A method and system for identifying average life assessment of batches of electronic products

Abstract

The application provides a kind of to identify batch electronic product average life evaluation method, it is related to the technical field of reliability engineering and life evaluation, solve the technical problem that average life evaluation method in prior art is difficult to accurately evaluate the average life of the batch electronic product.The method comprises establishing a segmented life distribution model, the segmented life distribution model includes a first stage life distribution model and a second stage life distribution model separated by a segmentation point time T, and the segmented life distribution model satisfies the parameter constraint relationship at the segmentation point time T;Wherein, the first stage life distribution model adopts an exponential distribution model with failure rate λ;The second stage life distribution model adopts a Weibull distribution model;Based on the segmented life distribution model, the average life of the batch electronic product is calculated and obtained by applying the average life analytical expression.The application can improve the calculation efficiency and calculation accuracy.

Description

Technical Field

[0001] This invention relates to the technical field of reliability engineering and life assessment, and in particular to a method and system for assessing the average life of a batch of electronic products. Background Technology

[0002] In electronic product reliability assessment, mean lifespan is a key indicator that directly impacts equipment maintenance strategies and spare parts planning. Certification batches of electronic products refer to those in the certification stage. These products have been used for a certain period, and their lifespan has stabilized. For the assessment of certification batches, industry practice often assumes a constant overall failure rate and uses an exponential distribution model for lifespan evaluation. This assumption stems from the fact that the product has passed environmental stress screening and reliability qualification tests, effectively controlling early failures. However, even with early failures eliminated, products inevitably encounter physical and chemical wear processes such as solder joint fatigue and insulation aging during later use, leading to an increasing failure rate over time. This phenomenon aligns with the classic "bathtub curve" theory (see below). Figure 1 This indicates that the product is simultaneously in the overlapping stage of the accidental failure period and the wear-out failure period.

[0003] There are two main types of commonly used life assessment methods in the current technology: one is the single exponential distribution model, which ignores later-stage wear and tear failures, leading to an overestimation of the average lifespan and potentially causing insufficient spare parts reserves or delays in maintenance plans; the other is the single Weibull distribution model, which, while able to describe the increasing failure rate stage, may overfit later-stage data, resulting in an underestimation of the average lifespan and wasting resources. Neither of these models can accurately characterize the complex characteristics of different failure mechanisms in the early and late stages of a batch of electronic products, leading to significant biases in the assessment results and affecting the scientific rigor of subsequent support decisions.

[0004] Therefore, there is an urgent need for a life assessment method that can simultaneously describe the failure rate variation characteristics of a product during both the incidental failure period and the wear-out failure period, in order to improve the accuracy and engineering applicability of mean life assessment. Summary of the Invention

[0005] This invention provides a method and system for assessing the average lifespan of a batch of electronic products, in order to solve the problem that existing average lifespan assessment methods are difficult to accurately assess the average lifespan of a batch of electronic products.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0007] A method for assessing the average lifespan of a batch of electronic products includes the following steps:

[0008] A segmented lifetime distribution model is established, comprising a first-stage lifetime distribution model and a second-stage lifetime distribution model separated by a segmentation point time T, wherein the segmented lifetime distribution models satisfy parameter constraints at the segmentation point time T; wherein, the first-stage lifetime distribution model adopts an exponential distribution model with a failure rate of λ to characterize the random failure period of the electronic products in the evaluation batch with a constant failure rate; the second-stage lifetime distribution model adopts a Weibull distribution model to characterize the wear-out failure period of the electronic products in the evaluation batch with an increasing failure rate.

[0009] Based on the segmented lifetime distribution model, the average lifetime of the electronic products in the evaluation batch is calculated and obtained by applying the average lifetime analytical expression.

[0010] Preferably, the method for determining the segmentation point time T includes:

[0011] Historical failure data of the batch of electronic products being evaluated are collected and analyzed to obtain an empirical failure rate function, which represents the trend of failure rate over time.

[0012] The first time point at which the first derivative of the empirical failure rate function exceeds a preset threshold is determined as the segmentation time T.

