A product reliability growth analysis method and system

Abstract

The present invention discloses a method and system for analyzing product reliability growth, and relates to the field of product reliability growth analysis. The method comprises: obtaining a product fault data set; according to the fault data set, counting the total working hours of the product in previous years and the number of occurrences of each fault mode; according to the improvement measure information, determining the evaluation algorithm of each fault mode; according to the evaluation algorithm, determining the effectiveness coefficient of each fault improvement; according to the total working hours in previous years, the number of occurrences of each fault mode and the effectiveness coefficient of each fault improvement, calculating the reliability index value after the fault improvement; judging whether the reliability index value after the fault improvement is greater than the required value of the reliability index; if so, determining the reliability growth of the product after the fault improvement. The present invention can quantify the impact of the fault improvement measures of electromechanical products on product reliability, and achieve the purpose of increasing product reliability.

Description

Technical Field

[0001] The present invention relates to the field of product reliability growth analysis, and in particular to a product reliability growth analysis method and system. Background Art

[0002] Reliability is a product performance indicator, indicating the ability of a product to maintain its functions and performance, and is an important indicator for measuring the quality of a product. The higher the reliability, the lower the possibility of product failure. However, product reliability is not achieved overnight, but is gradually improved through a series of technical means such as reliability design and testing. Therefore, when a product undergoes reliability growth activities, it is necessary to analyze the effectiveness of the growth activities and determine whether the reliability level has reached the required value. The analysis results can provide a decision-making basis for subsequent reliability growth activities or reliability growth activities for other similar products.

[0003] Reliability growth analysis technology has been developed since 1956 and can be roughly divided into the following categories. The first is the reliability growth tracking model, including the Duane model, the AMSAA model (US Army Materiel Systems Analysis Activity), etc.; the second is the reliability growth prediction model, including the ACPM model (ACPM, The AMSAA / Crow Projection Model) and the extended AMSAA model; the third is the small sample analysis technology, including the Bayesian analysis method. However, due to time and cost constraints, there will be a situation where there is no failure data after product growth when analyzing the effectiveness of reliability growth. In this case, the above analysis methods all have certain limitations.

[0004] Reliability growth tracking models, such as the Duane model and the AMSAA model, are suitable for reliability growth activities with timely correction strategies and reflect the correction effect in the growth curve after the test. The Bayesian evaluation method uses early data to construct a prior distribution, and then uses test data to update the probability distribution to obtain the posterior distribution. These growth analysis techniques require failure data after product improvements, and are not very applicable when failure data has not yet been generated during analysis.

[0005] The reliability growth prediction model uses the improved validity coefficient or growth factor to correct the data before improvement, so as to obtain the improved reliability data. However, both the improved validity coefficient and the growth factor are given subjectively according to the empirical value. Summary of the invention

[0006] The purpose of the present invention is to provide a product reliability growth analysis method and system, which can quantify the impact of fault improvement measures of electromechanical products on product reliability, judge whether the reliability of the improved product reaches the required value, and determine the failure mode and improvement measures that need to be improved based on the judgment result, so as to achieve the purpose of product reliability growth.

[0007] To achieve the above object, the present invention provides the following solutions:

[0008] A product reliability growth analysis method, the method comprising:

[0009] Obtain a product fault data set; the fault data set includes fault information, return to factory information, improvement measures information and reliability index requirement value; the fault information includes product number, fault occurrence time, fault mode and working hours;

[0010] According to the fault data set, the total working hours of the product over the years and the number of occurrences of each fault mode are counted;

[0011] Determining an evaluation algorithm for each of the failure modes according to the improvement measure information;

[0012] Determining each fault improvement effectiveness coefficient according to the evaluation algorithm;

[0013] Calculate the reliability index value after the fault improvement according to the total working hours over the years, the number of occurrences of each of the fault modes and the effectiveness coefficient of each of the fault improvements;

[0014] Determine whether the reliability index value after the fault is improved is greater than the reliability index requirement value;

[0015] If so, determine the reliability increase of the product after the fault improvement.

[0016] Optionally, the improvement measure information includes a list of failure modes to be improved, a reliability growth report and a failure mode improvement status table.

[0017] Optionally, counting the total working hours of the product over the years and the number of occurrences of each of the failure modes according to the failure data set specifically includes:

[0018] Determining the number of records of each fault mode according to the fault information;

[0019] determining, based on the reliability growth report, an unrecorded number of each failure mode in the failure data set;

[0020] Obtaining the total number of each failure mode according to the recorded number and the unrecorded number;

[0021] Determine an unimproved failure mode and an improved failure mode according to the list of failure modes to be improved and the failure information;

[0022] According to the total number of each of the failure modes, the unimproved failure mode and the improved failure mode, the number of each unimproved failure mode and the number of each improved failure mode are obtained;

[0023] Determining the improvement nodes of each improved fault mode according to the fault mode improvement status table;

[0024] According to the number of the improved nodes and each improved fault mode, determine the number of occurrences of each improved fault mode before the improved node and the number of occurrences after the improved node, and obtain the number of occurrences of each improved fault mode before improvement and the number of occurrences of each improved fault mode after improvement;

[0025] Determine the number of occurrences of each of the failure modes according to the total number of each of the failure modes, the number of each of the unimproved failure modes, the number of occurrences of each of the improved failure modes before improvement, and the number of occurrences of each of the improved failure modes after improvement;

[0026] The total working hours of the product over the years are calculated based on the working hours mentioned.

