Quality inspection passing rate attribution analysis method, program product, equipment and storage medium

Abstract

The invention provides a quality inspection passing rate attribution analysis method, a program product, equipment and a storage medium, and the method comprises the steps: calculating a Lorentz curve coefficient corresponding to an attribution dimension based on historical quality inspection data and current quality inspection data under each attribution dimension; the Lorentz curve coefficient is used for representing the fluctuation degree of the quality inspection passing rate under the attribution dimension; determining a target attribution dimension from the attribution dimensions based on the Lorentz curve coefficients of the attribution dimensions; calculating contribution degree data of the target attribution dimension according to the historical quality inspection data and the current quality inspection data; and according to the contribution degree data, determining the reason causing the fluctuation of the quality inspection passing rate under the target attribution dimension. By calculating the Lorentz curve coefficient, the target attribution dimension which should be disassembled most is quickly determined from the plurality of attribution dimensions, and the efficiency of cause analysis is improved. Therefore, targeted improvement is facilitated, and the quality inspection stability is improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a quality inspection pass rate attribution analysis method, program product, device and storage medium. Background Art

[0002] Quality inspection is an important part of warehouse operations. It mainly involves quality inspection staff conducting appearance quality inspections on the goods sent by sellers / merchants in accordance with standard operating instructions and quality inspection standards. Quality inspections ensure that the physical information of the goods is consistent with the system information, that the accessories are intact and complete, and that the goods are brand new and flawless. For goods that arrive in different batches or at different times, the pass rate of quality inspections may fluctuate, such as the pass rate of quality inspections becoming lower or higher. Current analysis tools cannot determine the reasons for the fluctuations in the pass rate of quality inspections. Summary of the invention

[0003] The purpose of the embodiments of the present application is to provide a quality inspection pass rate attribution analysis method, program product, device and storage medium, which are used to improve the above-mentioned problem that the cause of the fluctuation of the quality inspection pass rate cannot be determined.

[0004] In a first aspect, an embodiment of the present application provides a quality inspection pass rate attribution analysis method, including: based on historical quality inspection data and current quality inspection data under each attribution dimension, calculating the Lorenz curve coefficient corresponding to the attribution dimension; the attribution dimension refers to the classification factors that cause the quality inspection pass rate fluctuation; the Lorenz curve coefficient is used to characterize the degree of fluctuation of the quality inspection pass rate under the attribution dimension; based on the Lorenz curve coefficient of the attribution dimension, determining the target attribution dimension from the attribution dimension; according to the historical quality inspection data and the current quality inspection data, calculating the contribution data of the target attribution dimension; according to the contribution data, determining the cause of the quality inspection pass rate fluctuation under the target attribution dimension.

[0005] In the above implementation process, by calculating the Lorenz curve coefficient corresponding to each attribution dimension, the target attribution dimension that should be disassembled most is quickly determined from multiple attribution dimensions, avoiding processing of meaningless attribution dimensions, saving computing resources, and improving the efficiency of cause analysis. Based on historical quality inspection data and current time data, the contribution data of the target attribution dimension is calculated, and the proportion of the reasons for the change in the quality inspection pass rate is quantified through the contribution data, and then the reasons for the fluctuation of the quality inspection pass rate are determined, which is conducive to targeted improvements and improving the stability of the quality inspection pass rate.

[0006] Optionally, in an embodiment of the present application, the historical quality inspection data includes a historical quality inspection pass rate; before calculating the Lorenz curve coefficient corresponding to the attribution dimension based on the historical quality inspection data and the current quality inspection data under each attribution dimension, the method also includes: calculating the mean of the historical quality inspection pass rate and the standard deviation of the historical quality inspection pass rate based on the historical quality inspection pass rate; determining the upper control limit threshold and the lower control limit threshold of the current quality inspection data based on the mean of the historical quality inspection pass rate and the standard deviation of the historical quality inspection pass rate; the upper control limit threshold is the upper limit value of the fluctuation range of the quality inspection pass rate; the lower control limit threshold is the lower limit value of the fluctuation range of the quality inspection pass rate; based on the historical quality inspection data and the current quality inspection data under each attribution dimension, calculating the Lorenz curve coefficient corresponding to the attribution dimension, including: if the current quality inspection data exceeds the upper control limit threshold, or the current quality inspection data is lower than the lower control limit threshold, then based on the historical quality inspection data and the current quality inspection data, calculating the Lorenz curve coefficient corresponding to each attribution dimension.

[0007] In the above implementation process, by calculating the upper control threshold and lower control threshold of the current quality inspection data, it is possible to monitor in real time whether the pass rate corresponding to the current quality inspection data is within the normal fluctuation range, reducing the cost of manual monitoring. When the current quality inspection data exceeds the upper control threshold or is lower than the lower control threshold, the abnormal attribution analysis can be triggered to quickly determine the cause of the fluctuation in the quality inspection pass rate.

[0008] Optionally, in an embodiment of the present application, historical quality inspection data includes historical enumeration value quality inspection data under the attribution dimension; current quality inspection data includes current enumeration value quality inspection data under the attribution dimension; based on the historical quality inspection data and the current quality inspection data, the Lorenz curve coefficient corresponding to each attribution dimension is calculated, including: determining the quality inspection pass rate contribution and quality inspection quantity proportion corresponding to the enumeration value under the attribution dimension according to the historical enumeration value quality inspection data and the current enumeration value quality inspection data; the enumeration value under the attribution dimension refers to the sub-classification under the classification factor; the quality inspection pass rate contribution represents the influence of the sub-classification on the fluctuation of the quality inspection pass rate; the quality inspection quantity proportion represents the ratio of the quality inspection quantity of the sub-classification to the total quality inspection quantity under the attribution dimension; the quality inspection pass rate contribution and the quality inspection quantity proportion corresponding to the enumeration value are multiplied, and the product results corresponding to each enumeration value under the attribution dimension are added, and the Lorenz curve coefficient corresponding to the attribution dimension is obtained based on the addition result.

[0009] In the above implementation process, the Lorenz curve coefficient corresponding to the attribution dimension is determined by the contribution of the quality inspection pass rate and the proportion of quality inspection volume corresponding to the enumeration value under the attribution dimension. The Lorenz curve coefficient helps to accurately identify the attribution dimension that has the greatest impact on the quality inspection pass rate. By quantifying the contribution of each attribution dimension, the factors that have the most significant impact on the overall result can be quickly located.