[0013] Preferably, the cumulative fault distribution function of the segmented lifetime distribution model is defined as:

[0014]

[0015] in:

[0016] λ is the failure rate of the exponential distribution;

[0017] t represents working hours;

[0018] T represents the time at the breakpoint;

[0019] γ is the location parameter of the Weibull distribution, representing the minimum lifetime / guaranteed lifetime;

[0020] η is the scaling parameter of the Weibull distribution;

[0021] β is the shape parameter of the Weibull distribution.

[0022] Preferably, the parameter constraint relationship is as follows: at the segment point time T, the cumulative fault distribution function value of the exponential distribution model is equal to the cumulative fault distribution function value of the Weibull distribution model.

[0023] Preferably, the shape parameter β of the Weibull distribution is greater than 1.

[0024] Preferably, in the parameters of the cumulative fault distribution function of the segmented lifetime distribution model, the segment time T is greater than the location parameter γ of the three-parameter Weibull distribution.

[0025] Preferably, the analytical expression for the average lifetime is obtained by incomplete gamma function integration.

[0026] Preferably, the analytical expression for the average lifetime is calculated as follows:

[0027] The lifetime values ​​for the exponential distribution stage and the Weibull distribution stage are calculated respectively according to the segmented lifetime distribution model.

[0028] The lifetime values ​​for the exponential distribution phase are as follows:

[0029]

[0030] The lifetime value for the Weibull distribution stage is calculated using the following formula:

[0031]

[0032] The lifetime value calculation formula for the Weibull distribution stage uses t=(t-γ)+γ to split the integral into two parts:

[0033]

[0034] Define a key variable, let:

[0035] The lower limit of integration is:

[0036] The upper limit of integration corresponds to u→+∞;

[0037] Apply variable substitution; the result is:

[0038] t-γ=ηu,d t =ηd u

[0039] The integral of the first part is calculated as follows:

[0040]

[0041] Substitute the cumulative fault distribution function and simplify:

[0042]

[0043] Let v = u β That is, u = v 1 / β , The integral is converted to:

[0044]

[0045] in, It is an incomplete gamma function;

[0046] The integral of the second part is calculated as follows:

[0047]

[0048] The survival function of the Weibull distribution is:

[0049]

[0050] therefore:

[0051]

[0052] Therefore, the average lifespan of the batch of electronic products being tested is:

[0053]

[0054] Where Γ(·,·) is an incomplete gamma function.

[0055] Preferably, the Weibull distribution model is a two-parameter Weibull distribution model, having a scale parameter η and a shape parameter β.

[0056] To address the aforementioned technical problems, the present invention also provides a system for assessing the average lifespan of electronic products in a batch of inspection, used to implement the aforementioned method for assessing the average lifespan of electronic products in a batch of inspection, comprising:

[0057] At least one processor;

[0058] and a memory connected to the at least one processor;

[0059] The memory stores computer program instructions, which, when executed by the at least one processor, implement the method.

[0060] The beneficial effects of this invention are as follows: This invention provides a method and system for assessing the average lifespan of a batch of electronic products, which has the following advantages:

[0061] (1) The accuracy of the assessment is significantly improved: the segmented distribution model is adopted, which is more in line with the actual situation of different failure mechanisms in the early and late stages of the batch of electronic products and overcomes the problem of overestimating or underestimating the average life of the single distribution model.

[0062] (2) High computational efficiency: By establishing parameter constraint relationships and incomplete gamma function integration, the analytical expression of average lifetime is obtained. No complex numerical simulation is required, the calculation speed is fast, and it is suitable for real-time engineering evaluation.

[0063] (3) Strong applicability: The method is based on classical probability distribution and mature mathematical tools, and is easy to integrate and apply in the existing reliability assessment system.

[0064] (4) More accurate decision support: Provides more realistic average lifespan estimates, offering reliable input for logistical decisions such as maintenance, spare parts supply, etc., thereby improving equipment availability and economy. Attached Figure Description

[0065] Figure 1 This is a schematic diagram of the "bathtub curve" of electronic product failure rates.

