of coordinates of points. The risk limit point is M(S max ,O max ,D max ), that is, M(10,10,10); d i is the matrix A ODThe corresponding points of each risk event in (S) i ,O i ,D i ) to the risk limit point M. The smaller the distance, the greater the risk level, and the greater the value of the risk index x assigned; It is the average value of the sum of the distances between the corresponding points of the n risk events contained in each matrix element, and is used to measure the average risk level of a group of risk events represented by the matrix element; The smaller the value, the higher the average risk level. The values ​​are arranged from small to large, corresponding to the risk index values ​​from 25 to 1;

[0019] 3) Establish a risk preference difference model based on risk utility function;

[0020] 3.1 As the risk index x value increases, the assessor's attitude will change from risk-taking to conservative;

[0021] 3.2 Using the risk utility function to describe the evaluator's evaluation factor S for a specific risk level S n Risk attitude when:

[0022]

[0023] Where: U(x) is the risk utility value, U(x)∈(0,1); a n 、b n Severity S is S n The risk distribution parameters under each level are equally spaced;

[0024] 3.3 In order to make the graph of the risk utility function meet the definition domain requirements of the risk index x, its general expression is transformed as follows:

[0025]

[0026] 4) Determine the risk level of the failure mode and design the corresponding AP accordingly:

[0027] 4.1 According to the failure mode risk level, the risk level is allocated in the proportion of low R1, medium R2, and high R3, that is, the value of U(x) is divided into three value ranges: [0, 0.5], (0.5, 0.7], and (0.7, 1);

[0028] 4.20≤U(x)≤0.5, the failure mode is low risk, its AP is low, and the corresponding element in the risk matrix is ​​represented by L;

[0029] 4.30.5<U(x)≤0.7, the failure mode is medium risk, its AP is medium, and the corresponding element in the risk matrix is ​​represented by M;

[0030] 4.40.7<U(x)≤1, the failure mode is high risk, its AP is high, and the corresponding element in the risk matrix is ​​represented by H;

[0031] 5) Application of design methods:

[0032] 5.1 Apply the above-mentioned design methods and processes to design the AP of failure mode;

[0033] 5.2 Reveal the design rules implied in the design results.

[0034] The present invention can better utilize FMEA technology to carry out risk assessment, which can not only serve as a theoretical basis for AP design and improvement, but also help users understand the AP design process more clearly, thereby obtaining better application effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 is a flow chart of the AP design method of failure mode risk level in the present invention;

[0036] Figure 2 is a schematic diagram of a three-dimensional risk matrix in an embodiment of the present invention;

[0037] Figure 3 is an O×D plane risk matrix diagram in an embodiment of the present invention;

[0038] Figure 4a is the distribution of the type A risk index value in the embodiment of the present invention;

[0039] Figure 4b is the distribution of the type B risk index value in the embodiment of the present invention;

[0040] Figure 4c is the distribution of the C-type risk index value in the embodiment of the present invention;

[0041] Figure 5 is a schematic diagram of a risk utility function model in an embodiment of the present invention;

[0042] Figure 6a Schematic diagram of the distribution of AP tables in the FMEA manual in an embodiment of the present invention;

[0043] Figure 6b is a diagram showing the degree of fit of the Type A-Risk Preference Difference Model in an embodiment of the present invention;

[0044] Figure 6c is a diagram showing the degree of fit of the Type B-Risk Preference Difference Model in an embodiment of the present invention;

[0045] Figure 6d is a diagram showing the degree of fit of the Type C-Risk Preference Difference Model in an embodiment of the present invention;

[0046] Figure 7a It is the AP grade distribution diagram of the FMEA manual in the embodiment of the present invention;

[0047] Figure 7b is a diagram of the Type A-risk preference difference model in an embodiment of the present invention;

[0048] Figure 7c is a diagram of a Type B-risk preference difference model in an embodiment of the present invention;

[0049] Figure 7d 4 is a diagram of a Type C-risk preference difference model in an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The present invention is further described below in conjunction with the accompanying drawings and the AP examples in the FMEA manual. A method for designing the priority of measures for failure mode risk levels, the specific implementation steps of which are as follows (eg Figure 1 shown):

