Method and system for optimizing efficiency of sequential sampling inspection based on prior information
By obtaining prior information about the products to be tested and using a sequential sampling inspection model to evaluate their pass rate and compliance value, the sampling inspection process is optimized, solving the problems of sample selection bias and low efficiency in sampling inspection and achieving more efficient and accurate inspection results.
Patent Information
- Application Number
- CN202410648379.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-23
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-05-23
AI Technical Summary
In existing technologies, sampling inspections suffer from inaccurate results due to sample selection bias, and inappropriate sample size selection leads to low efficiency and an inability to effectively represent the overall situation.
By obtaining prior information about the products to be tested, assessing their prior compliance, and using a sequential sampling inspection model to evaluate the compliance value and optimize efficiency, we can match and optimize contingency plans to ensure that the sampled products meet the requirements and improve inspection efficiency.
This improves the accuracy and efficiency of sampling inspection, reduces resource waste, ensures that products meet inspection standards, and enhances production compliance and safety.
Smart Images

Figure CN118469394B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of sampling inspection technology, and in particular to a method and system for optimizing the efficiency of sequential sampling inspection based on prior information. Background Art
[0002] In current sampling inspection technology, the selection of sampling samples is prone to bias and is not enough to achieve complete randomness, that is, the sample cannot fully represent the population, so the accuracy of the test results will be affected; at the same time, the choice of sample size is too one-sided. If the sample selection is not appropriate, the efficiency of the sampling inspection analysis will be unreliable.
[0003] For example, the invention patent with announcement number CN104063772B discloses a method and device for verifying the reliability of logistics information; a method for verifying the reliability of logistics information includes: receiving logistics information actively pushed by a logistics information server; sampling the received logistics information; generating query conditions using the sampled logistics information samples, and sending query requests to the logistics information server; comparing the query results fed back by the logistics information server with the logistics information samples, and judging whether the logistics information pushed by the logistics information server is reliable based on the consistency of the comparison results; on the one hand, the above technical solution avoids the logistics information server from frequently passively receiving queries by actively pushing logistics information, and effectively balances the query load of the server corresponding to the logistics company; on the other hand, on the information query side, by sampling and reverse querying the pushed information, the reliability of the actively pushed logistics information can be verified.
[0004] For example, the invention patent with announcement number CN117670259B, the above application relates to the field of data processing technology, and specifically discloses a sample detection information management method, which aggregates sample information to form a data set and manages the data set information, wherein data is extracted from an information management database, and the positioning of the information in the data set and the information of the seated persons are completed through a positioning algorithm, so that the information before decision-making can be managed, and it is convenient to check the information of the seated persons, abandoning the manual statistical operation, completing the information conversion through the image acquisition operation of the scene, and completing the combination of multiple images to form a complete image subset of the scene, so as to complete the information recognition and decision result statistical operation, improve efficiency and accuracy while ensuring the authenticity of the information of the decision makers.
[0005] However, in the process of implementing the above-mentioned application embodiments, the present application discovered that the above-mentioned technology has at least the following technical problems: after sampling the received logistics information, the above-mentioned application did not determine whether the obtained sampled samples met the conditions, but directly generated query conditions. The query conditions obtained in this way may have problems of inaccurate and unreliable data. At the same time, the sampled logistics information cannot accurately reflect the overall situation, which will reduce the efficiency of sampling inspection; and the above-mentioned application did not inspect and evaluate the sample information before aggregating the sample information to form a data set, and could not ensure that the sample information obtained met the required standard state, which would lead to errors in the aggregated data set and could not improve the detection efficiency and accuracy of the sampled samples. Summary of the invention
[0006] In view of the deficiencies in the prior art, the present invention provides a method and system for optimizing the efficiency of sequential sampling inspection based on prior information, which can effectively solve the problems involved in the above-mentioned background technology.