[0066] Figure 2 This is a flowchart of a method for assessing the average lifespan of a batch of electronic products provided by the present invention;

[0067] Figure 3 This is a schematic diagram of the structure of an average life assessment system according to an embodiment of the present invention;

[0068] Figure 4 This is a comparative diagram of the evaluation results of different models used to illustrate the technical effects of the present invention.

[0069] In the figure: 10 is the data acquisition module; 20 is the parameter estimation module; 30 is the constraint solving module; 40 is the life calculation module; 50 is the result output module; 60 is the processor; 70 is the memory; S201 is the step of collecting historical fault data; S202 is the step of determining the segment point time T; 410 is the single exponential distribution model; 420 is the model of this invention; 430 is the single Weibull distribution model. Detailed Implementation

[0070] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of the present invention.

[0071] The terminology used in the embodiments of this application is for the purpose of describing the embodiments of this application only and is not intended to limit this application. Before further detailed description of the embodiments of this application, the nouns and terms involved in the embodiments of this application are explained, and the nouns and terms involved in the embodiments of this application are subject to the following interpretations.

[0072] (1) Qualification batch of electronic products: refers to a batch of electronic products that have passed environmental stress screening, reliability qualification tests, and other processes, and whose early failure problems have been effectively controlled, and have entered the accidental failure period and wear-out failure period. The failure rate characteristics of this type of product are consistent with the technical problem scenario to be solved in this application.

[0073] (2) Bathtub curve: such as Figure 1 The figure shows a classic curve describing the change in product failure rate over time, including an early failure period with a high failure rate, an occasional failure period (also known as a random failure period) with a relatively low and constant failure rate, and a wear-out failure period where the failure rate increases over time. The technical solution of this application is based on accurate modeling of the occasional failure period and the wear-out failure period.

[0074] (3) Segmented Lifetime Distribution Model: This refers to a composite probability distribution model divided by a segment time T. In the embodiments of this application, the first stage (0≤t<T), i.e., the accidental failure period, adopts an exponential distribution model; the second stage (t≥T), i.e., the wear-out failure period, adopts a Weibull distribution model. This segmentation method can more realistically reflect the failure pattern of the entire product life cycle.

[0075] (4) Analytical expression for mean lifespan: This refers to a closed-form mathematical formula used to calculate the mean lifespan of a product. This formula does not involve integration, limits, or iterative solutions. The evaluator only needs to substitute the known model parameters to directly obtain a unique numerical solution through a finite number of algebraic operations. This is a key characteristic that distinguishes it from computationally intensive methods that rely on numerical integration or Monte Carlo simulation.

[0076] (5) Incomplete Gamma Function: This is a generalization of the gamma function, defined as Γ(α, x)=∫[x,∞]t(d-1)e(-t)dt. In the embodiments of this application, this function is a key mathematical tool that can transform the reliability function integral of the Weibull distribution into a closed-form analytical solution, and is the theoretical basis for achieving efficient computation.

[0077] (6) Position parameter γ: In a three-parameter Weibull distribution, this parameter represents the minimum or guaranteed life of the product. Physically, it means that the product will not experience wear-out failure before this point in time; the wear-out process only begins at time t = γ. The existence of this parameter allows the model to more flexibly fit wear-out processes with a defined start-up delay.

[0078] (7) Scale parameter η: In the Weibull distribution, this parameter is also called characteristic lifetime. When t-γ = η, the cumulative failure probability of the product reaches a fixed value (approximately 63.2%). It is numerically related to the average lifetime of the product and reflects the dispersion of the product's lifetime.

[0079] (8) Shape parameter β: In the Weibull distribution, this parameter determines the shape of the failure rate curve, thus describing the failure mode of wear-out failure period. When β > 1, it indicates that the failure rate increases with time, which is consistent with the typical product aging and wear pattern. This is the wear-out failure period characteristic that this application focuses on.