[0010] Optionally, in an embodiment of the present application, the contribution data includes the contribution of capability change; the contribution of capability change indicates the impact of quality inspection capability change on quality inspection pass rate change under the target attribution dimension; the historical quality inspection data includes the historical enumeration value structure proportion and the historical enumeration value quality inspection pass rate under the attribution dimension; the historical enumeration value structure proportion indicates the proportion of the quality inspection amount of the enumeration value under the attribution dimension in the total historical quality inspection amount of the attribution dimension; the current quality inspection data includes the current enumeration value quality inspection pass rate under the attribution dimension; the contribution data of the target attribution dimension is calculated based on the historical quality inspection data and the current quality inspection data, including: calculating the difference between the current enumeration value quality inspection pass rate and the historical quality inspection pass rate of the same dimension, multiplying the difference by the historical enumeration value structure proportion, and obtaining the capability change contribution.

[0011] In the above implementation process, the contribution data under the same attribution dimension are disassembled, including the contribution of capability change. The contribution of capability change reflects the impact of changes in quality inspection capabilities on changes in quality inspection pass rate, and the quality inspection process is reviewed and optimized based on the contribution of capability change to reduce human errors and improve efficiency.

[0012] Optionally, in an embodiment of the present application, the contribution data also includes a contribution of structural distribution changes; the contribution of structural distribution changes indicates the impact of changes in product quality on changes in quality inspection pass rate under the target attribution dimension; the historical quality inspection data also includes historical quality inspection pass rates of the same dimension; the current quality inspection data also includes the current enumeration value structure proportion; the current enumeration value structure proportion indicates the proportion of the quality inspection amount of the enumeration value under the attribution dimension in the current total quality inspection amount of the attribution dimension; based on the historical quality inspection data and the current quality inspection data, the contribution data of the target attribution dimension is calculated, including: obtaining a first difference between the current enumeration value structure proportion and the historical enumeration value structure proportion, and a second difference between the current enumeration value quality inspection pass rate and the historical quality inspection pass rate of the same dimension; multiplying the first difference by the second difference to obtain the contribution of the structural distribution change of the attribution dimension.

[0013] In the above implementation process, after the contribution data is broken down, it also includes the contribution of structural distribution changes. The contribution of structural distribution changes can quantify the impact of changes in product quality on changes in quality inspection pass rates, which helps to clarify the proportion of quality factors in the overall change in quality inspection pass rates, thereby more accurately evaluating the room for improvement in product quality. After calculating the contribution of structural distribution changes for each target attribution dimension, it is also possible to determine the products or merchants that need to make quality improvements first.

[0014] Optionally, in an embodiment of the present application, the contribution data includes the contribution of capability change and the contribution of structural distribution change; based on the contribution data, the cause of the fluctuation of the quality inspection pass rate under the target attribution dimension is determined, including: if the contribution of capability change is greater than the contribution of structural distribution change, then it is determined that the cause of the fluctuation of the quality inspection pass rate is the change in quality inspection capability; the quality inspection capability includes quality inspection standards and / or the enforcement of quality inspection standards; if the contribution of capability change is less than the contribution of structural distribution change, then it is determined that the cause of the fluctuation of the quality inspection pass rate is the change in product quality.

[0015] In the above implementation process, the impact of fluctuations in quality inspection pass rate caused by changes in quality inspection capability is quantified through the contribution of capability change, and the impact of fluctuations in quality inspection pass rate caused by changes in product quality is quantified through the contribution of structural distribution change. Therefore, based on the contribution of capability change and the contribution of structural distribution change, the proportion of causes of fluctuations in quality inspection pass rate can be determined, which is conducive to targeted improvements and improving the stability of quality inspection pass rate.

[0016] Optionally, in an embodiment of the present application, based on the Lorenz curve coefficient of the attribution dimension, a target attribution dimension is determined from the attribution dimension, including: determining the attribution dimension whose Lorenz curve coefficient exceeds a preset threshold as the target attribution dimension; or determining the first several attribution dimensions with the largest Lorenz curve coefficient as the target attribution dimension.

[0017] In the above implementation process, without introducing the Lorenz curve coefficient, it may be necessary to conduct a detailed analysis of all attribution dimensions, which is not only time-consuming but also may involve a lot of unnecessary calculations. By determining the target attribution dimension from the attribution dimension through the Lorenz curve coefficient, only the truly important target attribution dimension can be deeply analyzed, thus saving time and computing resources, making the analysis process more efficient.

[0018] In the second aspect, the embodiment of the present application also provides a quality inspection pass rate attribution analysis device, including: a coefficient calculation module, which is used to calculate the Lorenz curve coefficient corresponding to the attribution dimension based on the historical quality inspection data and the current quality inspection data under each attribution dimension; the attribution dimension refers to the classification factor that causes the quality inspection pass rate fluctuation; the Lorenz curve coefficient is used to characterize the degree of fluctuation of the quality inspection pass rate under the attribution dimension; a dimension determination module, which is used to determine the target attribution dimension from the attribution dimension based on the Lorenz curve coefficient of the attribution dimension; a contribution data calculation module, which is used to calculate the contribution data of the target attribution dimension based on the historical quality inspection data and the current quality inspection data; a cause determination module, which is used to determine the cause of the quality inspection pass rate fluctuation under the target attribution dimension based on the contribution data.

[0019] In a third aspect, an embodiment of the present application further provides a computer program product, including computer program instructions, which, when executed by a processor, execute the method provided by the first aspect or any one of the implementations of the first aspect.

[0020] In a fourth aspect, an embodiment of the present application further provides an electronic device, comprising: a processor and a memory, the memory storing computer program instructions, and the computer program instructions, when executed by the processor, execute the method provided by the first aspect or any one of the implementations of the first aspect.

[0021] In a fifth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the method provided by the first aspect or any one of the implementations of the first aspect is executed.