[0027] Optionally, determining an evaluation algorithm for each of the failure modes according to the improvement measure information specifically includes:

[0028] Determining the category of each of the failure modes according to the reliability growth report;

[0029] According to the categories, an evaluation algorithm for each of the failure modes is determined; the evaluation algorithm includes a fuzzy comprehensive evaluation method, a failure mechanism model method and a cause importance analysis method.

[0030] Optionally, the method further includes:

[0031] According to the fault improvement effectiveness coefficient, the fault mode to be improved is determined.

[0032] A product reliability growth analysis system, the system comprising:

[0033] An acquisition module is used to acquire a product fault data set; the fault data set includes fault information, return information, improvement measures information and reliability index requirement value; the fault information includes product number, fault occurrence time, fault mode and working hours;

[0034] A statistical module, used to count the total working hours of the product over the years and the number of occurrences of each of the failure modes according to the failure data set;

[0035] An evaluation algorithm determination module, used to determine an evaluation algorithm for each of the failure modes according to the improvement measure information;

[0036] A coefficient determination module, used to determine the effectiveness coefficient of each fault improvement according to the evaluation algorithm;

[0037] A calculation module, used to calculate the improved reliability index value according to the total working hours over the years, the number of occurrences of each of the failure modes and the effectiveness coefficient of each of the failure improvements;

[0038] The judgment module is used to determine the reliability increase of the product after the fault is improved according to the reliability index value after the improvement being greater than the reliability index requirement value.

[0039] Optionally, the improvement measure information includes a list of failure modes to be improved, a reliability growth report and a failure mode improvement status table.

[0040] Optionally, the statistics module includes:

[0041] A record quantity determination submodule, used to determine the record quantity of each fault mode according to the fault information;

[0042] an unrecorded quantity determination submodule, configured to determine the unrecorded quantity of each fault mode in the fault data set according to the reliability growth report;

[0043] A total quantity determination submodule, used to obtain the total quantity of each fault mode according to the recorded quantity and the unrecorded quantity;

[0044] An improved fault mode determination submodule is used for determining an unimproved fault mode and an improved fault mode based on the list of fault modes to be improved and the fault information;

[0045] An improved failure mode quantity determination submodule, used to obtain the quantity of each unimproved failure mode and the quantity of each improved failure mode according to the total quantity of each failure mode, the unimproved failure mode and the improved failure mode;

[0046] An improved node determination submodule, used to determine the improved nodes of each improved fault mode according to the fault mode improvement status table;

[0047] A submodule for determining the number of occurrences of a fault mode before and after improvement, for determining the number of occurrences of each improved fault mode before and after the improved node according to the number of the improved nodes and each improved fault mode, and obtaining the number of occurrences of each improved fault mode before improvement and the number of occurrences of each improved fault mode after improvement;

[0048] The occurrence number determination submodule of each of the fault modes is used to determine the occurrence number of each of the fault modes according to the total number of each of the fault modes, the number of each of the unimproved fault modes, the occurrence number of each of the improved fault modes before improvement, and the occurrence number of each of the improved fault modes after improvement;

[0049] The hour determination submodule is used to count the total working hours of the product over the years based on the working hours.

[0050] Optionally, the evaluation algorithm determination module includes:

[0051] A category determination submodule, used to determine the category of each of the failure modes according to the reliability growth report;

[0052] The method determination submodule is used to determine the evaluation algorithm of each of the fault modes according to the category; the evaluation algorithm includes a fuzzy comprehensive evaluation method, a fault mechanism model method and a cause importance analysis method.

[0053] Optionally, the system further comprises:

[0054] The module for determining the fault to be improved is used to determine the fault mode to be improved according to the fault improvement effectiveness coefficient.

[0055] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0056] The present invention provides a method for analyzing product reliability growth, including: obtaining a product failure data set; the failure data set includes failure information, return information, improvement measures information and reliability index requirement value; the failure information includes product number, failure occurrence time, failure mode and working hours; according to the failure data set, statistically analyzing the total working hours of the product over the years and the number of occurrences of each failure mode; according to the improvement measures information, determining the evaluation algorithm of each failure mode; according to the evaluation algorithm, determining the effectiveness coefficient of each failure improvement; according to the total working hours over the years, the number of occurrences of each failure mode and the effectiveness coefficient of each failure improvement, calculating the reliability index value after the failure improvement; judging whether the reliability index value after the failure improvement is greater than the reliability index requirement value; if so, determining the reliability growth of the product after the failure improvement. The present invention analyzes the characteristics of product failure data, determines data collection and statistical methods, and classifies failure modes, and constructs a method for determining failure improvement effectiveness coefficients of different categories of failure modes according to the classification results, quantifies the impact of product failure improvement measures on product reliability according to failure data and failure improvement effectiveness coefficients, judges whether the reliability of the improved product reaches the required value, and determines the failure mode and improvement measures that need to be improved according to the judgment result, so as to achieve the purpose of product reliability growth. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0058] Figure 1 A flow chart of the product reliability growth analysis method provided by the present invention;

[0059] Figure 2 A working principle diagram of the product reliability growth analysis method provided by the present invention;

[0060] Figure 3 A statistical flow chart of the occurrence frequency of the fault mode provided by the present invention;

[0061] Figure 4 A failure mode classification flow chart provided by the present invention;

[0062] Figure 5 A schematic diagram of an evaluation algorithm for a fault mode provided by the present invention;