[0080] See attached document Figure 2 This invention provides a method for assessing the average lifespan of a batch of electronic products. It aims to address the technical problems of existing technologies that use a single distribution model to assess lifespan, resulting in large deviations in assessment results due to the inability to accurately characterize the complex failure characteristics of a product from the accidental failure period to the wear-out failure period, and the inefficiency of some segmented models that rely on numerical calculations. The core idea of ​​this method is to construct a segmented probability model that better reflects physical reality and derive an analytical expression for the average lifespan of this model, thereby achieving accurate and efficient lifespan assessment.

[0081] The present invention specifically includes the following steps:

[0082] First, a segmented lifespan distribution model is established. This model divides the product's lifespan into two stages separated by a segmentation point T. The first stage corresponds to the product's accidental failure period, characterized by a relatively constant failure rate caused by external random events. The second stage corresponds to the product's wear-out failure period, characterized by a significantly increasing failure rate over time due to cumulative damage from aging and wear of internal components. To accurately characterize these two distinctly different physical processes, this invention employs an exponential distribution model with a constant failure rate λ in the first stage lifespan distribution model and a Weibull distribution model that describes the increasing failure rate in the second stage lifespan distribution model. This model combination allows for a more accurate fit than any single model. Figure 1 The accidental failure and wear-out phases shown in the "bathtub curve" are the fundamental reason for improving the accuracy of the assessment.

[0083] Then, based on the segmented lifetime distribution model, the average lifetime analytical expression is applied to calculate and obtain the average lifetime of the batch of electronic products.

[0084] In this embodiment of the invention, the method for determining the segmentation point time T includes:

[0085] Historical failure data of the batch of electronic products being evaluated are collected and analyzed to obtain an empirical failure rate function, which represents the trend of failure rate over time.

[0086] The first time point at which the first derivative of the empirical failure rate function exceeds a preset threshold is determined as the segmentation time T.

[0087] The design principle is that the first derivative of the failure rate represents the rate of change of the failure rate. When a product transitions from a period of random failure with a constant failure rate to a period of wear and tear failure with an increasing failure rate, this derivative value will experience a significant jump. By setting a threshold to capture this "jump" point, the T value can be determined objectively and reproducibly, avoiding the uncertainty and arbitrariness brought about by relying solely on subjective engineering experience, thereby improving the scientificity and reliability of the entire evaluation process.

[0088] In this embodiment of the invention, the lifetime distribution of the batch of electronic products is composed of an exponential distribution in the first segment and a three-parameter Weibull distribution in the second segment, and its cumulative failure distribution function is defined as:

[0089]

[0090] in:

[0091] λ: Failure rate of exponential distribution;

[0092] f: Working hours;

[0093] T: Breakpoint time;

[0094] γ: Weibull distribution location parameter, minimum lifetime / guaranteed lifetime;

[0095] η: Scaling parameter of the Weibull distribution;

[0096] β: Shape parameter of the Weibull distribution, β > 1.

[0097] In this embodiment of the invention, to ensure the continuity of the distribution function at the segmentation point time T, a parameter constraint relationship must be satisfied at the segmentation point time T. The parameter constraint relationship is as follows:

[0098]

[0099] Simplifying, we get:

[0100]

[0101] Meanwhile, due to the non-negativity of the parameters, T > γ should be satisfied, that is, the boundary time point T of the combined distribution should be greater than the location parameter γ of the three-parameter Weibull distribution; and since the later failure mode is a wear-out failure mode, the parameter β > 1 should be guaranteed to ensure that the failure rate in the later stage is increasing.

[0102] In this embodiment of the invention, the analytical expression for average lifetime is obtained by incomplete gamma function integration. Specifically:

[0103] Calculate the lifetime values ​​for the exponential distribution segment and the Weibull distribution segment according to the distribution model.