[0022] By adopting a quality inspection pass rate attribution analysis method, program product, device and storage medium provided by this application, by calculating the Lorenz curve coefficient corresponding to each attribution dimension, the target attribution dimension that should be disassembled most is quickly determined from multiple attribution dimensions, avoiding processing of meaningless attribution dimensions, saving computing resources, and improving the efficiency of cause analysis. Based on historical quality inspection data and current time data, the contribution data of the target attribution dimension is calculated, and the proportion of the reasons for the change in the quality inspection pass rate is quantified through the contribution data, and then the reasons for the fluctuation of the quality inspection pass rate are determined, which is conducive to targeted improvements and improving the stability of the quality inspection pass rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0024] Figure 1 A flow chart of a quality inspection pass rate attribution analysis method provided in an embodiment of the present application;

[0025] Figure 2 A schematic diagram of the structure of a quality inspection pass rate attribution analysis device provided in an embodiment of the present application;

[0026] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0027] The following embodiments of the technical solution of the present application are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present application, and are therefore only used as examples, and cannot be used to limit the scope of protection of the present application.

[0028] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application.

[0029] In the description of the embodiments of the present application, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0030] See also Figure 1 A flow chart of a quality inspection pass rate attribution analysis method provided by an embodiment of the present application is shown. The quality inspection pass rate attribution analysis method provided by an embodiment of the present application can be applied to an electronic device, which may include a physical device such as a server, a PC, a tablet computer, or a smart phone, or a virtual device such as a virtual machine or a container. The electronic device may be a single device, or a combination of multiple devices or a cluster of a large number of devices. The quality inspection pass rate attribution analysis method may include:

[0031] Step S110: Based on the historical quality inspection data and current quality inspection data under each attribution dimension, calculate the Lorenz curve coefficient corresponding to the attribution dimension; the attribution dimension refers to the classification factors that cause the fluctuation of the quality inspection pass rate; the Lorenz curve coefficient is used to characterize the degree of fluctuation of the quality inspection pass rate under the attribution dimension.

[0032] Step S120: Determine a target attribution dimension from the attribution dimensions based on the Lorenz curve coefficient of the attribution dimension.

[0033] Step S130: Calculate the contribution data of the target attribution dimension based on the historical quality inspection data and the current quality inspection data.

[0034] Step S140: Determine the reasons for the fluctuation of the quality inspection pass rate under the target attribution dimension based on the contribution data.

[0035] In step S110, the attribution dimension refers to the classification factors that cause the fluctuation of the quality inspection pass rate, such as product category, merchant type, order type, merchant qualification, execution project and / or defect label, etc. Each classification factor may include one or more enumeration values, and the enumeration value under the attribution dimension refers to the sub-classification under the classification factor, for example, the sub-classification of the product category may include electronic products, clothing, shoes, etc.; the sub-classification of the merchant type may include individual merchants, corporate merchants or other merchants, etc. The sub-types of the order type include brand direct delivery, spot goods or consignment, etc.

[0036] Historical quality inspection data refers to the quality inspection-related data generated in the quality inspection process in history. Current quality inspection data refers to the quality inspection-related data generated in the quality inspection process in the latest period. The time range of historical quality inspection data can be set according to demand. For example, if the current quality inspection data is the quality inspection data of September of the current year, then the quality inspection data can be the quality inspection data of August of the current year, or the quality inspection data of January to August of the current year, or the quality inspection data of a certain period in history that is preset in advance. Historical quality inspection data includes historical quality inspection pass quantity, historical quality inspection quantity, historical quality inspection pass rate, historical enumeration value quality inspection data, etc. Among them, enumeration value quality inspection data refers to the quality inspection data of subcategories under classification factors, such as the quality inspection pass rate of electronic products under the commodity category. Historical quality inspection data can be directly obtained by calling data interfaces and other methods.

[0037] The corresponding Lorenz curve coefficient is calculated for each attribution dimension. The Lorenz curve coefficient is used to characterize the degree of fluctuation of the quality inspection pass rate under the attribution dimension. The larger the Lorenz curve coefficient is, the greater the degree of fluctuation of the quality inspection pass rate under the attribution dimension is, and the more attention and disassembly are needed. On the contrary, the smaller the Lorenz curve coefficient is, the less fluctuation of the quality inspection pass rate under the attribution dimension is, and the attribution dimension can be left unprocessed. The calculation method of the Lorenz curve coefficient will be introduced later.

[0038] In step S120, based on the Lorenz curve coefficient of each attribution dimension, the degree of fluctuation of the quality inspection pass rate of each attribution dimension can be determined, and the first several attribution dimensions with the largest Lorenz curve coefficient can be determined as the target attribution dimension, or the attribution dimension whose Lorenz curve coefficient exceeds the preset threshold can be determined as the target attribution dimension. For the attribution dimension whose Lorenz curve coefficient does not exceed the preset threshold, it means that the degree of fluctuation of the quality inspection pass rate of the attribution dimension is small, and no subsequent analysis is required for the time being.

[0039] In step S130, the contribution data refers to the influence of different reasons on the change of quality inspection pass rate under the target attribution dimension. The contribution data may have one or more items, each corresponding to one reason. The larger the contribution data value, the greater the influence of the reason on the change of quality inspection pass rate. Exemplarily, the contribution data may include the contribution of capability change and the contribution of structural distribution change. The contribution of capability change represents the influence of the change of quality inspection capability on the change of quality inspection pass rate under the target attribution dimension. The contribution of structural distribution change represents the influence of the change of product quality on the change of quality inspection pass rate under the target attribution dimension. If the contribution data includes only one item of data, the contribution data can be compared with a preset threshold. If the contribution data is greater than the preset threshold, it is considered that the reason corresponding to the contribution data is the reason for the fluctuation of the pass rate.

[0040] Taking the contribution data of capability change contribution as an example, the contribution data of the target attribution dimension is calculated. For example, the historical enumeration value structure ratio in the historical quality inspection data and the historical enumeration value quality inspection pass rate under the attribution dimension are obtained, and the current enumeration value quality inspection pass rate in the current quality inspection data is obtained. The historical enumeration value structure ratio is used as the weight data, and the difference between the current enumeration value quality inspection pass rate and the historical quality inspection pass rate of the same dimension is weighted to obtain the capability change contribution.