[0063] Figure 6 A flow chart of a method for determining a fault improvement effectiveness coefficient based on a fuzzy comprehensive evaluation method provided by the present invention;

[0064] Figure 7 An evaluation system diagram provided by the present invention;

[0065] Figure 8 A flow chart of a method for determining a fault improvement effectiveness coefficient based on a fault mechanism model provided by the present invention;

[0066] Fig. 9 A flow chart of a method for determining a fault improvement effectiveness coefficient based on cause importance analysis provided by the present invention;

[0067] Fig.10 A schematic diagram of the failure mechanism evolution provided by the present invention;

[0068] Fig.11 The fault evolution diagram of the fault mode 6 provided by the present invention;

[0069] Fig.12 This is a schematic diagram of the structure of the product reliability growth analysis system provided by the present invention.

[0070] Explanation of symbols:

[0071] Acquisition module—1, statistics module—2, evaluation algorithm determination module—3, coefficient determination module—4, calculation module—5, judgment module—6. DETAILED DESCRIPTION

[0072] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0073] The purpose of the present invention is to provide a product reliability growth analysis method and system, which can quantify the impact of fault improvement measures of electromechanical products on product reliability, judge whether the reliability of the improved product reaches the required value, and determine the failure mode and improvement measures that need to be improved based on the judgment result, so as to achieve the purpose of product reliability growth.

[0074] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0075] The present invention provides a product reliability growth analysis method, which aims to solve the problem of no fault data after growth during analysis. The fault data before improvement is corrected by the fault improvement effectiveness coefficient to analyze the reliability level after improvement and determine whether it reaches the required value. Therefore, before analysis, it is necessary to collect and count the fault data and determine the fault improvement effectiveness coefficient. In order to determine the fault improvement effectiveness coefficient more scientifically and reasonably, the method classifies the fault modes and determines the method for determining the fault improvement effectiveness coefficient of each category of fault mode according to the characteristics of different categories of fault modes.

[0076] like Figure 1 and Figure 2 As shown, the present invention provides a product reliability growth analysis method, comprising:

[0077] Step S1: Obtain the product fault data set; the fault data set includes fault information, return information, improvement measures information and reliability index requirement value; the fault information includes product number, fault occurrence time, fault mode and working hours; specifically, the improvement measures information includes a list of fault modes to be improved, a reliability growth report and a fault mode improvement status table.

[0078] In actual applications, when products enter the use stage and reach the level of mass production, the generated fault data is very large. The fault data that needs to be processed include the following categories: (1) Fault information: Fault information refers to the information generated by the product due to the fault, which is collected in the "Fault Situation Table". The data in the table includes product number, fault occurrence time (calendar time), fault mode, faulty parts and their numbers; (2) Return information: Return information refers to the information of products returned to the main manufacturer due to faults or other reasons. Among them, the information returned due to faults overlaps with the fault information in item (1). The data is collected in the "Return Situation Table". Specific information includes: product number, return time (calendar time), product working time (working hours), and departure time (calendar time); (3) Corrective action information: Corrective action information refers to the relevant information of the fault mode for which improvement measures are proposed, including "Fault Mode List to be Improved", "Reliability Growth Report" and "Fault Mode Improvement Situation Table". The "Reliability Growth Report" describes in detail the failure modes that need to be improved, analyzes the possible causes of failures, and proposes corresponding corrective measures; the "Failure Mode Improvement Status Table" clarifies the implementation nodes of the corrective measures for each failure mode, including the implementation nodes for new products and the implementation nodes for returned products. (4) Reliability index requirement value.

[0079] Furthermore, a reliability growth report refers to a technical report that analyzes and improves a certain failure mode in order to reduce the failure rate of that failure mode. The failure mode comes from the "failure mode list to be improved", and each failure mode on the list has a corresponding reliability growth report.

[0080] The contents of the reliability growth report include: failure mode description, failure cause (failure mechanism), improvement measures, and implementation arrangements of measures. Failure mode description refers to the manifestation and characteristics of the failure mode. Failure cause refers to finding out the cause of the failure mode. If the cause is simple, the failure mechanism can be described directly; if the cause of the failure is complex, the possible causes of the failure mode can be listed one by one through the fault tree analysis method and the final cause of the failure can be found one by one. Improvement measures refer to the measures that can be improved in design, process, material, geometry, etc. proposed for the cause of the failure to reduce the impact of the cause of the failure. The implementation arrangement of measures refers to the node arrangement for the implementation of each improvement measure in subsequent products.

[0081] Step S2: According to the fault data set, the total working hours of the product over the years and the number of occurrences of each fault mode are counted.

[0082] S2 specifically includes:

[0083] Step S21: Determine the number of records of each fault mode according to the fault information.

[0084] Step S22: According to the reliability growth report, determine the number of unrecorded fault modes in the fault data set. Specifically, due to the huge amount of product failure data, the many locations involved, and the inconsistent understanding of the fault judgment criteria, the product failure data may be lost. In order to analyze accurately, the data set needs to be completed as much as possible to improve the fault data set.

[0085] Step S23: Obtain the total number of each failure mode according to the recorded number and the unrecorded number.

[0086] Step S24: Determine the unimproved failure mode and the improved failure mode according to the list of failure modes to be improved and the failure information.

[0087] Step S25: according to the total number of each failure mode, the unimproved failure mode and the improved failure mode, the number of each unimproved failure mode and the number of each improved failure mode are obtained.

