A co-platform product-based reliability test design method

Abstract

The application discloses a reliability test design method based on a common platform product, wherein step 1 is common platform product module definition, and each component module of the product determined according to the module definition is listed, i.e., module 1, module 2,..., and module n, wherein n is the number of modules; step 2 is to determine test stress, including determining test stress type and test stress level; step 3 is to determine a stress sensitivity coefficient matrix of each module of the common platform product and the test stress; step 4 is to determine a change state matrix of each module of a newly developed product and the original common platform product; step 5 is to calculate a test stress priority coefficient matrix; and step 6 is to determine a test scheme according to the test stress priority coefficient matrix. The method can quantitatively describe the correlation between the newly developed product and each test stress, thereby stress can be selected in a targeted manner, the test project of the newly developed product is trimmed, the development cycle of the common platform product is shortened, the test cost is reduced, and the competitiveness of the product is improved.

Description

Technical Field

[0001] This invention provides a reliability test design method based on a common platform product, which relates to the field of product reliability test technology. Background Technology

[0002] Reliability testing is a crucial step in product design and development. It involves ensuring a product maintains functional reliability throughout its specified lifespan under all expected usage, transportation, and storage conditions. Reliability testing exposes a product to natural or artificial environmental conditions to evaluate its performance under real-world usage, transportation, and storage conditions, and to analyze the extent and mechanism of environmental influences. By using various environmental testing equipment to simulate high and low temperatures, high temperature and high humidity, and temperature variations in climatic environments, the product's performance in its usage environment is accelerated, verifying whether it meets the quality objectives anticipated during research, design, and manufacturing, thus providing an overall evaluation of the product. For different products, after evaluating their actual application environment, appropriate testing protocols are determined according to relevant standards, such as GB / T 2423.

[0003] The commonly used process for determining a reliability test plan is as follows: Figure 1 While the existing reliability testing protocols mentioned above are relatively comprehensive for testing new products, they are time-consuming and costly.

[0004] In product design, to reduce design and production costs, products are often designed as platforms, meaning the product design is modularized and platform-based. Depending on different usage requirements, the design of some modules can be adjusted, minimizing design changes while meeting those requirements. Newly developed products have relatively mature platforms and modules, reducing both R&D and production costs and improving product reliability. For reliability testing of these shared-platform products, information from reliability tests of similar products on the same platform is usually available. However, existing comprehensive testing methods do not effectively utilize this information, resulting in longer product development cycles, failing to meet the current market demands for rapid product iteration, and leading to high testing costs due to the large number of test items, thus reducing the product's market competitiveness. Summary of the Invention

[0005] The purpose of this invention is to provide a standardized method that utilizes test information from similar products on a shared platform to quantify the correlation of reliability test stresses for newly developed products, determine the priority of test stresses, and thus select stresses in a targeted manner. This enables the tailoring of test items in the test plan, shortens the development cycle of shared platform products, reduces test costs, and thereby enhances product competitiveness.

[0006] This invention provides a reliability test design method based on a common platform product. By defining a stress sensitivity coefficient matrix and a change state matrix for each module, a test stress priority coefficient matrix is ​​calculated. This method can quantitatively describe the correlation between the newly developed product and each test stress, thereby selectively choosing stresses and tailoring test items for the newly developed product.

[0007] In this invention, a shared platform product refers to a series of products based on the same hardware design platform, whose module composition is the same or similar, and whose environmental stresses in application are the same or similar. An original shared platform product refers to a mature product developed based on the aforementioned hardware design platform that has been verified; a newly developed product refers to a new product developed based on the aforementioned hardware design platform that requires experimental design.

[0008] Compared to traditional fault mechanism models, this invention mainly includes the following three innovations:

[0009] (1) The correlation between the various modules of the common platform product and the test stress is described by defining the stress sensitivity coefficient matrix. The form is simple and easy to use.

[0010] (2) By defining the test stress priority coefficient matrix, the correlation between the newly developed product and each test stress is quantitatively described, so as to design test schemes in a targeted manner, which facilitates the tailoring of test items, shortens the development cycle of the new product, and reduces test costs.

[0011] (3) The test stress priority coefficient matrix is ​​obtained by multiplying the stress sensitivity coefficient matrix and the change state matrix of each module. The method is simple, scientific and accurate.