[0041] Taking the contribution data of capability change as an example, the contribution data of the target attribution dimension is calculated. For example, the first difference between the current enumeration value structure proportion and the historical enumeration value structure proportion, and the second difference between the current enumeration value quality inspection pass rate and the historical quality inspection pass rate of the same dimension are obtained; the first difference is multiplied by the second difference to obtain the contribution of the structural distribution change of the attribution dimension.

[0042] In step S140, the cause corresponding to the contribution data with the largest value can be determined as the cause of the fluctuation of the quality inspection pass rate. Continuing with the above embodiment, the contribution data includes the contribution of capability change and the contribution of structural distribution change, where the contribution of capability change is greater than the contribution of structural distribution change, then it can be considered that the quality inspection capability, which is the main cause of the fluctuation of the quality inspection pass rate, has changed. Of course, this does not mean that the quality of the goods will not cause the fluctuation of the quality inspection pass rate, but relatively speaking, the fluctuation of the quality inspection pass rate caused by the quality of the goods is small, and we can first focus on the changes caused by the quality inspection capability.

[0043] In the above implementation process, by calculating the Lorenz curve coefficient corresponding to each attribution dimension, the target attribution dimension that should be disassembled most is quickly determined from multiple attribution dimensions, avoiding processing of meaningless attribution dimensions, saving computing resources, and improving the efficiency of cause analysis. Based on historical quality inspection data and current time data, the contribution data of the target attribution dimension is calculated, and the proportion of the reasons for the change in the quality inspection pass rate is quantified through the contribution data, and then the reasons for the fluctuation of the quality inspection pass rate are determined, which is conducive to targeted improvements and improving the stability of the quality inspection pass rate.

[0044] Optionally, in an embodiment of the present application, the historical quality inspection data includes a historical quality inspection pass rate; wherein the historical period can be freely set, and the historical quality inspection pass rate can be calculated by dividing the historical quality inspection pass quantity by the historical quality inspection quantity.

[0045] Based on the historical quality inspection data and the current quality inspection data under each attribution dimension, before calculating the Lorenz curve coefficient corresponding to the attribution dimension, the method further includes:

[0046] The mean of the historical quality inspection pass rate and the standard deviation of the historical quality inspection pass rate are calculated based on the historical quality inspection pass rate. For example, one cycle is one month, and the calculation method of the mean of the historical quality inspection pass rate can be: collect the historical quality inspection pass rate of each cycle (January to August) respectively, and calculate the average of the historical quality inspection pass rate of each cycle from January to August. The standard deviation of the historical quality inspection pass rate can be calculated using the standard deviation formula, substituting the historical quality inspection pass rate of each cycle and the mean of the historical quality inspection pass rate.

[0047] Based on the historical quality inspection pass rate mean and historical quality inspection pass rate standard deviation, determine the upper control limit threshold and lower control limit threshold of the current quality inspection data; the upper control limit threshold is the upper limit value of the quality inspection pass rate fluctuation range; the lower control limit threshold is the lower limit value of the quality inspection pass rate fluctuation range.

[0048] For example, the upper control limit threshold can be recorded as UCL, and the lower control limit threshold can be recorded as LCL, and the calculation logic is UCL = μ + nσ, LCL = μ - nσ; where μ is the mean of the historical quality inspection pass rate, σ is the standard deviation of the historical quality inspection pass rate, and n is a multiple, which can be dynamically adjusted according to the business requirements for stability in different scenarios. For example, n can be set to 2, 3 or 4, etc., and can generally be set to 3. The basis for adjustment is that if the range from the upper control limit threshold to the lower control limit threshold needs to be narrowed, the multiple can be reduced; if the range from the upper control limit threshold to the lower control limit threshold needs to be expanded, the multiple can be enlarged.

[0049] After determining the upper control limit threshold and the lower control limit threshold, the Lorenz curve coefficient corresponding to the attribution dimension is calculated based on the historical quality inspection data and current quality inspection data under each attribution dimension, including:

[0050] If the current quality inspection data exceeds the upper control limit threshold, or the current quality inspection data is lower than the lower control limit threshold, the Lorenz curve coefficient corresponding to each attribution dimension is calculated based on the historical quality inspection data and the current quality inspection data.

[0051] The upper control limit threshold and the lower control limit threshold can be regarded as conditional thresholds for automatically triggering attribution analysis, wherein, if the current quality inspection data exceeds the upper control limit threshold, it means that the current quality inspection data exceeds the upper limit value of the quality inspection pass rate fluctuation range, and attribution analysis is required. Then, the quality inspection pass rate analysis method is started, and the step of calculating the Lorenz curve coefficient corresponding to each attribution dimension based on the historical quality inspection data and the current quality inspection data in step S110 is executed.

[0052] If the current quality inspection data is lower than the control lower limit threshold, it means that the current quality inspection data is at the lower limit of the quality inspection pass rate fluctuation range, and attribution analysis is also required. Then, the quality inspection pass rate analysis method is started, and the step S110 is executed to calculate the Lorenz curve coefficient corresponding to each attribution dimension based on the historical quality inspection data and the current quality inspection data.

[0053] In the implementation process of the above embodiment: by calculating the upper control threshold and the lower control threshold of the current quality inspection data, it is possible to monitor in real time whether the pass rate corresponding to the current quality inspection data is within the normal fluctuation range, thereby reducing the cost of manual monitoring. When the current quality inspection data exceeds the upper control threshold or is lower than the lower control threshold, an abnormal attribution analysis can be triggered to quickly determine the cause of the fluctuation in the quality inspection pass rate.

[0054] Optionally, in an embodiment of the present application, the historical quality inspection data includes the historical enumeration value quality inspection data under the attribution dimension; the time range of the historical quality inspection data can be set according to the demand, and the historical quality inspection data here can be regarded as the base period quality inspection data, and the base period can be the previous cycle of the current time, for example, if the current time is February, the base period is January, and the base period can also be a preset fixed period. Taking the attribution dimension as the merchant type as an example, the sub-classification under the merchant type includes personal merchants, corporate merchants and other merchants, wherein the historical enumeration value quality inspection data can include the historical enumeration value quality inspection amount, the historical enumeration value quality inspection amount ratio, the historical enumeration value quality inspection pass amount and the historical enumeration value quality inspection pass rate and other data. Then the historical enumeration value quality inspection data includes the historical quality inspection data of personal merchants, the historical quality inspection data of corporate merchants and other historical merchant quality inspection data, wherein the historical quality inspection data of personal merchants includes the historical quality inspection amount of personal merchants, the historical quality inspection amount ratio of personal merchants and the historical quality inspection pass rate of personal merchants, etc., and the same is true for corporate merchants and other merchants.