[0088] Step S26: Determine the improvement nodes for each improved fault mode according to the fault mode improvement status table.

[0089] Step S27: According to the number of improved nodes and improved fault modes, determine the number of occurrences of each improved fault mode before the improved node and the number of occurrences after the improved node, and obtain the number of occurrences of each improved fault mode before the improvement and the number of occurrences of each improved fault mode after the improvement. Specifically, counting the number of fault occurrences includes counting the number of occurrences of all faults and the number of occurrences of the proposed improved fault mode. And determine whether the occurrence of the proposed fault mode occurs before or after the improvement. Figure 3 As shown in the figure, X is the node of new product improvement in the "Failure Mode Improvement Table", indicating that products with numbers greater than or equal to X are improved products when they leave the factory, and X is represented by the product number; Y is the node of returned repair product improvement in the "Failure Mode Improvement Table", indicating that products with numbers less than X are in an improved state after leaving the factory at time Y, and Y is represented by calendar time.

[0090] S27 specifically includes:

[0091] Step S271: If the product number of the failure mode for which the improvement measure is proposed is greater than or equal to the product number of the new product, the failure mode for which the improvement measure is proposed is the improved failure mode; specifically, the number of failures that occur after the implementation of the improvement measure for the i-th failure mode for which the improvement measure is proposed is l i +1.

[0092] Step S272: If the product number of the failure mode for which the improvement measure is proposed is smaller than the product number of the new product, and the return time of the product for which the improvement measure is proposed is earlier than the improvement time node of the repaired product, then the failure mode for which the improvement measure is proposed is the failure mode before the improvement; specifically, the number of failures that occurred before the implementation of the improvement measure for the i-th failure mode for which the improvement measure is proposed is m i +1.

[0093] Step S273: If the product number of the failure mode for which the improvement measures are proposed is smaller than the product number of the new product, and the return time of the product for which the improvement measures are proposed is later than the repair product improvement time node, then determine whether the last factory delivery time of the product for which the improvement measures are proposed is later than the repair product improvement time node.

[0094] Step S274: If yes, the failure mode for which the improvement measure is proposed is the improved failure mode; specifically, the number of failures that occur after the implementation of the improvement measure in the i-th failure mode for which the improvement measure is proposed is l i +1.

[0095] Step S275: If not, the failure mode for which the improvement measure is proposed is the failure mode before the improvement; specifically, the number of failures that occurred before the implementation of the improvement measure in the i-th failure mode for which the improvement measure is proposed is m. i +1.

[0096] Step S28: Determine the number of occurrences of each failure mode according to the total number of each failure mode, the number of each unimproved failure mode, the number of occurrences of each improved failure mode before improvement, and the number of occurrences of each improved failure mode after improvement.

[0097] Step S29: Count the total working hours of the product over the years based on the working hours.

[0098] Step S3: Determine the evaluation algorithm for each failure mode according to the improvement measure information.

[0099] S3 specifically includes:

[0100] Step S31: Determine the category of each fault mode according to the reliability growth report. Specifically, the classification of fault modes is to more reasonably perform quantitative calculation of the fault improvement effectiveness coefficient. Therefore, the fault modes are classified based on "quantitative determination of the fault improvement effectiveness coefficient". Figure 4As shown in the figure, the fault modes are divided into Class A fault modes and Class B fault modes. Class A is a fault mode that does not require improvement; Class B is a fault mode for which improvement measures are proposed. Further, according to the characteristics of the fault mechanism and improvement measures, Class B fault modes are divided into Class B1, B2 and B3 fault modes. Among them, Class B1 fault mode is that there is no mathematical model to describe the fault mechanism, or the fault improvement measures cannot be reflected in the input of the mechanism model; Class B2 fault mode is that there is a mathematical model to describe the fault mechanism and the fault improvement measures can be fully reflected in the input of the mechanism model; Class B3 fault mode is other situations except Class B1 fault mode and Class B2 fault mode.

[0101] Step S32: Determine the evaluation algorithm for each fault mode according to the category; the evaluation algorithm includes fuzzy comprehensive evaluation method, fault mechanism model method and cause importance analysis method. Specifically, according to the result of fault mode classification, the corresponding fault improvement effectiveness coefficient is determined by category. The fuzzy comprehensive evaluation method, fault mechanism model and cause importance analysis are used to quantitatively determine the fault improvement effectiveness coefficient. Figure 5 As shown, the evaluation algorithm for type B1 failure mode adopts the fuzzy comprehensive evaluation method; the evaluation algorithm for type B2 failure mode adopts the failure mechanism model method; and the evaluation algorithm for type B3 failure mode adopts the cause importance analysis method.

[0102] Step S4: Determine the effectiveness coefficient of each fault improvement according to the evaluation algorithm.

[0103] Specifically, the specific methods for determining the effectiveness coefficient of each fault improvement are as follows:

[0104] like Figure 6 As shown in the figure, the specific steps of the fuzzy comprehensive evaluation method adopted by the evaluation algorithm of the B1 type fault mode include:

[0105] Step 101: First, analyze the factors that affect the effectiveness of improvement measures and establish an evaluation system; e.g. Figure 7 As shown in the figure, among them, the accuracy of fault cause, rationality of improvement measures and feasibility of implementation plan are the criterion layer; the complexity of fault mechanism and so on are the factors affecting the criterion layer, which are called factor layer.