[0051] Step 1: The risk of failure mode is comprehensively described by three risk factors: severity factor (S), frequency factor (O) and detection factor (D):

[0052] The FMEA manual uses a 1-10 point system to quantitatively evaluate S, O, and D, with 1 and 10 representing the lowest and highest risks, respectively. Here, the risk levels of S, O, and D are divided into five risk levels: very high (level V), high (level IV), medium (level III), low (level II), and very low (level I) according to their quantitative evaluation results. Table 1 gives the specific information on the classification of the three risk factors.

[0053] Table 1 Risk factor classification

[0054]

[0055]

[0056] The risk level of failure mode is divided into three risk levels: high, medium and low, and AP is set. Table 2 gives the specific information of the risk level classification of failure mode.

[0057] Table 2 Failure mode classification

[0058] Failure Mode Risk Level AP <![CDATA[Low (R1)]]> Low <![CDATA[(R2) in China]]> middle <![CDATA[High (R3)]]> high

[0059] Step 2: Construct a risk matrix of failure modes and define risk index:

[0060] Construct a three-dimensional risk matrix A including S, O and D SOD , using the form of graded function to express comprehensive consideration of A SODrisks, namely:

[0061] R=f(S,O,D)=[R ijk ];

[0062] Where: R ijk is the risk level corresponding to S level i, O level j and D level k in the risk matrix, and the values ​​of i, j and k are I, II, III, IV and V (the three-dimensional risk matrix is ​​shown in Figure 2 shown).

[0063] In order to integrate the three risk factors into a single risk utility value, the three-dimensional risk matrix needs to be reduced in dimension based on factor S to form an O×D risk matrix A consisting of the risk levels of factors O and D. OD (S), at any level of S, the levels of O and D form a planar risk matrix diagram with 5×5 matrix elements (such as Figure 3 As shown in the figure), each element in the matrix represents a set of risk events.

[0064] The risk index x is defined here. Its value is an integer with a range of [1,25]. The value of the risk index represents the priority of a group of risk events contained in the matrix element. The larger the risk index, the greater the risk.

[0065] In the above matrix diagram, the quantitative evaluation score of the risk factor is regarded as the coordinate value of the point, and the risk event contained in each matrix element can be expressed in the form of the coordinates of the point, where the risk limit point is M(S max ,O max ,D max ), that is, M(10,10,10), is the point with the greatest risk. The closer to this point, the greater the risk. The distribution of risk index values ​​in the risk matrix is ​​determined by calculating the average value of the sum of the distances between each point contained in the matrix element and the plane risk limit point. The distribution rule formula is as follows:

[0066]

[0067] Where: Q S , Q O and Q D are the weight coefficients of risk factors respectively. i is the matrix A OD The corresponding points of each risk event in (S) i ,O i ,D i ) to the risk limit point M. The smaller the distance, the greater the risk level and the larger the value of the assigned risk index x. It is the average value of the sum of the distances between the corresponding points of the n risk events contained in each matrix element, and is used to measure the average risk level of a group of risk events represented by the matrix element. The smaller the value, the higher the average risk level. The values ​​are arranged from small to large, corresponding to the values ​​of risk index x from 25 to 1;

[0068] Taking the weight coefficient as an example with one decimal place retained, Table 3 shows the combination of risk factor weight coefficients.

[0069] Table 3 Risk factor weight coefficient (retain one decimal place as an example)

[0070]

[0071] According to the above formula and the given weight coefficient, different distributions of risk index values ​​in the plane risk matrix can be obtained. The four weight combinations correspond to three types of risk index value distributions, namely Type A (0.5-0.3-0.2), Type B (0.5-0.4-0.1, 0.6-0.3-0.1) and Type C (0.7-0.2-0.1) (e.g. Figures 4a to 4c shown).

[0072] Step 3: Establish a risk preference difference model based on risk utility function:

[0073] As the risk index x value increases, the evaluator's attitude will change from risk-taking to conservative, that is, from risk-loving to risk-averse. Therefore, a risk utility function model is constructed here to describe this possible attitude change (such as Figure 5 shown).