[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: The first aspect of the present invention provides a method for optimizing the efficiency of sequential sampling inspections based on prior information, including: statistics of prior qualified products: obtaining prior information of each product to be inspected, evaluating the prior qualification of each product to be inspected, and counting the products to be inspected corresponding to a prior qualification higher than a set prior qualification threshold, which are recorded as prior qualified products; inspection compliance value evaluation: through a sequential sampling inspection model, sequential sampling inspection operations are performed on each prior qualified product to obtain each sampled product, the inspection status of each sampled product is analyzed, and the inspection compliance value of each sampled product is evaluated; sampling inspection efficiency optimization: the inspection compliance value of each sampled product is differenced with a preset inspection compliance preset value to obtain the inspection compliance deviation value of each sampled product, and an inspection efficiency optimization plan for each sampled product is matched, thereby achieving the sequential sampling inspection efficiency optimization state of each sampled product.
[0008] As a further method, the sequential sampling inspection efficiency optimization state of each sampled product is achieved, and the specific analysis process is: the inspection compliance value of each sampled product is subtracted from the preset inspection compliance preset value to obtain the inspection compliance deviation value of each sampled product, and the inspection efficiency optimization plan corresponding to each inspection compliance deviation value interval defined in the product information library is matched to obtain the inspection efficiency optimization plan for each sampled product, thereby achieving the sequential sampling inspection efficiency optimization state of each sampled product.
[0009] As a further method, the inspection compliance value of each sampled product is specifically analyzed as follows: by analyzing the inspection status of each sampled product, the inspection status information of each sampled product is obtained, and the compressive strength of each sampled product within the sampling test period is extracted, and the difference is processed with the set compression boundary strength to obtain the median value of the compression strength of each sampled product within the sampling test period, and the correction factor corresponding to the median value of the compression strength defined in the product information library is multiplied to obtain the compression strength influence coefficient S of each sampled product. j , and analyze the coefficient of influence of the inspection status of each sampled product; comprehensively obtain the inspection compliance value of each sampled product, the specific formula is:
[0010]
[0011] where γ j It is the inspection compliance value of the jth sampled product, j is the number of each sampled product, j = 1, 2, 3, ..., n, n is the number of sampled products, T j It is expressed as the inspection status influence coefficient of the jth sampled product, S j It is expressed as the compression strength influence coefficient of the jth sampling product, θ1 is expressed as the weight factor corresponding to the predefined inspection state influence coefficient, θ2 is expressed as the weight factor corresponding to the predefined compression strength influence coefficient, and e is a natural constant.
[0012] As a further method, the specific analysis process for obtaining each sampled product is as follows: randomly extracting sampling samples from each prior qualified product, performing sequential sampling inspection operations on the sampling samples through a sequential sampling inspection model, and evaluating the standard qualification of the sampling samples based on the inspection results of the sampling samples. If the standard qualification of the sampling samples is higher than the standard qualification threshold, the sampling samples are recorded as each sampled product; otherwise, continue to randomly extract sampling samples and perform sequential sampling inspection operations.
[0013] As a further method, the specific analysis process of each prior qualified product is as follows: the prior qualified degree of each product to be tested is compared with the set prior qualified threshold value. If the prior qualified degree of a product to be tested is higher than the set prior qualified threshold value, the product to be tested is recorded as a prior qualified product. Thus, the products to be tested whose prior qualified degree is higher than the set prior qualified threshold value are statistically recorded as prior qualified products.
[0014] As a further method, the specific analysis process of the prior qualification of each product to be tested is as follows: analyzing the physical property influence coefficient of each product to be tested, multiplying it with the weight factor corresponding to the set physical property influence coefficient, to obtain the physical property influence index of each product to be tested; analyzing the material property influence coefficient of each product to be tested, multiplying it with the weight factor corresponding to the set material property influence coefficient, to obtain the material property influence index of each product to be tested; adding the physical property influence index of each product to be tested and the material property influence index of each product to be tested to obtain the prior qualification of each product to be tested.