[0055] The current quality inspection data includes the current enumeration value quality inspection data under the attribution dimension. The current enumeration value quality inspection data includes the current enumeration value quality inspection quantity, the current enumeration value quality inspection quantity ratio, the current enumeration value quality inspection pass quantity, and the current enumeration value quality inspection pass rate.

[0056] Based on the historical quality inspection data and the current quality inspection data, the Lorenz curve coefficient corresponding to each attribution dimension is calculated, including: determining the quality inspection pass rate contribution and quality inspection quantity proportion corresponding to the enumeration value under the attribution dimension according to the historical enumeration value quality inspection data and the current enumeration value quality inspection data; the enumeration value under the attribution dimension refers to the sub-classification under the classification factor; the quality inspection pass rate contribution represents the impact of the sub-classification on the fluctuation of the quality inspection pass rate; the quality inspection quantity proportion represents the ratio of the quality inspection quantity of the sub-classification to the total quality inspection quantity under the attribution dimension.

[0057] Continuing with the above embodiment, for each enumeration value (individual merchants, corporate merchants and other merchants), the corresponding quality inspection pass rate contribution and quality inspection quantity ratio are calculated. The calculation method of the quality inspection pass rate contribution is: quality inspection pass rate contribution = {[historical quality inspection pass quantity total + (current enumeration value quality inspection pass quantity - historical enumeration value quality inspection pass quantity)] / [historical quality inspection quantity total + (historical enumeration value quality inspection quantity - current enumeration value quality inspection quantity)]} - historical total quality inspection pass rate. Among them, the historical quality inspection quantity total is the sum of the quality inspection quantities of each historical enumeration value. The quality inspection quantity ratio is the historical enumeration value quality inspection quantity ratio, which represents the ratio of the quality inspection quantity of the sub-category to the total quality inspection quantity under the attribution dimension, and can be obtained by dividing the historical enumeration value quality inspection quantity by the historical quality inspection quantity total.

[0058] After obtaining the quality inspection pass rate contribution, the Lorenz curve coefficient corresponding to the attribution dimension is obtained according to the quality inspection pass rate contribution and quality inspection volume proportion corresponding to the enumeration value under the attribution dimension. For example, the Lorenz curve coefficient = |1-[sum(quality inspection pass rate contribution*quality inspection volume proportion)]|.

[0059] In the implementation process of the above embodiment: the Lorenz curve coefficient corresponding to the attribution dimension is determined by the contribution of the quality inspection pass rate and the proportion of the quality inspection volume corresponding to the enumeration value under the attribution dimension. The Lorenz curve coefficient helps to accurately identify the attribution dimension that has the greatest impact on the quality inspection pass rate. By quantifying the contribution of each attribution dimension, the factors that have the most significant impact on the overall result can be quickly located.

[0060] Optionally, in an embodiment of the present application, the contribution data includes a contribution of capability change; the contribution of capability change represents the impact of changes in quality inspection capability on changes in quality inspection pass rate under the target attribution dimension.

[0061] The historical quality inspection data includes the historical enumeration value structure ratio and the historical enumeration value quality inspection pass rate under the attribution dimension; the historical enumeration value structure ratio indicates the ratio of the quality inspection quantity of the enumeration value under the attribution dimension to the total historical quality inspection quantity of the attribution dimension; the current quality inspection data includes the current enumeration value quality inspection pass rate under the attribution dimension. Among them, the historical enumeration value structure ratio can be obtained by dividing the historical (base period) enumeration value quality inspection quantity by the total historical (base period) quality inspection quantity. The historical enumeration value quality inspection pass rate can be obtained by dividing the historical (base period) enumeration value quality inspection pass quantity by the base period enumeration value quality inspection quantity. The current enumeration value quality inspection pass rate can be obtained by dividing the current period enumeration value quality inspection pass quantity by the current period enumeration value quality inspection quantity. Of course, the above data can also be directly obtained by calling the quality inspection data interface and other methods to obtain the statistical data.

[0062] Based on historical quality inspection data and current quality inspection data, the contribution data of the target attribution dimension is calculated, including:

[0063] Calculate the difference between the current enumeration value quality inspection pass rate and the historical quality inspection pass rate of the same dimension, multiply the difference by the historical enumeration value structure ratio, and obtain the contribution to capability change.

[0064] In the implementation process of the above embodiment: the contribution data under the same attribution dimension is disassembled, including the contribution of capability change, and the contribution of capability change is used to reflect the impact of the change in quality inspection capability on the change in quality inspection pass rate, so as to review and optimize the quality inspection process according to the contribution of capability change, so as to reduce human errors and improve efficiency.

[0065] Optionally, in the embodiment of the present application, the contribution data also includes a contribution of a structural distribution change; the contribution of a structural distribution change indicates the impact of a change in product quality on a change in the quality inspection pass rate under the target attribution dimension;

[0066] The historical quality inspection data also includes the historical quality inspection pass rate of the same dimension; the current quality inspection data also includes the current enumeration value structure ratio; the current enumeration value structure ratio indicates the ratio of the quality inspection quantity of the enumeration value under the attribution dimension to the current quality inspection total of the attribution dimension. The current enumeration value structure ratio can be obtained by dividing the current (current period) enumeration value quality inspection quantity by the current quality inspection total.

[0067] According to the historical quality inspection data and the current quality inspection data, the contribution data of the target attribution dimension is calculated, including: obtaining the first difference between the current enumeration value structure proportion and the historical enumeration value structure proportion, and the second difference between the current enumeration value quality inspection pass rate and the historical quality inspection pass rate of the same dimension; multiplying the first difference by the second difference to obtain the contribution of the structural distribution change of the attribution dimension.