[0104] The lifetime values ​​for the exponential distribution segment are as follows:

[0105]

[0106] The formula for calculating the lifetime value of the Weibull distribution segment is as follows:

[0107]

[0108] (1) Integral decomposition

[0109] Using t = (t - γ) + γ, the integral can be split into two parts:

[0110]

[0111] (2) Variable substitution and incomplete gamma function

[0112] 1) Define key variables, let:

[0113]

[0114] The lower limit of integration is:

[0115]

[0116] The upper limit of integration corresponds to u→+∞;

[0117] After replacement:

[0118] f-γ=ηu,d t =ηd u

[0119] 2) First part of the integration:

[0120]

[0121] Substitute the cumulative fault distribution function and simplify:

[0122]

[0123] Let v = u β (i.e., u = v) 1 / β , The integral is converted to:

[0124]

[0125] in, It is an incomplete gamma function.

[0126] 3) Part Two: Integral

[0127]

[0128] The survival function of the Weibull distribution is:

[0129]

[0130] therefore:

[0131]

[0132] Therefore, the average lifespan of the batch of electronic products being tested is:

[0133]

[0134] In an alternative implementation, the Weibull distribution model can also be a simpler two-parameter Weibull distribution model, which has a scale parameter η and a shape parameter β, and its position parameter γ is set to 0. This model is suitable for electronic products where the wear and tear process can be considered to be accumulating from the beginning of product use (t=0), i.e., there is no absolutely wear-free "safe period". Using a two-parameter model can reduce one parameter to be estimated, simplifying the model structure and parameter estimation process. This implementation is designed to broaden the applicability of the method of the present invention, enabling it to cover more types of product failure modes, and providing a choice between accuracy and simplicity.

[0135] like Figure 2 and Figure 3 As shown, this invention provides a specific embodiment of a method for assessing the average lifespan of a batch of electronic products, aiming to accurately and efficiently assess the average lifespan of a batch of electronic products. The core of this method lies in establishing a segmented lifespan distribution model that accurately reflects the actual failure patterns of the products, and applying an analytical expression for average lifespan that does not require numerical integration for calculation.

[0136] In a basic embodiment, the method first executes step S201 to collect historical failure data of the target electronic product. Subsequently, the core of the method is to establish a segmented lifespan distribution model. This model divides the product's lifespan into two stages at the segmentation point T. The first stage (0 ≤ t < T) is defined as the random failure period, where failures are considered to be caused by random events, and its failure rate λ is assumed to be constant; therefore, an exponential distribution model is used to describe it. The second stage (t ≥ T) is defined as the wear-out failure period, where failures are mainly caused by the aging and wear accumulation of internal components, and its failure rate increases over time; therefore, a Weibull distribution model is used to describe it. This segmented modeling approach decomposes the complex full-lifespan failure behavior into two simple stages with clear physical mechanisms that are easy to model, thereby achieving an accurate approximation of the actual situation overall.

[0137] See Figure 3The present invention also provides a system 300 for assessing the average lifespan of a batch of electronic products. This system is a hardware carrier for implementing any of the above-described methods. The system can be a general-purpose computer, an embedded device, or a dedicated testing instrument. System 300 includes at least a processor 60 and a memory 70 connected to the processor 60. The memory 70 stores computer program instructions, which, when executed by the processor 60, enable the implementation of the aforementioned average lifespan assessment methods.

[0138] Specifically, the system includes a data acquisition module 10, responsible for acquiring historical lifespan data or failure time data of electronic products from external databases, files, or sensor interfaces. The data acquisition module 10 transmits the acquired data to the parameter estimation module 20. The parameter estimation module 20 is one of the core computing units of the system, and it incorporates various statistical estimation algorithms. For example, it can use the maximum likelihood estimation method to estimate the failure rate λ of the exponential distribution using data before the segment point T, and to estimate the shape parameter β and scale parameter η of the Weibull distribution using data after T. Furthermore, the parameter estimation module 20 can also integrate the aforementioned objective method for determining the segment point T, i.e., determining the value of T by analyzing data, fitting an empirical failure rate function, and taking its derivative.