[0106] Step 102: Next, weight analysis is performed, and the initial weight of the factor layer is obtained by using the analytic hierarchy process (AHP), and the final weight of the factor layer is corrected by the entropy weight method; since in the process of reliability growth, analyzing the cause of the failure, proposing improvement measures, and implementing the improvement measures are closely linked and indispensable, the weights of the criterion layer in this method are the same.

[0107] Step 103: Use the fuzzy comprehensive evaluation method to score the factor layer. In this process, the compatibility and the expert's prior weight (i.e., the inverse of the number of experts) are used to obtain the expert's weight, and then the factor layer score is obtained; the final value of the fault improvement effectiveness coefficient is obtained by aggregating the scores and weights layer by layer.

[0108] like Figure 8 As shown in the figure, the fault mechanism model method used in the evaluation algorithm of the B2 fault mode specifically includes: obtaining reliability-related indicators through the fault mechanism model, and obtaining the fault improvement effectiveness coefficient directly according to the relationship between the fault improvement effectiveness coefficient and the reliability and MTTB, and the relationship is:

[0109]

[0110] The cause importance analysis method used in the evaluation algorithm of Class B3 failure mode includes:

[0111] Step 301: First, the proposed improvement measures are divided into two groups according to whether they can be described by a mechanism model.

[0112] Step 302: Calculate the corresponding improvement effectiveness using the above fuzzy comprehensive evaluation method and the fault mechanism model, that is, calculate the improvement effectiveness coefficient of some measures.

[0113] Step 303: Analyze the evolution process of the fault mechanism and form a fault mechanism evolution diagram; Fig. 9 shown.

[0114] Step 304: Fig.10 is a schematic diagram of the fault mechanism evolution. From the schematic diagram of the fault mechanism evolution, the in-degree of each event and the adjacency matrix A of each event can be obtained. The similarity matrix S and the transfer probability matrix M are calculated in sequence. The out-degree refers to the number of lines pointed out by arrows in the evolution diagram; the in-degree refers to the number of lines pointed into by arrows in the evolution diagram; the transfer probability is the probability that event one causes event two. If the transfer probability has real data, the transfer probability matrix M is corrected according to the real data to obtain the corrected transfer probability matrix P. According to the in-degree and transfer probability, the in-degree potential of each event can be obtained. The in-degree potential of the cause event targeted by the improvement measures is normalized to obtain the cause importance. Finally, the fault improvement effectiveness coefficient is obtained according to the improvement effectiveness coefficient of some improvement measures and the cause importance of their corresponding events.

[0115] Step S5: Calculate the reliability index value after fault improvement based on the total working hours over the years, the number of occurrences of each fault mode and the effectiveness coefficient of each fault improvement. Specifically, the fault data set is corrected using the fault improvement effectiveness coefficient, and the corrected fault data set is used as the number of failures that should occur in the product under the same time and improved technical status, and the reliability index value after fault improvement is calculated according to the model.

[0116] Step S6: Determine whether the reliability index value after the fault is improved is greater than the reliability index requirement value.

[0117] Step S7: If so, determine the reliability growth of the product after the fault improvement. Specifically, when the reliability index value after the fault improvement is greater than the required value of the reliability index, the current reliability growth strategy is effective. Among them, the reliability growth strategy is a list of fault modes that need to be improved and the corresponding improvement measures for each fault mode that needs to be improved. When the reliability index value after the fault improvement is less than or equal to the required value of the reliability index, the current reliability growth strategy needs to be further optimized; specifically including: (1) continue to improve the improved fault mode, that is, according to the level of the fault improvement effectiveness coefficient, continue to improve the fault mode with a low coefficient, and propose new improvement measures; (2) re-screen the unimproved fault mode, select the fault mode with large improvement space for improvement, and implement corresponding improvement measures according to the fault mode.

[0118] In addition, it also includes: determining the fault mode to be improved according to the fault improvement effectiveness coefficient.

[0119] In practical applications, before building the model, several assumptions of the model are first explained.

[0120] (1) Assume that faults are independent of each other.

[0121] (2) It is assumed that the improvements will not cause new failure modes.

[0122] (3) Assume that at the evaluation node, the system has x types of faults, and the number of faults for each fault is n i , 1≤i≤x; among them, there are y types of faults for which improvement measures are proposed, and there are z types of faults for which improvement measures are implemented and improved fault data are generated; the number of improved faults for each fault for which improvement measures are proposed and improved fault data are generated is l i , 0≤i≤z.

[0123]

[0124] Where T is the total engine operating time during the statistical period; k is the total number of faults for which no improvement measures were proposed; m iis the number of failures that occurred before the implementation of the i-th failure mode for which improvement measures are proposed; l i is the number of failures that occurred after the i-th failure mode for which improvement measures were proposed; d i is the failure improvement effectiveness coefficient of the i-th failure mode for which improvement measures are proposed. And, n i =m i +l i ,0≤i≤z;

[0125] As a specific implementation of this embodiment, taking engine reliability growth analysis as an example, this method is further described in detail.

[0126] 1. Collect data.

[0127] (1) Fault information: According to the “Fault Status Table”, collect the product number, fault occurrence time (calendar time), fault mode and working hours.

[0128] (2) Return information: Based on the “Return Status Table”, collect the product number, return time (calendar time), and departure time (calendar time).

[0129] (3) Corrective action information: The fault list and fault mode improvements are shown in Table 1.