[0068] In the implementation process of the above embodiment: after the contribution data is disassembled, it also includes the contribution of structural distribution change, which quantifies the impact of product quality changes on the change of quality inspection pass rate through the contribution of structural distribution change, which helps to clarify the proportion of quality factors in the overall change of quality inspection pass rate, so as to more accurately evaluate the improvement space of product quality. And after calculating the contribution of structural distribution change for each target attribution dimension, it is also possible to determine the products or merchants that need to improve their quality first.

[0069] Optionally, in the embodiment of the present application, the contribution data includes the contribution of capability change and the contribution of structure distribution change; according to the contribution data, the reasons for the fluctuation of the quality inspection pass rate under the target attribution dimension are determined, including:

[0070] If the contribution of capability change is greater than the contribution of structural distribution change, this indicates that the fluctuation of the quality inspection pass rate is mainly affected by the quality inspection capability, and the cause of the fluctuation of the quality inspection pass rate is determined to be the change of quality inspection capability; quality inspection capability includes quality inspection standards and / or the enforcement of quality inspection standards. For example, whether the quality inspection standard itself has changed, or whether the quality inspection standard has been appropriately updated to adapt to the new product quality requirements. Or whether the quality inspectors consistently perform quality inspections according to established standards. This may involve the training, supervision and strictness of the execution process of personnel. It is understandable that this situation does not mean that the fluctuation of the quality inspection pass rate is completely unaffected by the quality of the goods, but that the main cause of the abnormality is the change in quality inspection capability relative to the quality of the goods. This conclusion is used to give targeted opinions, not to give a definite and unique answer. For example, you can first check the problems related to quality inspection capability, and then verify the problems related to product quality based on the contribution of structural distribution changes.

[0071] If the contribution of capacity change is less than the contribution of structural distribution change, it is determined that the cause of the fluctuation in the quality inspection pass rate is the change in product quality. If the contribution of capacity change is less than the contribution of structural distribution change, this indicates that the change in product quality is the main factor affecting the quality inspection pass rate. This may include: supplier changes, new or changed suppliers may affect the quality of raw materials or products. Product design issues, design defects may cause product quality problems, thereby affecting the quality inspection pass rate. Production process changes, any changes in the production process, such as process adjustments or operational errors, may affect the quality of the final product.

[0072] For example, the dimension Lorenz curve coefficient of the "first-level category" in the attribution dimension is the largest, and the enumeration values ​​(subcategories) of the first-level category include footwear categories and clothing categories, etc. The contribution of the ability change of the enumeration value "footwear category" is -3%, and the contribution of the structural distribution change is -20%, indicating that the reason for the decrease in the quality inspection pass rate is mainly due to the change in the quality of the goods in the footwear category. It is necessary to inventory the source of goods of the merchants and clear out the low-quality sweeping merchants. In another example, the contribution of the ability change of the "clothing category" is +30%, and the contribution of the structural distribution change is +2%, indicating that the change in the quality inspection pass rate (mainly an increase here) is mainly due to the change in the quality inspection capacity in the warehouse. It may be that the quality inspection standards and / or the enforcement of the quality inspection standards have changed, and the goods that have completed the quality inspection need to be checked to check whether the quality inspectors have missed defects in the inspected goods.

[0073] In the implementation process of the above embodiment: the impact of fluctuations in quality inspection pass rate caused by changes in quality inspection capability is quantified by the contribution of capability change, and the impact of fluctuations in quality inspection pass rate caused by changes in product quality is quantified by the contribution of structural distribution change. Therefore, based on the contribution of capability change and the contribution of structural distribution change, the proportion of causes of fluctuations in quality inspection pass rate can be determined, which is conducive to targeted improvements and improving the stability of quality inspection pass rate.

[0074] Optionally, in the embodiment of the present application, based on the Lorenz curve coefficient of the attribution dimension, determining the target attribution dimension from the attribution dimension includes:

[0075] The attribution dimension whose Lorenz curve coefficient exceeds the preset threshold is determined as the target attribution dimension; or the first several attribution dimensions with the largest Lorenz curve coefficient are determined as the target attribution dimensions. In one embodiment, the preset threshold may be 0.4, 0.5 or 0.6, etc. The Lorenz curve coefficient exceeding the preset threshold indicates that the attribution dimension has a large fluctuation and can be determined as the target attribution dimension. The preset threshold can be set according to demand, and this application does not limit this.

[0076] In the implementation process of the above embodiment: without introducing the Lorenz curve coefficient, it may be necessary to perform a detailed analysis of all attribution dimensions, which is not only time-consuming but also may involve a large number of unnecessary calculations. By determining the target attribution dimension from the attribution dimension through the Lorenz curve coefficient, only the truly important target attribution dimension can be deeply analyzed, thereby saving time and computing resources, making the analysis process more efficient.

[0077] In an optional embodiment, after determining the reasons for the fluctuation of the quality inspection pass rate under the target attribution dimension, an attribution report can be generated based on the reasons for the fluctuation of the quality inspection pass rate and the analysis process, and corresponding suggestions can be given in combination with the reasons for the fluctuation of the quality inspection pass rate. For example, if the reason for the fluctuation of the quality inspection pass rate is a change in quality inspection capabilities, it can be suggested to review the quality inspection process, strengthen training, and improve supervision. If the reason for the fluctuation of the quality inspection pass rate is a change in product quality, it can be suggested to evaluate the supply chain, improve product design, and optimize the product production process.

[0078] See also Figure 2 The structure diagram of the quality inspection pass rate attribution analysis device provided in the embodiment of the present application is shown; the embodiment of the present application provides a quality inspection pass rate attribution analysis device 200, including:

[0079] The coefficient calculation module 210 is used to calculate the Lorenz curve coefficient corresponding to the attribution dimension based on the historical quality inspection data and the current quality inspection data under each attribution dimension; the attribution dimension refers to the classification factors that cause the fluctuation of the quality inspection pass rate; the Lorenz curve coefficient is used to characterize the degree of fluctuation of the quality inspection pass rate under the attribution dimension;

[0080] A dimension determination module 220, configured to determine a target attribution dimension from the attribution dimensions based on the Lorenz curve coefficient of the attribution dimension;

[0081] The contribution data calculation module 230 is used to calculate the contribution data of the target attribution dimension based on the historical quality inspection data and the current quality inspection data;

[0082] The cause determination module 240 is used to determine the cause of the fluctuation of the quality inspection pass rate under the target attribution dimension based on the contribution data.