[0074] The risk utility function model constructed here is a concave function below a certain risk level, indicating that the assessor's attitude towards risk is risk-loving and adventurous; above a certain risk level, it is a convex function, indicating that the assessor's attitude towards risk is risk-averse and relatively conservative. The risk utility function model can better describe the change of the decision maker's risk attitude with the increase of the risk index x value within any severity level.

[0075] Step 4: Determine the risk level of the failure mode and design the AP accordingly:

[0076] The risk level is allocated according to the failure mode risk level of low (R1), medium (R2), and high (R3) with a rating ratio of 5:2:3, that is, the value of U(x) is divided into three value ranges: [0,0.5], (0.5,0.7], and (0.7,1);

[0077] 0≤U(x)≤0.5, the failure mode is low risk, its AP is low, and the corresponding element in the risk matrix is ​​represented by L;

[0078] 0.5<U(x)≤0.7, the failure mode is medium risk, its AP is medium, and the corresponding element in the risk matrix is ​​represented by M;

[0079] 0.7<U(x)≤1, the failure mode is high risk, its AP is high, and the corresponding element in the risk matrix is ​​represented by H;

[0080] Without loss of generality, the risk distribution parameters are equally spaced, and the specific distribution of failure mode risk levels in the risk matrix under different risk attitudes can be obtained; for the convenience of comparison with the FMEA manual, the AP table in the manual (such as Figure 6a The risk matrix constructed by the present invention is shown in the form of the risk matrix. The value of the risk distribution parameter in each model can be the risk distribution parameter determined when the AP table is most consistent. Ⅲ When the level is , the consistency between the present invention and the AP table is shown here (such as Figure 6b to Figure 6d shown).

[0081] Table 4 shows the risk distribution parameters of the risk preference difference model constructed based on the risk utility function at different severity levels when the degree of fit with the AP table is the highest.

[0082] Table 4 Risk distribution parameters determined at different severity levels

[0083]

[0084]

[0085] Step 5: Application of design method:

[0086] In the above steps, the present invention provides the risk distribution parameters that correspond to the closest consistency with the AP table. In order to more intuitively represent the distribution of failure mode risk levels of AP and analyze the design rules implied by AP, the corresponding three-dimensional risk level distribution diagrams (i.e., AP distribution diagrams) are constructed. Each cube in the diagram represents a combination of S, O, and D. The L-marked area represents an acceptable risk, and the corresponding AP is low; the M-marked area represents a risk that needs to be reasonably controlled, and the corresponding AP is medium; the H-marked area represents an unacceptable risk, and the corresponding AP is high. The AP distribution diagram obtained according to the present invention is compared with the AP level distribution diagram given in the fifth edition of the FMEA manual (the AP level distribution diagram given in the manual is as follows: Figure 7a As shown, the AP distribution obtained by the present invention is as follows Figure 7b to Figure 7d shown).

[0087] FIG7 shows the failure mode risk level distribution when the design is performed according to the AP design method proposed in the present invention and the AP in the FMEA manual is most consistent. Each three-dimensional risk level distribution diagram contains 125 cubes, and Table 5 shows the consistency comparison results between them.

[0088] Table 5. Comparison results of consistency

[0089]

[0090] By comparing the three-dimensional risk level distribution diagram, the consistency comparison results with the AP table under different risk index value distribution conditions are obtained. It can be seen from the table that under the distribution of type A risk index values, the consistency of the risk preference difference model based on the risk utility function constructed by the present invention with the FMEA manual standard reaches 100%; under the distribution of type B risk index values, the consistency of the two is 96%; under the distribution of type C risk index values, the consistency of the two is 98.4%. It can be seen that the risk level distribution diagram generated by the type A-risk preference difference model can better reflect the distribution law of AP in the FMEA manual, and its consistency is exactly the same. The risk factor weights behind it are: severity S (0.5), frequency O (0.3), and detection D (0.2), which shows that the AP method of the fifth edition of the FMEA manual pays considerable attention to severity S, while considering detection D at a significantly lower level. At the same time, the risk preference difference model constructed based on the risk utility function can well describe the change of the decision maker's risk preference from risk preference to risk aversion as the risk index value increases. It can better reflect the decision-making attitude of the evaluator in the AP design process and can well characterize the distribution of AP.