[0139] The parameter estimation module 20 passes the estimated parameters (λ, T, η, β) to the constraint solving module 30. The constraint solving module 30 applies the aforementioned parameter constraint relationship, λT = ((T-γ) / η)^β, to solve for the remaining unknown parameters in the model, typically the location parameter γ. It solves for the value of γ through algebraic operations, thereby determining all parameters of the entire segmented lifetime distribution model.

[0140] Subsequently, all determined model parameters (λ, T, η, β, γ) are fed into the lifespan calculation module 40. The lifespan calculation module 40 internally contains the core achievement of this invention—the analytical expression for average lifespan. It automatically selects whether to apply the formula corresponding to the three-parameter model or the simplified formula corresponding to the two-parameter model based on whether the input parameter Y is 0. This module calls a standard mathematical library function for calculating incomplete gamma function values, ultimately calculating the product's average lifespan E.

[0141] Finally, the lifespan calculation module 40 transmits the calculated average lifespan value E to the result output module 50. The result output module 50 is responsible for presenting the final result in a user-friendly manner, such as displaying the value on a graphical user interface screen, generating a PDF report containing a detailed evaluation process and results, or providing the result to other management systems (such as a spare parts management system or a maintenance decision support system) via a network interface. Throughout the process, the processor 60 is responsible for executing instructions from the various modules in the memory 70, coordinating the flow of data between modules, and ultimately completing the entire evaluation task.

[0142] To more clearly illustrate the implementation process and technical effects of the present invention, a specific embodiment will be described below. This embodiment incorporates several preferred technical features of the present invention and represents a preferred implementation of the present invention. Suppose it is necessary to evaluate the average lifespan of a batch of marine DC uninterruptible power supplies.

[0143] 1. Data and Parameter Settings

[0144] (1) The failure rate of the exponential distribution in the first segment is λ = 0.05 times / year;

[0145] (2) The segmentation time is 10 years. After 10 years of use by users, the failure rate increases significantly.

[0146] (3) The Weibull distribution parameters for the latter part are β = 2.5 and η = 8 years;

[0147] The position parameter γ of the Weibull distribution in the latter part is constrained by parameters. The solution was performed, and γ was calculated to be 3.94 years.

[0148] 2. Average lifespan calculation

[0149] Substitute into the analytical formula to calculate:

[0150]

[0151] 3. Comparative Analysis

[0152] (1) If the simplified treatment assumes that the failure rate follows a single exponential distribution, its average lifetime is...

[0153] (2) If the simplified treatment assumes that the failure rate follows a single three-parameter Weibull distribution, its mean lifetime is:

[0154]

[0155] The comparison of average lifetime assessment results for each distribution is shown in the table below:

[0156] Table 1. Differences in Average Lifetime Assessment Results for Different Distributions

[0157] 1 Combination distribution 16.64 / 2 Exponential distribution 20 +20.19％ 3 Three-parameter Weibull distribution 11.03 -33.71％

[0158] Therefore, it is evident from the analysis results of the product lifespan data of the above-mentioned batch of products that, compared with a single distribution model, a combined distribution can effectively improve the accuracy of average lifespan assessment.

[0159] like Figure 4 As shown, if the traditional single exponential distribution model 410 is used for evaluation, the average lifespan is estimated to be 1 / 0.05 = 20 years, an overestimation of 20.19%. If a single three-parameter Weibull distribution model 430 is used, the average lifespan is estimated to be approximately 11.03 years, an underestimation of 33.71%. The 16.64-year result obtained by the model 420 of this invention falls between the two, effectively overcoming the evaluation bias of the single model. This high-precision and high-efficiency evaluation method provides reliable data support for subsequent maintenance strategy formulation and spare parts inventory planning.

[0160] In the description of this specification, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0161] Of course, the present invention may have other various embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art can make various corresponding changes and modifications according to the present invention, but these changes and modifications should all fall within the protection scope of the claims of the present invention.