[0083] Optionally, in an embodiment of the present application, the quality inspection pass rate attribution analysis device 200, the historical quality inspection data includes the historical quality inspection pass rate; it also includes an automatic trigger module, which is used to calculate the historical quality inspection pass rate mean and the historical quality inspection pass rate standard deviation based on the historical quality inspection pass rate; according to the historical quality inspection pass rate mean and the historical quality inspection pass rate standard deviation, determine the control upper limit threshold and control lower limit threshold of the current quality inspection data; the control upper limit threshold is the upper limit value of the quality inspection pass rate fluctuation range; the control lower limit threshold is the lower limit value of the quality inspection pass rate fluctuation range; if the current quality inspection data exceeds the control upper limit threshold, or the current quality inspection data is lower than the control lower limit threshold, then based on the historical quality inspection data and the current quality inspection data, calculate the Lorenz curve coefficient corresponding to each attribution dimension.

[0084] Optionally, in an embodiment of the present application, in the quality inspection pass rate attribution analysis device 200, the historical quality inspection data includes historical enumeration value quality inspection data under the attribution dimension; the current quality inspection data includes current enumeration value quality inspection data under the attribution dimension; the coefficient calculation module 210 is specifically used to determine the quality inspection pass rate contribution and quality inspection quantity proportion corresponding to the enumeration value under the attribution dimension according to the historical enumeration value quality inspection data and the current enumeration value quality inspection data; the enumeration value under the attribution dimension refers to the sub-classification under the classification factor; the quality inspection pass rate contribution represents the influence of the sub-classification on the fluctuation of the quality inspection pass rate; the quality inspection quantity proportion represents the ratio of the quality inspection quantity of the sub-classification to the total quality inspection quantity under the attribution dimension; the quality inspection pass rate contribution and the quality inspection quantity proportion corresponding to the enumeration value are multiplied, and the product results corresponding to each enumeration value under the attribution dimension are added, and the Lorenz curve coefficient corresponding to the attribution dimension is obtained based on the addition result.

[0085] Optionally, in an embodiment of the present application, in the quality inspection pass rate attribution analysis device 200, contribution data includes contribution of capability change; the capability change contribution indicates the impact of quality inspection capability change on quality inspection pass rate change under the target attribution dimension; historical quality inspection data includes historical enumeration value structure proportion and historical enumeration value quality inspection pass rate under the attribution dimension; the historical enumeration value structure proportion indicates the proportion of the quality inspection amount of the enumeration value under the attribution dimension to the total historical quality inspection amount of the attribution dimension; current quality inspection data includes the current enumeration value quality inspection pass rate under the attribution dimension; the contribution data calculation module 230 is specifically used to calculate the difference between the current enumeration value quality inspection pass rate and the historical quality inspection pass rate of the same dimension, and multiply the difference by the historical enumeration value structure proportion to obtain the capability change contribution.

[0086] Optionally, in an embodiment of the present application, in the quality inspection pass rate attribution analysis device 200, the contribution data also includes the contribution of structural distribution changes; the contribution of structural distribution changes indicates the impact of changes in product quality on changes in quality inspection pass rate under the target attribution dimension; the historical quality inspection data also includes the historical quality inspection pass rate of the same dimension; the current quality inspection data also includes the current enumeration value structure proportion; the current enumeration value structure proportion indicates the proportion of the quality inspection amount of the enumeration value under the attribution dimension in the current total quality inspection amount of the attribution dimension; the contribution data calculation module 230 specifically obtains the first difference between the current enumeration value structure proportion and the historical enumeration value structure proportion, and the second difference between the current enumeration value quality inspection pass rate and the historical quality inspection pass rate of the same dimension; the first difference is multiplied by the second difference to obtain the contribution of the structural distribution change of the attribution dimension.

[0087] Optionally, in an embodiment of the present application, the quality inspection pass rate attribution analysis device 200, the contribution data includes the contribution of capability change and the contribution of structural distribution change; the cause determination module 240 is specifically used to determine that if the contribution of capability change is greater than the contribution of structural distribution change, then the cause of the fluctuation in the quality inspection pass rate is determined to be a change in quality inspection capability; the quality inspection capability includes quality inspection standards and / or the intensity of enforcement of quality inspection standards; if the contribution of capability change is less than the contribution of structural distribution change, then the cause of the fluctuation in the quality inspection pass rate is determined to be a change in product quality.

[0088] Optionally, in an embodiment of the present application, the quality inspection pass rate attribution analysis device 200, the dimension determination module 220, is specifically used to determine the attribution dimension whose Lorenz curve coefficient exceeds a preset threshold as the target attribution dimension; or to determine the first several attribution dimensions with the largest Lorenz curve coefficient as the target attribution dimension.

[0089] It should be understood that the device corresponds to the above-mentioned quality inspection pass rate attribution analysis method embodiment, and can execute the various steps involved in the above-mentioned method embodiment. The specific functions of the device can be found in the description above. To avoid repetition, the detailed description is appropriately omitted here. The device includes at least one software function module that can be stored in a memory in the form of software or firmware or solidified in the operating system (OS) of the device.

[0090] See also Figure 3 The electronic device 300 provided in the embodiment of the present application includes: a processor 310 and a memory 320, wherein the memory 320 stores machine-readable instructions executable by the processor 310, and when the machine-readable instructions are executed by the processor 310, the above method is executed.

[0091] Figure 3 Each component shown in can be implemented by hardware, software or a combination thereof. The electronic device 300 may be a physical device, such as a server, a PC, etc., or a virtual device, such as a virtual machine, a virtualized container, etc. Moreover, the electronic device 300 is not limited to a single device, but may also be a combination of multiple devices or a cluster consisting of a large number of devices.

[0092] An embodiment of the present application further provides a storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above method is executed.

[0093] Among them, the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, disk or optical disk.

[0094] In several embodiments provided by the embodiments of the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to the multiple embodiments of the embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and a part of the module, program segment or code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0095] In addition, the functional modules in each embodiment of the present application can be integrated together to form an independent part, or each module can exist separately, or two or more modules can be integrated to form an independent part.