[0130] Table 1 Failure mode improvement table

[0131]

[0132] Note: 1. A new engine is an engine that has just been delivered from the factory; a troubleshooting engine is an engine that was previously delivered and returned to the factory for repair due to various reasons.

[0133] 2. No. 240 is the product number, Figure 2 The X in the figure indicates that when the number of the new product is greater than or equal to 240, the improvement measures for failure mode 3 have been implemented.

[0134] 3. 2018.1 is the execution node of the three improvement measures for the fault mode of the troubleshooting machine. Figure 2 The Y in the figure indicates that the troubleshooting machine was improved in January 2018.

[0135] (4) Reliability index requirement: MTBF ≥ 400 hours.

[0136] 2. Analyze, compile statistics and process data.

[0137] (1) Improve the fault data set.

[0138] There are 1366 fault data in the “Fault Condition Table”, and 29 fault data mentioned in the “Detailed Reliability Growth Plan” but not recorded in the “Fault Condition Table”. Therefore, there are 1395 fault data in total, that is, N=1395.

[0139] (2) Count the number of faults that occur and determine whether the fault data is a fault before or after improvement.

[0140] according to Figure 2 The number of failures before and after the improvement of the six failure modes in Table 1 is obtained, as shown in Table 2.

[0141] Table 2 Failure mode occurrence frequency table

[0142]

[0143]

[0144] (3) Count the total working hours of the products over the years. The result is T = 545012 hours.

[0145] 3. Classify failure modes.

[0146] The failure modes are classified and the classification results are shown in Table 3.

[0147] Table 3 Classification results

[0148]

[0149] 4. Quantitatively determine the fault improvement effectiveness coefficient.

[0150] (1) Method for determining the fault improvement effectiveness coefficient based on fuzzy comprehensive evaluation method.

[0151] 1) Determine the weights.

[0152] The weights of each factor layer obtained by AHP-entropy weight method are:

[0153] W=(0.6174 0.3826 0.3100 0.4189 0.2711 0.3479 0.1993 0.4527)

[0154] The weight of each criterion in the criterion layer is: W = (1 / 3 1 / 3 1 / 3).

[0155] 2) Fuzzy comprehensive evaluation.

[0156] Taking fault mode 4 as an example, four experts scored each underlying factor. The expert weights were calculated based on the scoring results, and the final scores of the underlying factors were obtained. The scores of the four experts and their calculation process are shown in the table.

[0157] Table 4 Experts' Scoring

[0158]

[0159]

[0160] The scores of the criterion layer obtained by aggregating the underlying factors are 0.7878, 0.8680, and 0.6806, respectively, and the fault improvement effectiveness coefficient is: 0.7788.

[0161] (2) Method for determining the fault improvement effectiveness coefficient based on the fault mechanism model.

[0162] Fault improvement effectiveness coefficients are calculated for fault modes 1 to 5 using their respective mechanism models. Taking fault mode 1 as an example, the reliability after improvement calculated based on the mechanism model is:

[0163]

[0164] By the attached Figure 7 It can be obtained that the fault improvement effectiveness coefficient is:

[0165]

[0166] (3) Method for determining the fault improvement effectiveness coefficient based on cause importance analysis.

[0167] Fault mode 6 is analyzed by cause importance to obtain the fault improvement effectiveness coefficient. Figure 8 The technical route gradually calculates its fault improvement effectiveness coefficient.

[0168] 1) Analysis of failure mechanism and improvement measures.

[0169] After analysis, the failure mechanism of failure mode 6 evolved from three cause events; there are six improvement measures. Among them, measures 1-4 are proposed for cause events A and B, and the improvement effectiveness coefficient of the measures can be calculated using the mechanism model (low cycle fatigue life model); measures 5-6 are proposed for cause event C, and the improvement effectiveness coefficient is calculated using the fuzzy comprehensive evaluation method.

[0170] The MTBF before and after the improvement is calculated using the low cycle fatigue life model to be 11703 and 48925, so the improvement effectiveness coefficient of this part of the measures is:

[0171]

[0172] The improvement effectiveness coefficient of measures 5-6 calculated by fuzzy comprehensive evaluation method is 0.8286.

[0173] 2) Analysis of fault evolution process.

[0174] The failure evolution process of failure mode 6 is as follows: Fig.11 As shown. According to the fault evolution diagram of fault mode 6, calculate Fig.11 The importance of the five events in the above formula is calculated, and then the importance ranking and weight of the cause events A, B and C are obtained. Fig.11 The in-and-out degrees of each event are shown in Table 5.

[0175] Table 5 The entry and exit table of each event of power turbine blade fracture failure

[0176]

[0177] The adjacency matrix A is:

[0178] The similarity matrix S is:

[0179] The transmission probability is:

[0180] According to the known data, the probability that event B causes event A is 0.38, and the probability that event C transmits to event D is 0.8. Therefore, the modified transmission probability matrix is:

[0181]

[0182] Since the events are in an “or” relationship, the in-and-out potential of each event can be calculated according to the formula as shown in Table 6.

[0183]

[0184] Table 6 The entry and exit potential of each event

[0185]

[0186] In this process, only events A, B, and C are cause events. Therefore, in order to give the importance of causes A, B, and C, the in-and-out potentials of events A, B, and C are normalized, and the importance of events A, B, and C is obtained as shown in Table 7.