Claims

1. A method for assessing the average lifespan of a batch of electronic products, characterized in that, Specifically, the steps include the following: A segmented lifetime distribution model is established, comprising a first-stage lifetime distribution model and a second-stage lifetime distribution model separated by a segmentation point time T, wherein the segmented lifetime distribution models satisfy parameter constraints at the segmentation point time T; wherein, the first-stage lifetime distribution model adopts an exponential distribution model with a failure rate of λ to characterize the random failure period of the electronic products in the evaluation batch with a constant failure rate; the second-stage lifetime distribution model adopts a Weibull distribution model to characterize the wear-out failure period of the electronic products in the evaluation batch with an increasing failure rate. Based on the segmented lifetime distribution model, the average lifetime of the electronic products in the evaluation batch is calculated and obtained by applying the average lifetime analytical expression.

2. The method for assessing the average lifespan of electronic products in a batch according to claim 1, characterized in that, The method for determining the segmentation point time T includes: Historical failure data of the batch of electronic products being evaluated are collected and analyzed to obtain an empirical failure rate function, which represents the trend of failure rate over time. The first time point at which the first derivative of the empirical failure rate function exceeds a preset threshold is determined as the segmentation time T.

3. The method for assessing the average lifespan of electronic products in a batch according to claim 2, characterized in that, The cumulative fault distribution function of the segmented lifetime distribution model is defined as: in: λ is the failure rate of the exponential distribution; t represents working hours; T represents the time at the breakpoint; γ is the location parameter of the Weibull distribution, representing the minimum lifetime / guaranteed lifetime; η is the scaling parameter of the Weibull distribution; β is the shape parameter of the Weibull distribution.

4. The method for assessing the average lifespan of electronic products in a batch according to claim 3, characterized in that, The parameter constraint relationship is as follows: at the segment point time T, the cumulative fault distribution function value of the exponential distribution model is equal to the cumulative fault distribution function value of the Weibull distribution model.

5. The method for assessing the average lifespan of a batch of electronic products according to claim 3, characterized in that, The shape parameter β of the Weibull distribution is greater than 1.

6. The method for assessing the average lifespan of electronic products in a batch according to claim 5, characterized in that, In the parameters of the cumulative fault distribution function of the segmented lifetime distribution model, the segment time T is greater than the location parameter γ of the three-parameter Weibull distribution.

7. The method for assessing the average lifespan of a batch of electronic products according to any one of claims 3-6, characterized in that, The analytical expression for the average lifetime is obtained by incomplete gamma function integration.

8. The method for assessing the average lifespan of electronic products in a batch according to claim 7, characterized in that, The method for calculating the analytical expression of the average lifetime is as follows: The lifetime values ​​for the exponential distribution stage and the Weibull distribution stage are calculated respectively according to the segmented lifetime distribution model. The lifetime values ​​for the exponential distribution phase are as follows: The lifetime value for the Weibull distribution stage is calculated using the following formula: The lifetime value calculation formula for the Weibull distribution stage uses t=(t-γ)+γ to split the integral into two parts: Define a key variable, let: The lower limit of integration is: The upper limit of integration corresponds to u→+∞; Apply variable substitution; the result is: t-γ=ηu,d t =ηd u The integral of the first part is calculated as follows: Substitute the cumulative fault distribution function and simplify: Let v = u β That is, u = v 1 / β , The integral is converted to: in, It is an incomplete gamma function; The integral of the second part is calculated as follows: The survival function of the Weibull distribution is: therefore: Therefore, the average lifespan of the batch of electronic products being tested is: Where Γ(·,·) is an incomplete gamma function.

9. The method for assessing the average lifespan of a batch of electronic products according to claim 1, characterized in that, The Weibull distribution model is a two-parameter Weibull distribution model, with a scale parameter η and a shape parameter β.

10. A system for assessing the average lifespan of electronic products in a batch, used to implement the method for assessing the average lifespan of electronic products in a batch as described in any one of claims 1-9, characterized in that, include: At least one processor; and a memory connected to the at least one processor; The memory stores computer program instructions, which, when executed by the at least one processor, implement the method as described in any one of claims 1-9.

Images