[0012] This invention discloses a reliability test design method based on a shared platform product. First, the shared platform product is analyzed and modules are defined. Second, environmental analysis is used to determine the environmental stresses that have a significant impact on the shared platform product, and thus the test stresses are determined. Then, a stress sensitivity coefficient matrix and a change state matrix for each module of the newly developed product are defined and determined. Finally, the test stress priority coefficient matrix is ​​calculated to determine the test plan. Figure 2 As shown, the specific steps are as follows:

[0013] Step 1: Define the common platform product modules. Based on the actual design of the common platform product, agree on the levels to be analyzed, thus completing the module definition. Generally, select the module levels that may change during design modifications. List the product's constituent modules as determined by the module definitions: Module 1, Module 2, ..., Module n, where n is the number of modules.

[0014] Step 2: Determine the test stress.

[0015] Step 201: Determine the type of test stress. Analyze the usage environment profile of the common platform product and determine the environmental stresses that have a significant impact on the product, including temperature, vibration, electrical stress, humidity, etc.

[0016] Step 202: Determine the test stress level. Based on the overall design of the common platform product or the user's requirements, select the test standard and determine the test stress level according to the indicators proposed in the usage profile or overall design.

[0017] Step 3: Determine the stress sensitivity coefficient matrix of each module of the common platform product and the test stress.

[0018] Step 301: Develop a correlation table between modules and test stresses. Based on past testing experience or through expert evaluation, determine the correlation between each module of the common platform product and the test stress. If a module is affected by the corresponding test stress, it is considered correlated; otherwise, it is considered uncorrelated.

[0019] Step 302: Define the stress sensitivity coefficient matrix. Using the number of test stresses m as the number of rows and the number of modules n as the number of columns, define an m×n matrix A as the stress sensitivity coefficient matrix for each module of the product and the test stresses.

[0020]

[0021] Where m is the experimental stress number;

[0022] n: the number of modules;

[0023] The value a in the i-th row and j-th column of the matrix ij : Defined as the stress sensitivity coefficient between the j-th module and the i-th test stress of the common platform product. If the module and the test stress are related, the stress sensitivity coefficient is defined as 1; if the module and the test stress are not related, the stress sensitivity coefficient is defined as 0.

[0024] Step 303: Determine the stress sensitivity coefficient matrix of each module of the common platform product and the test stress. Based on the correlation table of modules and test stress and equation (1), the stress sensitivity coefficient matrix of each module of the product and the test stress can be determined.

[0025] Step 4: Determine the change status matrix of each module of the newly developed product and the original shared platform product.

[0026] Step 401: Based on the changes to each module of the newly developed product and the original common platform product, formulate a change status table for each module of the newly developed product and the original common platform product.

[0027] Step 402: Define the change state matrix for each module of the newly developed product and the original shared platform product. Using the number of modules n as the number of rows, define an n×1 matrix B as the change state matrix for each module of the newly developed product and the original shared platform product.

[0028]

[0029] Where the value b in the j-th row of the matrix j This is defined as the state coefficient indicating whether the j-th module has changed compared to the original shared platform product in the newly developed product. If there is a change, the state coefficient is defined as 1; otherwise, it is defined as 0.

[0030] Step 403: Determine the change status matrix of each module of the newly developed product and the original shared platform product. Based on the change status table of each module of the newly developed product and the original shared platform product and equation (2), the change status matrix of each module of the newly developed product and the original shared platform product can be determined.

[0031] Step 5: Calculate the test stress priority coefficient matrix.

[0032] Step 501: Define the test stress priority coefficient matrix. Using the number of test stresses m as the number of rows, define an m×1 matrix C as the test stress priority coefficient matrix.

[0033]

[0034] Wherein, the value c in the i-th row of the matrix i Defined as the priority coefficient for the i-th test stress of the common platform product;

[0035] Matrix C can be obtained by multiplying matrices A and B.

[0036] Step 502: Calculate the test stress priority coefficient matrix.

[0037] According to the definition that matrix C can be obtained by multiplying matrix A and matrix B, we get the following formula:

[0038] C = A × B (4)

[0039]

[0040] According to the rules of matrix multiplication, we can obtain:

[0041] c i =a i1 ×b1+a i2 ×b² + … + a in ×b n (6)

[0042] In equation (6), c iThis is the priority coefficient for the i-th test stress. If c i A priority coefficient of 0 means that the product of the sensitivity coefficient of each module and the i-th test stress and the change state coefficient of that module is 0. This indicates that the change in the newly developed product is unrelated to the i-th stress, and therefore the test stress is a non-sensitive stress for the new product and can be disregarded. Similarly, if the priority coefficient is greater than or equal to 1, it indicates that the test stress is a sensitive stress related to the change in the new product and needs to be selected. A higher priority coefficient indicates a greater correlation between the new product and the test stress, and therefore it should be selected and given special attention in the test plan.