[0015] The second aspect of the present invention provides a system for a method for optimizing the efficiency of sequential sampling inspection based on prior information, including: a priori qualified product statistics module, which is used to obtain the prior information of each product to be inspected, evaluate the prior qualification of each product to be inspected, and count the products to be inspected corresponding to a priori qualification higher than the set prior qualification threshold, which are recorded as prior qualified products, and thus enter the inspection compliance value evaluation module; the inspection compliance value evaluation module, which is used to perform sequential sampling inspection operations on each priori qualified product through a sequential sampling inspection model to obtain each sampled product, analyze the inspection status of each sampled product, evaluate the inspection compliance value of each sampled product, and enter the sampling inspection efficiency optimization module; the sampling inspection efficiency optimization module, which is used to perform difference processing on the inspection compliance value of each sampled product and a preset inspection compliance preset value to obtain the inspection compliance deviation value of each sampled product, and match the inspection efficiency optimization plan of each sampled product, so as to achieve the sequential sampling inspection efficiency optimization state of each sampled product.
[0016] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:
[0017] (1) The present invention provides a method and system for optimizing the efficiency of sequential sampling inspection based on prior information. First, each prior qualified product is obtained. Then, each sampled product is obtained through a sequential sampling inspection model, and the inspection compliance value of each sampled product is evaluated. Then, an inspection efficiency optimization plan for each sampled product is matched. By optimizing the sequential sampling inspection efficiency of each sampled product, the sampling sample capacity can be specified in advance, thereby reducing the waste of resources that may be caused by an excessively large sample capacity. At the same time, the decision of accepting or rejecting the batch of products can be made more quickly and accurately, thereby improving the overall inspection efficiency.
[0018] (2) The present invention obtains prior information of each product to be tested and evaluates the prior qualification of each product to be tested, so that more accurate decisions can be made when performing sequential sampling operations, and the sampling samples selected during sequential sampling can be more random; at the same time, the evaluation of prior qualification can help inspectors determine which products are more likely to pass the inspection and which products require more inspection operations, which helps to concentrate inspection resources on products that are most likely to have problems, thereby improving the overall efficiency of product inspection.
[0019] (3) The present invention obtains each sampled product through a sequential sampling inspection model, analyzes the inspection status of each sampled product, and by evaluating the inspection compliance value of each sampled product, the inspectors can ensure that the product meets the prescribed inspection standards, which helps to ensure the safety and reliability of the product; at the same time, it can enable the inspectors to discover possible problems in the production process and take effective measures in a timely manner to improve the production compliance of the product. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.
[0021] Figure 1 The figure is a schematic flow chart of the method steps of the present invention.
[0022] Figure 2 It is a schematic diagram of system module connection of the present invention.
[0023] Figure 3 It is the apparent density broken line involved in the present invention.
[0024] As shown in the attached figure: 1. Apparent density growth straight line; 2. First horizontal straight line. DETAILED DESCRIPTION
[0025] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0026] Reference Figure 1 As shown, the first aspect of the present invention provides a method for optimizing the efficiency of sequential sampling inspection based on prior information, including: statistics of prior qualified products: obtaining prior information of each product to be tested, evaluating the prior qualification of each product to be tested, and counting the products to be tested whose prior qualification is higher than the set prior qualification threshold, which are recorded as prior qualified products.
[0027] In this embodiment, the above-mentioned products to be tested may be household products, mechanical products, building materials and the like, and their application scenarios may include but are not limited to the product sampling and testing process of household products after production on the production line.
[0028] Specifically, the physical property influence coefficient of each product to be tested is analyzed in the following process:
[0029] Based on the prior information of each product to be tested, the physical property correlation information of each product to be tested and the material property correlation information of each product to be tested are extracted.
[0030] According to the physical property correlation information of each product to be tested, wherein the physical property correlation information includes the length, width and height of each product to be tested, wherein the specific values of the length, width and height of each product to be tested can be obtained by the length sensor carried by the production testing equipment itself, and the apparent volume of each product to be tested is obtained by the volume algorithm. It should be explained that the apparent volume of each product to be tested in this embodiment includes the solid volume of the product and the closed pore volume of the product.
[0031] In this embodiment, the above-mentioned apparent density refers to the mass per unit volume of the product to be tested, which usually includes the pores inside the product and the gaps between particles; it is different from the true density of the product, also called volume density or solid density, which refers only to the density of the solid part of the product, excluding the pores therein.