[0187] Table 7 Cause event weight values

[0188]

[0189]

[0190] 3) Determination of fault improvement effectiveness coefficient.

[0191] Measures 1-4 are proposed for cause events A and B, and measures 5-6 are proposed for cause event C. Therefore, the fault improvement effectiveness coefficient is:

[0192] d=(0.1525+0.3686)×0.7608+0.4789×0.8286=0.7933.

[0193] 5. Calculate the system reliability level.

[0194] The obtained fault improvement effectiveness coefficients are shown in Table 8.

[0195] Table 8 Improvement effectiveness coefficients for each failure mode

[0196]

[0197] The MTBF of the improved engine is:

[0198] Compared with the required value of 400 hours, the analysis shows that the reliability level after improvement has reached the required value. At the same time, the failure improvement effectiveness coefficient shows that the next work should focus on failure mode 3, failure mode 1 and failure mode 5. Because the improvement effect of these two failure modes is the weakest.

[0199] like Fig.12 As shown, the present invention also provides a product reliability growth analysis system, which is applied to a product reliability growth analysis method provided by the present invention, and the system includes:

[0200] Acquisition module 1 is used to obtain the product fault data set; the fault data set includes fault information, return information, improvement measures information and reliability index requirement value; the fault information includes product number, fault occurrence time, fault mode and working hours; specifically, the improvement measures information includes a list of fault modes to be improved, a reliability growth report and a fault mode improvement status table.

[0201] The statistical module 2 is used to count the total working hours of the product over the years and the number of occurrences of each failure mode based on the failure data set.

[0202] The evaluation algorithm determination module 3 is used to determine the evaluation algorithm of each fault mode according to the improvement measure information.

[0203] The coefficient determination module 4 is used to determine the effectiveness coefficient of each fault improvement according to the evaluation algorithm.

[0204] The calculation module 5 is used to calculate the improved reliability index value according to the total working hours in previous years, the number of occurrences of each failure mode and the effectiveness coefficient of each failure improvement.

[0205] The judgment module 6 is used to determine the reliability increase of the product after the fault is improved according to the reliability index value after improvement being greater than the reliability index requirement value.

[0206] In addition, the system also includes: a fault to be improved determining module, which is used to determine the fault mode to be improved according to the fault improvement effectiveness coefficient.

[0207] Among them, the statistics module 2 includes:

[0208] A record quantity determination submodule is used to determine the record quantity of each fault mode according to the fault information;

[0209] The unrecorded quantity determination submodule is used to determine the unrecorded quantity of each failure mode in the failure data set according to the reliability growth report.

[0210] The total quantity determination submodule is used to obtain the total quantity of each fault mode according to the recorded quantity and the unrecorded quantity.

[0211] The improved fault mode determination submodule is used for the list of fault modes to be improved and the fault information, and determines the unimproved fault modes and the improved fault modes.

[0212] The improved fault mode quantity determination submodule is used to obtain the quantity of each unimproved fault mode and the quantity of each improved fault mode according to the total quantity of each fault mode, the unimproved fault mode and the improved fault mode.

[0213] The improvement node determination submodule is used to determine the improvement nodes of each improved fault mode according to the fault mode improvement status table.

[0214] The submodule for determining the number of occurrences of the fault mode before and after improvement is used to determine the number of occurrences of each improved fault mode before the improved node and the number of occurrences after the improved node according to the number of improved nodes and each improved fault mode, so as to obtain the number of occurrences of each improved fault mode before improvement and the number of occurrences of each improved fault mode after improvement.

[0215] The occurrence number determination submodule of each fault mode is used to determine the occurrence number of each fault mode according to the total number of each fault mode, the number of each unimproved fault mode, the occurrence number of each improved fault mode before improvement and the occurrence number of each improved fault mode after improvement.

[0216] The hour determination submodule is used to count the total working hours of the product over the years based on the working hours.

[0217] Among them, the evaluation algorithm determination module 3 includes:

[0218] The category determination submodule is used to determine the category of each failure mode according to the reliability growth report.

[0219] The method determination submodule is used to determine the evaluation algorithm of each fault mode according to the category; the evaluation algorithm includes fuzzy comprehensive evaluation method, fault mechanism model method and cause importance analysis method.

[0220] This method quantitatively determines the effectiveness of growth activities through fuzzy comprehensive evaluation method, fault mechanism model and cause importance analysis method, calculates the fault improvement effectiveness coefficient; then uses this coefficient to correct the data before improvement to predict the reliability level after improvement. This method can first solve the situation where there is no fault data during growth analysis; secondly, compared with the reliability growth prediction model, its determination of the fault improvement effectiveness coefficient is more objective; finally, the analysis results of growth effectiveness can provide a decision-making basis for subsequent growth activities.