[0043] Step 6: Determine the test plan based on the test stress priority coefficient matrix.

[0044] Based on the definition and calculation results of the test stress priority coefficient matrix in step 5, each test stress can be matched with its corresponding priority coefficient. Based on the definition of the priority coefficient, it can be determined whether the stress should be selected in the test plan for the newly developed product, and whether it should be a stress of key concern.

[0045] The advantages of this invention are: By utilizing test information from similar products on a shared platform, and by defining a stress sensitivity coefficient matrix and a change state matrix for each module, a test stress priority coefficient matrix is ​​calculated. This matrix can quantitatively describe the correlation between the newly developed product and each test stress, thereby allowing for targeted selection of stresses. This enables the tailoring of test items for the newly developed product, shortens the development cycle of the shared platform product, reduces test costs, and ultimately enhances the product's competitiveness. Attached Figure Description

[0046] Figure 1 Determine the process for existing reliability testing protocols

[0047] Figure 2 This is a flowchart of the reliability test design method for products based on a common platform according to the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the various embodiments of the present invention will be described in detail below with reference to the accompanying drawings. However, those skilled in the art will understand that many technical details have been presented in the various embodiments of the present invention to enable the reader to better understand the present invention. However, the technical solutions claimed in the present invention can be implemented even without these technical details and various changes and modifications based on the following embodiments.

[0049] This invention discloses a reliability test design method for products based on a common platform, as shown in the appendix. Figure 2 The specific implementation steps are shown below:

[0050] Step 1: Define the common platform product modules. Based on the actual design of the common platform product, agree on the levels to be analyzed, thus completing the module definition. Generally, select the module levels that may change during design modifications. List the constituent modules of the common platform product determined according to the module definitions: Module 1, Module 2, ..., Module n, where n is the number of modules.

[0051] Step 2: Determine the test stress.

[0052] Step 201: Determine the type of test stress. Analyze the usage environment profile and identify environmental stresses that have a significant impact on the common platform product, such as temperature, vibration, electrical stress, and humidity.

[0053] Step 202: Determine the test stress level. Based on the overall design of the common platform product or the user's requirements, select the test standard and determine the test stress level according to the indicators proposed in the usage profile or overall design.

[0054] Step 3: Determine the stress sensitivity coefficient matrix of each module of the common platform product and the test stress.

[0055] Step 301: Develop a correlation table between modules and test stress. Based on past testing experience or expert evaluation, determine the correlation between each module of the common platform product and the test stress. If a module is affected by the corresponding test stress, it is considered correlated; otherwise, it is considered uncorrelated. A correlation table can be developed as shown in Table 1 below:

[0056] Table 1. Correlation between Module and Test Stress

[0057]

[0058]

[0059] Step 302: Define the stress sensitivity coefficient matrix. Using the number of test stresses m as the number of rows and the number of modules n as the number of columns, define an m×n matrix A as the stress sensitivity coefficient matrix for each module of the common platform product and the test stresses.

[0060]

[0061] Where m is the experimental stress number;

[0062] n: the number of modules;

[0063] The value a in the i-th row and j-th column of the matrix ij : Defined as the stress sensitivity coefficient between the j-th module and the i-th test stress of the common platform product. If the module and the test stress are related, the stress sensitivity coefficient is defined as 1; if the module and the test stress are not related, the stress sensitivity coefficient is defined as 0.

[0064] Step 303: Determine the stress sensitivity coefficient matrix of each module of the common platform product and the test stress. According to Table 1 and Equation (7), the stress sensitivity coefficient matrix of each module of the common platform product and the test stress can be determined.

[0065] Step 4: Determine the change status matrix of each module of the newly developed product and the original shared platform product.

[0066] Step 401: Create a change status table for each module of the newly developed product and the original shared platform product. Based on the changes to each module of the newly developed product and the original shared platform product, the table below (Table 2) can be created:

[0067] Table 2. Change Status Table of Each Module for Newly Developed Products and Existing Shared Platform Products

[0068]

[0069] Step 402: Define the change state matrix for each module of the newly developed product and the original shared platform product. Using the number of modules n as the number of rows, define an n×1 matrix B as the change state matrix for each module of the newly developed product and the original shared platform product.

[0070]

[0071] Where the value b in the j-th row of the matrix j This is defined as the state coefficient indicating whether the j-th module has changed compared to the original shared platform product in the newly developed product. If there is a change, the state coefficient is defined as 1; otherwise, it is defined as 0.