[0032] The mass of each product to be tested is obtained, wherein the mass of each product to be tested can be obtained by a mass sensor carried by the production testing equipment itself, and the mass of each product to be tested and the apparent volume of each product to be tested are comprehensively processed to obtain the apparent density of each product to be tested, wherein the specific processing process of the apparent density of each product to be tested is: ratio processing of the mass of each product to be tested and the volume of each product to be tested is performed to obtain the apparent density of each product to be tested, and an apparent density polyline of the product to be tested is constructed, from which the maximum angle value of the apparent density change of the product to be tested is extracted.
[0033] According to the apparent density of each product to be tested, the apparent density polyline of the product to be tested is constructed, such as Figure 3 As shown, the horizontal axis of the apparent density line is the product to be tested, and the unit is pieces; the vertical axis is the apparent density, and the unit is grams / cubic centimeter.
[0034] According to the apparent density polyline, an apparent density growth straight line 1 between the minimum apparent density value and the maximum apparent density value is located, and according to Figure 3The first horizontal straight line 2 in the figure is obtained, and the minimum angle value of the intersection between the apparent density growth straight line 1 and the first horizontal straight line 2 is obtained, which is recorded as the maximum angle value of the apparent density change of the product to be tested.
[0035] The length and width of each product to be detected are subjected to difference processing with the length threshold and width threshold defined in the product information library, respectively, to obtain the length deviation value and width deviation value of each product to be detected.
[0036] The maximum angle reference value of apparent density change, the length deviation limit value and the width deviation limit value are extracted from the product information library.
[0037] The data defined in the above-mentioned product information library is obtained by fitting based on the historical data of multiple product production lines. For example, the length deviation limit value is obtained by obtaining multiple length deviation values of the product through different historical data on the product production line, and the inspection personnel determine whether the product status under different deviation values meets the inspection standards. The length deviation values that meet the inspection standards are averaged and used as the length deviation limit value. In this way, the data in the product information library is obtained, and these data will change due to different factors such as the sensor status on the product production line, the status of the product, and the status of the inspection personnel. This embodiment does not make any special limitations on this.
[0038] The influence coefficients of the physical properties of each product to be tested are comprehensively analyzed. In a specific embodiment, the analytic hierarchy process can be used to invite experts to compare the physical properties in pairs, so as to determine their relative weights in the hierarchy. By calculating the comprehensive scores of these weights, the influence coefficients of the physical properties of each product to be tested can be obtained.
[0039] The above-mentioned physical property influence coefficient of each product to be detected is obtained by comprehensive analysis of the maximum angle value of the apparent density change of the product to be detected, the length deviation value of each product to be detected, and the width deviation value of each product to be detected in this embodiment, so as to determine the value of the influence degree of the physical property of each product to be detected. In this embodiment, a more accurate calculation method is used to obtain it, and the specific expression is:
[0040]
[0041] Where W iIt is expressed as the coefficient of influence of the physical properties of the ith product to be tested. In this embodiment, the larger the maximum angle value of the apparent density change of the product to be tested is, the larger the span of the apparent density of the product to be tested is, and the more uneven the material apparent density of the product to be tested is, that is, the more unstable the state of the product to be tested is, which will affect the length deviation value and the width deviation value of each product to be tested, causing them to increase, resulting in a decrease in the performance of the product to be tested. Therefore, a correlation analysis is performed on this expression to ensure that the physical properties of the product to be tested are sufficient to meet the required requirements.
[0042] i represents the serial number of each product to be tested, i=1, 2, 3, ..., m, and m represents the number of products to be tested.
[0043] JD represents the maximum angle value of the apparent density change of the product to be tested, which refers to the minimum angle value of the intersection between the apparent density growth line of the product to be tested and the first horizontal line. The larger the maximum angle value of the apparent density change, the larger the change span of the apparent density of the product to be tested.
[0044] JD′ represents the reference value of the maximum angle of apparent density change, which refers to the standard value of the maximum angle of apparent density change.
[0045] CD i It is expressed as the length deviation value of the i-th product to be detected, which refers to the deviation between the actual measured length of the product to be detected and the defined length threshold.