[0221] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0222] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A product reliability growth analysis method, characterized in that: The method comprises: Obtain a product fault data set; the fault data set includes fault information, return information, improvement measures information and reliability index requirement value; the fault information includes product number, fault occurrence time, fault mode and working hours, wherein the improvement measures information includes a list of fault modes to be improved, a reliability growth report and a fault mode improvement status table; According to the fault data set, the total working hours of the product over the years and the number of occurrences of each fault mode are counted, specifically including: determining the number of records of each fault mode according to the fault information; determining the number of unrecorded fault modes in the fault data set according to the reliability growth report; obtaining the total number of each fault mode according to the number of records and the number of unrecorded fault modes; determining the unimproved fault mode and the improved fault mode according to the list of fault modes to be improved and the fault information; obtaining the number of each unimproved fault mode and the number of each improved fault mode according to the total number of each fault mode, the unimproved fault mode and the improved fault mode; obtaining the number of each unimproved fault mode and the number of each improved fault mode according to the A table of improved fault modes, determining the improved nodes of each improved fault mode; determining the number of occurrences of each improved fault mode before the improved node and the number of occurrences after the improved node according to the number of improved nodes and each improved fault mode, and obtaining the number of occurrences of each improved fault mode before the improvement and the number of occurrences of each improved fault mode after the improvement; determining the number of occurrences of each fault mode according to the total number of each fault mode, the number of each unimproved fault mode, the number of occurrences of each improved fault mode before the improvement and the number of occurrences of each improved fault mode after the improvement; and counting the total working hours of the product over the years according to the working hours; Determining an evaluation algorithm for each of the failure modes according to the improvement measure information; Determining each fault improvement effectiveness coefficient according to the evaluation algorithm; Calculate the reliability index value after the fault improvement according to the total working hours over the years, the number of occurrences of each of the fault modes and the effectiveness coefficient of each of the fault improvements; Determine whether the reliability index value after the fault is improved is greater than the reliability index requirement value; If so, determine the reliability increase of the product after the fault improvement; According to the fault improvement effectiveness coefficient, the fault mode to be improved is determined, specifically: assuming that the faults are independent of each other; assuming that the improvement will not cause new fault modes; assuming that at the evaluation node, the system has x types of faults, and the number of faults of each fault is n i , 1≤i≤x; among them, there are y types of faults for which improvement measures are proposed, and there are z types of faults for which improvement measures are implemented and improved fault data are generated; the number of improved faults for each fault for which improvement measures are proposed and improved fault data are generated is l i , the formula is as follows: Where T is the total engine operating time during the statistical period; k is the total number of faults for which no improvement measures were proposed; m i is the number of failures that occurred before the implementation of the i-th failure mode for which improvement measures are proposed; l i is the number of failures that occurred after the i-th failure mode for which improvement measures were proposed; d i is the failure improvement effectiveness coefficient of the i-th failure mode for which improvement measures are proposed, n i =m i +l i ,0≤i≤z.

2. The product reliability growth analysis method according to claim 1, characterized in that: Determining the evaluation algorithm of each of the failure modes according to the improvement measure information specifically includes: Determining the category of each of the failure modes according to the reliability growth report; According to the categories, an evaluation algorithm for each of the failure modes is determined; the evaluation algorithm includes a fuzzy comprehensive evaluation method, a failure mechanism model method and a cause importance analysis method.

3. A product reliability growth analysis system, characterized in that: The system comprises: An acquisition module is used to acquire a product fault data set; the fault data set includes fault information, return information, improvement measures information and reliability index requirement value; the fault information includes product number, fault occurrence time, fault mode and working hours, wherein the improvement measures information includes a list of fault modes to be improved, a reliability growth report and a fault mode improvement status table; The statistical module is used to count the total working hours of the product over the years and the number of occurrences of each failure mode according to the failure data set, specifically including: A record quantity determination submodule, used to determine the record quantity of each fault mode according to the fault information; an unrecorded quantity determination submodule, configured to determine the unrecorded quantity of each fault mode in the fault data set according to the reliability growth report; A total quantity determination submodule, used to obtain the total quantity of each fault mode according to the recorded quantity and the unrecorded quantity; An improved fault mode determination submodule is used for determining an unimproved fault mode and an improved fault mode based on the list of fault modes to be improved and the fault information; An improved failure mode quantity determination submodule, used to obtain the quantity of each unimproved failure mode and the quantity of each improved failure mode according to the total quantity of each failure mode, the unimproved failure mode and the improved failure mode; An improved node determination submodule, used to determine the improved nodes of each improved fault mode according to the fault mode improvement status table; A submodule for determining the number of occurrences of a fault mode before and after improvement, for determining the number of occurrences of each improved fault mode before and after the improved node according to the number of the improved nodes and each improved fault mode, and obtaining the number of occurrences of each improved fault mode before improvement and the number of occurrences of each improved fault mode after improvement; The occurrence number determination submodule of each of the fault modes is used to determine the occurrence number of each of the fault modes according to the total number of each of the fault modes, the number of each of the unimproved fault modes, the occurrence number of each of the improved fault modes before improvement, and the occurrence number of each of the improved fault modes after improvement; The hour determination submodule is used to count the total working hours of the product over the years according to the working hours; An evaluation algorithm determination module, used to determine an evaluation algorithm for each of the failure modes according to the improvement measure information; A coefficient determination module, used to determine the effectiveness coefficient of each fault improvement according to the evaluation algorithm; A calculation module, used to calculate the improved reliability index value according to the total working hours over the years, the number of occurrences of each of the failure modes and the effectiveness coefficient of each of the failure improvements; The judgment module is used to determine the reliability increase of the product after the fault is improved according to the reliability index value after the improvement being greater than the reliability index requirement value.

4. The product reliability growth analysis system according to claim 3, characterized in that: The evaluation algorithm determination module includes: A category determination submodule, used to determine the category of each of the failure modes according to the reliability growth report; The method determination submodule is used to determine the evaluation algorithm of each of the fault modes according to the category; the evaluation algorithm includes a fuzzy comprehensive evaluation method, a fault mechanism model method and a cause importance analysis method.

Images