[0072] Step 403: Determine the change status matrix of each module of the newly developed product and the original shared platform product. According to Table 2 and Equation (8), the change status matrix of each module of the newly developed product and the original shared platform product can be determined.

[0073] Step 5: Calculate the test stress priority coefficient matrix.

[0074] Step 501: Define the test stress priority coefficient matrix. Using the number of test stresses m as the number of rows, define an m×1 matrix C as the test stress priority coefficient matrix.

[0075]

[0076] Wherein, the value c in the i-th row of the matrix i Defined as the priority coefficient of the i-th test stress for a common platform product.

[0077] Matrix C can be obtained by multiplying matrices A and B.

[0078] Step 502: Calculate the test stress priority coefficient matrix.

[0079] According to the definition that matrix C can be obtained by multiplying matrix A and matrix B, we get the following formula:

[0080] C = A × B (10)

[0081]

[0082] According to the rules of matrix multiplication, we can obtain:

[0083] c i =a i1 ×b1+a i2 ×b² + … + a in ×b n (12)

[0084] In equation (12), c i That is, the priority coefficient of the i-th test stress if c i A priority coefficient of 0 means that the product of the sensitivity coefficient of each module and the i-th test stress, and the change coefficient of that module, is 0. This indicates that the change in the newly developed product is unrelated to the i-th stress, and therefore the test stress is a non-sensitive stress for the new product and can be disregarded. Similarly, if the priority coefficient is greater than or equal to 1, it indicates that the test stress is a sensitive stress related to the change in the new product and needs to be selected. Furthermore, the larger the priority coefficient, the greater the correlation between the new product and the test stress, and the more important it is to select and focus on it in the test plan.

[0085] Step 6: Determine the test plan based on the test stress priority coefficient matrix.

[0086] Based on the definition and calculation results of the test stress priority coefficient matrix in step 5, each test stress can be mapped to its corresponding priority coefficient. According to the definition of the priority coefficient, it can be determined whether the stress should be selected for the new product testing plan and whether it should be considered a key stress. The results are shown in Table 3 below.

[0087] Table 3 Test Plan for Newly Developed Products

[0088]

[0089] The steps of this invention will be explained below using a communication product from a certain company's shared platform as an example:

[0090] Background: A company has a series of common platform communication products. It is currently designing a test plan for a newly developed product. The original common platform similar products have undergone complete reliability testing.

[0091] Step 1, Product Module Definition: After evaluation, the primary components of this communication product are used as the module definition. The product is divided into 4 modules: chassis, power module, communication module, and display module.

[0092] Step 2, determine the test stresses. The common platform product is sensitive to five stresses: high temperature, low temperature, vibration, surge, and mold.

[0093] Step 3: Determine the stress sensitivity coefficient matrix for each module of the product and the test stress.

[0094] Based on past testing experience, the correlation table 4 between a certain communication product module and test stress is as follows:

[0095] Table 4 Correlation between a certain communication product module and test stress

[0096]

[0097]

[0098] Based on the definition of the stress sensitivity coefficient matrix and Table 4 above, a 5×4 matrix A is defined as the stress sensitivity coefficient matrix of each module and the test stress of this communication product, with the number of test stresses (5) as the number of rows and the number of modules (4) as the number of columns.

[0099]

[0100] Step 4: Determine the change status matrix of each module of the newly developed product and the original shared platform product.

[0101] The change status of each module of a newly developed communication product and the original shared platform product is shown in Table 5 below:

[0102] Table 5. Change Status Table of Each Module of a Newly Developed Communication Product and an Existing Shared Platform Product

[0103] Module chassis Power module Communication module Display module Has it been changed? change No changes No changes change

[0104] Based on the definition of the change state matrix for each module and Table 5 above, the number of product modules is 4, which can form a 4×1 change state matrix B for the modules of the newly developed product and the original shared platform product:

[0105]

[0106] Step 5: Calculate the test stress priority coefficient matrix.

[0107] According to the method of the present invention, the test stress priority coefficient matrix C can be obtained by the following formula:

[0108]

[0109] Step 6: Determine the test plan based on the test stress priority coefficient matrix.

[0110] Based on the definition of the test stress priority coefficient matrix and the results calculated in step 5, each test stress is mapped to its corresponding priority coefficient. Then, according to the definition of the priority coefficient, it is determined whether a particular stress should be selected for a test plan of a newly developed communication product, and whether it should be designated as a stress of key concern.