[0046] CD' represents the length deviation limit value, which refers to the maximum value allowed for the length deviation.
[0047] In this embodiment, the length deviation value of the i-th product to be detected is set to 0.5, the length deviation limit value is set to 0.8, and the correction factor corresponding to the length deviation value is set to 0.62. Then, in this expression The partial calculation result is 0.39.
[0048] KD i It is expressed as the width deviation value of the i-th product to be detected, which refers to the deviation between the actual measured width of the product to be detected and the defined width threshold.
[0049] KD′ represents the width deviation limit value, which refers to the maximum value allowed for the width deviation.
[0050] b1 represents the correction factor corresponding to the predefined maximum angle value of the apparent density change, b2 represents the correction factor corresponding to the predefined length deviation value, and b3 represents the correction factor corresponding to the predefined width deviation value, wherein the correction factor corresponding to the maximum angle value of the apparent density change, the correction factor corresponding to the length deviation value, and the correction factor corresponding to the width deviation value are all calculated by collecting the apparent density of the product at different angles, different length deviation values, and different width deviation values, and using polynomial fitting to calculate the specific values of the correction factors; in this embodiment, the correction factor corresponding to the maximum angle value of the apparent density change ranges from 0.33 to 0.43, the correction factor corresponding to the length deviation value ranges from 0.58 to 0.68, the correction factor corresponding to the width deviation value ranges from 0.4 to 0.6, and e is a natural constant.
[0051] Furthermore, the material property influence coefficient of each product to be tested is specifically analyzed as follows:
[0052] According to the material property correlation information of each product to be tested, the temperature value of each product to be tested at each testing time point is extracted from the material property correlation information, wherein each testing time point is obtained by dividing the testing period into each testing time point, and the determination of the testing period is obtained by the testing personnel through analyzing historical data and combining their own experience; the temperature value can be obtained by a temperature sensor, and the temperature mean of each product to be tested in the testing period is obtained by mean processing, and the tensile strength of each product to be tested in the testing period is obtained by Arrhenius model analysis, wherein the Arrhenius model is used to describe the change of the tensile strength of the product with temperature. When the temperature rises, the microstructure of the product may change, resulting in a decrease in tensile strength.
[0053] The temperature value of each product to be detected at the detection termination time point is obtained, wherein the detection termination time point is the last time point in the detection time points, and the temperature average value of each product to be detected in the detection cycle is recorded as the temperature value of each product to be detected at the detection initial time point, and the temperature value of each product to be detected at the detection initial time point is processed by difference with the temperature value of each product to be detected at the detection termination time point to obtain the temperature deviation value of each product to be detected in the detection cycle, and match it with the thermal conductivity corresponding to each temperature deviation value interval defined in the product information library. The specific matching process is: set each temperature deviation value interval, determine the deviation value interval to which the temperature deviation value of each product to be detected in the detection cycle belongs, assign the thermal conductivity corresponding to the temperature deviation value interval to the product to be detected corresponding to the temperature deviation value, and obtain the thermal conductivity of each product to be detected in the detection cycle.
[0054] It should be explained that, in this embodiment, a constant heating device is applied to each product to be tested to increase its temperature. The heating time is the duration corresponding to a detection cycle. By obtaining the temperature value of each product to be tested at the detection termination time point as described above, the temperature difference of each product to be tested before and after heating is obtained, so as to obtain the thermal conductivity of each product to be tested during the detection cycle.
[0055] From the product information library, the tensile reference strength and thermal conductivity reference values are extracted.
[0056] A comprehensive analysis is performed on the influence coefficient of the material properties of each product to be tested. In a specific embodiment, a decision tree algorithm can be used to predict the relationship between product performance and different material properties through a large amount of historical data. The influence coefficient of the material properties of each product to be tested can be determined by training the model and observing its sensitivity to different input features.