[0091] The AP design method oriented to failure mode risk level can be used for the design and improvement theory of AP corresponding to failure mode risk level. It can also help users understand the AP design process more clearly and provide support for risk assessment using AP in FMEA technology.

Claims

1. A method for designing the priority of measures based on the failure mode risk level, the steps of which are as follows: 1) The risk of failure mode is comprehensively described by three risk factors: severity factor S, frequency factor O and detection factor D, and the risk level is defined: 1.1S, O and D were quantitatively evaluated using a 1-10 point system, with 1 and 10 representing the lowest and highest risks, respectively; 1.2 The risk levels of S, O and D are divided into five risk levels: very high, high, medium, low and very low on a 10-point scale based on their quantitative evaluation results; 1.3 Based on the logical combination of S, O and D risk levels, the risk level of failure mode FM is divided into three risk levels: low R1, medium R2 and high R3, which correspond to low AP, medium AP and high AP respectively; 2) Construct a risk matrix of failure modes and define risk index: 2.1 Constructing a three-dimensional risk matrix A including S, O and D SOD ; 2.2 Taking the most concerned risk factor S among the three risk factors as the benchmark, dimensionality reduction processing is performed to form a two-dimensional risk matrix A consisting of risk factor O and risk level D. OD (S); 2.3A OD (S) is a 5×5 risk matrix, containing 25 matrix elements; 2.4 Define the risk index x, which is an integer in the range [1,25] and is used to measure the matrix A OD (S) is the relative risk level of a group of risk events contained in each element. The larger the risk index value, the greater the risk; 2.5 The risk index allocation rule is as follows: Where: Q S , Q O and Q D are the weight coefficients of risk factors respectively. The quantitative evaluation scores of risk factors are regarded as the coordinate values ​​of points. Then each risk event is represented by a set of coordinates of points. The risk limit point is M(S max ,O max ,D max ), that is, M(10,10,10); d i is the matrix A OD The corresponding points of each risk event in (S) i ,O i ,D i ) to the risk limit point M. The smaller the distance, the greater the risk level, and the greater the value of the risk index x assigned; E is the average value of the sum of the distances between the corresponding points of the n risk events contained in each matrix element and the point M, which is used to measure the average risk level of a group of risk events represented by the matrix element; The smaller the E value, the higher the average risk level. Arrange the E values ​​from small to large, corresponding to the risk index values ​​from 25 to 1; 3) Establish a risk preference difference model based on risk utility function; 3.1 As the risk index x value increases, the assessor's attitude will change from risk-taking to conservative; 3.2 Using the risk utility function to describe the evaluator's evaluation factor S for a specific risk level S n Risk attitude when: Where: U(x) is the risk utility value, U(x)∈(0,1); a n , b n Severity S is S n The risk distribution parameters under each level are equally spaced; 3.3 In order to make the graph of the risk utility function meet the definition domain requirements of the risk index x, its general expression is transformed as follows: 4) Determine the risk level of the failure mode and design the corresponding AP accordingly: 4.1 According to the failure mode risk level, the risk level is allocated in the proportion of low R1, medium R2, and high R3, that is, the value of U(x) is divided into three value ranges: [0, 0.5], (0.5, 0.7], and (0.7, 1); 4.20≤U(x)≤0.5, the failure mode is low risk, its AP is low, and the corresponding element in the risk matrix is ​​represented by L; 4.30.5<U(x)≤0.7, the failure mode is medium risk, its AP is medium, and the corresponding element in the risk matrix is ​​represented by M; 4.40.7<U(x)≤1, the failure mode is high risk, its AP is high, and the corresponding element in the risk matrix is ​​represented by H; 5) Application of design methods: 5.1 Apply the above-mentioned design methods and processes to design the AP of failure mode; 5.2 Reveal the design rules implied in the design results.

Images