[0111] Table 6 Test Plan for a Newly Developed Communication Product

[0112] Test stress Priority coefficient Test plan during this product change high temperature 2 Select this stress and focus on Low temperature 1 Select this stress vibration 2 Select this stress and focus on surge 0 Do not select this stress mold 1 Select this stress

[0113] The results are shown in Table 6 above.

Claims

1. A reliability test design method based on a common platform product, wherein the common platform product refers to a series of products based on the same hardware design platform, whose module composition is the same or similar, and whose environmental stress in application is the same or similar; characterized in that: The steps of this method are as follows: Step 1: Define common platform product modules, listing each component module of the product as determined by the module definition: Module 1, Module 2, ..., Module n， in n Number of modules; Step 2, determine the test stress, including determining the type and level of the test stress; Step 3: Determine the stress sensitivity coefficient matrix for each module of the common platform product and the test stress; Step 4: Determine the change status matrix for each module of the newly developed product and the original shared platform product; Step 5: Calculate the test stress priority coefficient matrix; Step 6: Determine the test plan based on the test stress priority coefficient matrix; Based on the definition and calculation results of the test stress priority coefficient matrix in step 5, each test stress is matched with its corresponding priority coefficient. Based on the definition of the priority coefficient, it is determined whether the stress should be selected in the test plan for the newly developed product, and whether it should be regarded as a key stress.

2. The reliability test design method based on a common platform product according to claim 1, characterized in that: The specific process of step 3 is as follows: Step 301: Develop a correlation table between modules and test stresses; Based on past test experience or through expert evaluation, determine the correlation between each module of the product and the test stress. If a module is affected by the corresponding test stress, it is considered to be related; if it is not affected, it is considered to be unrelated. Step 302, define the stress sensitivity coefficient matrix; using the experimental stress number m The number of rows is represented by the number of modules. n For the number of columns, set a matrix A , is defined as the stress sensitivity coefficient matrix of each module of the common platform product and the test stress; （1） in, m : represents the experimental stress number; n : Number of modules; Matrix number i Line number j Column values a ij Defined as a common platform product j The module and the first i The stress sensitivity coefficient between test stresses is defined as 1 if the module and the test stress are related, and 0 if the module and the test stress are not related. Step 303: Determine the stress sensitivity coefficient matrix of each module of the common platform product and the test stress; according to the correlation table of modules and test stress and formula (1), determine the stress sensitivity coefficient matrix of each module of the product and the test stress.

3. The reliability test design method for a product based on a common platform according to claim 2, characterized in that: The specific process of step 4 is as follows: Step 401: Based on the changes to each module of the newly developed product and the original common platform product, formulate a change status table for each module of the newly developed product and the original common platform product; Step 402: Define the change status matrix for each module of the newly developed product and the original shared platform product; By number of modules n Set a number for the number of rows. matrix B , defined as the change status matrix of each module of the newly developed product and the original common platform product; （2） Among them, the matrix number j row value b j Defined as the newly developed product compared to the original shared platform product, the first j Does each module have a changed state coefficient? If it has changed, the state coefficient is defined as 1; if it has not changed, the state coefficient is defined as 0. Step 403: Determine the change status matrix of each module of the newly developed product and the original common platform product; based on the change status table of each module of the newly developed product and the original common platform product and formula (2), determine the change status matrix of each module of the newly developed product and the original common platform product.

4. The reliability test design method for a product based on a common platform according to claim 3, characterized in that: The specific process of step 5 is as follows: Step 501, define the test stress priority coefficient matrix; based on the test stress number m Set a number for the number of rows. m A matrix of ×1 C , is defined as the test stress priority coefficient matrix; （3） Among them, the first in the matrix i row value c i Defined as a common platform product i Priority coefficient for each test stress; matrix C From the matrix A sum matrix B The result is obtained by multiplication. Step 502: Calculate the test stress priority coefficient matrix; Based on matrix C, from matrix A sum matrix B The definition obtained by multiplication is as follows: （4） （5） According to the rules of matrix multiplication, we get: （6） In formula (6) c i That is, the first i If the priority coefficient of each test stress is c i A value of 0 means that each module and the first i The product of the sensitivity coefficient of the test stress and the change state coefficient of the module is 0, meaning that the change in this newly developed product and the first... i If a stress is not related, it means that the test stress is a non-sensitive stress of the newly developed product and should not be selected. Similarly, if the priority coefficient is greater than or equal to 1, it means that the test stress is a sensitive stress of the newly developed product in this change and needs to be selected. The larger the priority coefficient, the greater the correlation between the newly developed product and the test stress, and it should be selected and given special attention in the test plan.

Images