[0096] The above description is only an optional implementation manner of the embodiments of the present application, but the protection scope of the embodiments of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or replacements within the technical scope disclosed in the embodiments of the present application, which should be covered within the protection scope of the embodiments of the present application.

Claims

1. A quality inspection pass rate attribution analysis method, characterized in that: include: Based on the historical quality inspection data and current quality inspection data under each attribution dimension, the Lorenz curve coefficient corresponding to the attribution dimension is calculated; the attribution dimension refers to the classification factors that cause the fluctuation of the quality inspection pass rate; the Lorenz curve coefficient is used to characterize the degree of fluctuation of the quality inspection pass rate under the attribution dimension; determining a target attribution dimension from the attribution dimensions based on the Lorenz curve coefficients of the attribution dimensions; Calculate the contribution data of the target attribution dimension according to the historical quality inspection data and the current quality inspection data; Based on the contribution data, determine the reasons for the fluctuation of the quality inspection pass rate under the target attribution dimension.

2. The method according to claim 1, characterized in that The historical quality inspection data includes a historical quality inspection pass rate; before calculating the Lorenz curve coefficient corresponding to the attribution dimension based on the historical quality inspection data and the current quality inspection data under each attribution dimension, the method further includes: Calculate the mean of the historical quality inspection pass rate and the standard deviation of the historical quality inspection pass rate according to the historical quality inspection pass rate; According to the mean value of the historical quality inspection pass rate and the standard deviation of the historical quality inspection pass rate, determine the upper control limit threshold and the lower control limit threshold of the current quality inspection data; the upper control limit threshold is the upper limit value of the fluctuation range of the quality inspection pass rate; the lower control limit threshold is the lower limit value of the fluctuation range of the quality inspection pass rate; The calculating of the Lorenz curve coefficient corresponding to each attribution dimension based on the historical quality inspection data and the current quality inspection data under each attribution dimension includes: If the current quality inspection data exceeds the upper control limit threshold, or the current quality inspection data is lower than the lower control limit threshold, the Lorenz curve coefficient corresponding to each attribution dimension is calculated based on the historical quality inspection data and the current quality inspection data.

3. The method according to claim 1, characterized in that The historical quality inspection data includes the historical enumeration value quality inspection data under the attribution dimension; the current quality inspection data includes the current enumeration value quality inspection data under the attribution dimension; based on the historical quality inspection data and the current quality inspection data, the Lorenz curve coefficient corresponding to each attribution dimension is calculated, including: According to the historical enumeration value quality inspection data and the current enumeration value quality inspection data, determine the quality inspection pass rate contribution and quality inspection quantity proportion corresponding to the enumeration value under the attribution dimension; the enumeration value under the attribution dimension refers to the sub-classification under the classification factor; the quality inspection pass rate contribution represents the influence of the sub-classification on the quality inspection pass rate fluctuation; the quality inspection quantity proportion represents the ratio of the quality inspection quantity of the sub-classification to the total quality inspection quantity under the attribution dimension; The contribution degree of the quality inspection pass rate and the proportion of the quality inspection quantity corresponding to the enumeration value are multiplied, and the product results corresponding to each enumeration value under the attribution dimension are added, and the Lorenz curve coefficient corresponding to the attribution dimension is obtained based on the addition result.

4. The method according to claim 1, characterized in that: The contribution data includes capability change contribution; the capability change contribution represents the impact of quality inspection capability change on quality inspection pass rate change under the target attribution dimension; The historical quality inspection data includes the historical enumeration value structure ratio and the historical enumeration value quality inspection pass rate under the attribution dimension; the historical enumeration value structure ratio represents the ratio of the quality inspection amount of the enumeration value under the attribution dimension to the total historical quality inspection amount of the attribution dimension; the current quality inspection data includes the current enumeration value quality inspection pass rate under the attribution dimension; The step of calculating the contribution data of the target attribution dimension according to the historical quality inspection data and the current quality inspection data includes: The difference between the current enumeration value quality inspection pass rate and the historical quality inspection pass rate of the same dimension is calculated, and the difference is multiplied by the historical enumeration value structure ratio to obtain the capability change contribution.

5. The method according to claim 4, characterized in that The contribution data also includes a contribution of structural distribution changes; the contribution of structural distribution changes represents the impact of changes in product quality on changes in quality inspection pass rate under the target attribution dimension; The historical quality inspection data also includes the historical quality inspection pass rate of the same dimension; the current quality inspection data also includes the current enumeration value structure ratio; the current enumeration value structure ratio represents the ratio of the quality inspection amount of the enumeration value under the attribution dimension to the current total quality inspection amount of the attribution dimension; Calculating the contribution data of the target attribution dimension according to the historical quality inspection data and the current quality inspection data includes: Obtain a first difference between the current enumeration value structure proportion and the historical enumeration value structure proportion, and a second difference between the current enumeration value quality inspection pass rate and the historical same dimension quality inspection pass rate; multiply the first difference by the second difference to obtain the structural distribution change contribution of the attribution dimension.

6. The method according to claim 1, characterized in that The contribution data includes the contribution of capability change and the contribution of structure distribution change; the reasons for the fluctuation of the quality inspection pass rate under the target attribution dimension are determined according to the contribution data, including: If the contribution of capability change is greater than the contribution of structure distribution change, it is determined that the cause of the fluctuation of the quality inspection pass rate is the change of quality inspection capability; the quality inspection capability includes quality inspection standards and / or the strength of implementation of quality inspection standards; If the contribution of capability change is less than the contribution of structural distribution change, it is determined that the cause of the fluctuation in quality inspection pass rate is the change in product quality.

7. The method according to any one of claims 1 to 6, characterized in that: The determining a target attribution dimension from the attribution dimensions based on the Lorenz curve coefficient of the attribution dimension includes: The attribution dimension whose Lorenz curve coefficient exceeds a preset threshold is determined as the target attribution dimension; or the first several attribution dimensions with the largest Lorenz curve coefficient are determined as the target attribution dimension.

8. A computer program product, characterized in that The method comprises computer program instructions, and when the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is executed.

9. An electronic device, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the method according to any one of claims 1 to 7 is executed.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the method according to any one of claims 1 to 7 is executed.

Images