[0057] The above-mentioned material property influence coefficient of each product to be tested is obtained by comprehensive analysis of the tensile strength of each product to be tested during the test period and the thermal conductivity of each product to be tested during the test period in this embodiment, and is used to determine the value of the material property influence degree of each product to be tested. In this embodiment, a more accurate calculation method is used to obtain the value, and the specific expression is:
[0058]
[0059] Wherein αi represents the coefficient of influence of the material properties of the i-th product to be tested. In this embodiment, if the tensile strength of the product to be tested is greater, it means that the product is not easy to be deformed or aged due to external factors such as temperature, humidity, etc., thereby maintaining the performance of the product stable, so that the thermal conductivity of the product is reduced accordingly. At the same time, this expression is simplified so that product testers can better solve the negative impact brought about by the different material properties of each product to be tested.
[0060] i represents the serial number of each product to be tested, i=1, 2, 3, ..., m, and m represents the number of products to be tested.
[0061] LSi is represented by the tensile strength of the i-th product to be tested within the test cycle, which refers to the ability of the product to be tested to resist fracture when subjected to tensile stress; the greater the tensile strength, the less likely the product is to break when subjected to tensile force.
[0062] LS' stands for tensile reference strength, which refers to the standard value specified for tensile strength.
[0063] DRi is expressed as the thermal conductivity of the i-th product to be tested during the test cycle, which refers to the ability of the product to be tested to conduct heat; the higher the thermal conductivity, the faster the product to be tested transfers heat under a temperature gradient.
[0064] DR' is the thermal conductivity reference value, which is the standard value specified for thermal conductivity.
[0065] b4 represents the correction factor corresponding to the predefined tensile strength, and b5 represents the correction factor corresponding to the predefined thermal conductivity, wherein the correction factor corresponding to the tensile strength and the correction factor corresponding to the thermal conductivity are both obtained by designing tensile and thermal conductivity experiments to measure the tensile strength and thermal conductivity under different conditions. According to the experimental data, the relationship between the tensile strength and thermal conductivity and various influencing factors is analyzed, and the mathematical model between the tensile strength and thermal conductivity and the influencing factors is fitted using regression analysis. The correction factor is the parameter in the mathematical model, and e is a natural constant.
[0066] In this embodiment, the correction factor corresponding to the tensile strength ranges from 0.55 to 0.6, and the correction factor corresponding to the tensile strength is set to 0.58; the correction factor corresponding to the thermal conductivity ranges from 0.57 to 0.65, and the correction factor corresponding to the thermal conductivity is set to 0.62; the tensile reference strength is set to 120 MPa; and the thermal conductivity reference value is set to 0.15 watts per meter Kelvin.
[0067] In this embodiment, the material property influence coefficient of the i-th product to be tested is as shown in Table 1:
[0068] Table 1 The influence coefficient of material properties of the i-th product to be tested
[0069]
[0070] In this embodiment, as the tensile strength of the product to be tested during the testing cycle and the thermal conductivity of the product to be tested during the testing cycle increase, the material property influence coefficient of the product to be tested will also increase significantly. It can be seen that there is a positive correlation between the material property influence coefficient of the product to be tested and the tensile strength of the product to be tested during the testing cycle and the thermal conductivity of the product to be tested during the testing cycle. In order to reduce the influence of different material properties of the product to be tested on the test results, the tensile strength of the product to be tested during the testing cycle and the thermal conductivity of the product to be tested during the testing cycle should be reduced to ensure that the state of the product to be tested is optimal.
[0071] Specifically, the specific analysis process of the priori qualification of each product to be tested is as follows:
[0072] The physical property influence coefficient of each product to be tested is multiplied by the weight factor corresponding to the set physical property influence coefficient to obtain the physical property influence index of each product to be tested; the material property influence coefficient of each product to be tested is multiplied by the weight factor corresponding to the set material property influence coefficient to obtain the material property influence index of each product to be tested.
[0073] The weight factors corresponding to the above-mentioned physical property influence coefficients and the weight factors corresponding to the material property influence coefficients are evaluated by inviting experts in related fields to determine the degree of influence of different material properties on performance. Experts can give weight factors for different material properties based on their own experience and professional knowledge; in this embodiment, the weight factors corresponding to the physical property influence coefficients range from 0.45 to 0.52; the weight factors corresponding to the material property influence coefficients range from 0.47 to 0.55.
[0074] The physical property influencing index of each product to be tested is added to the material property influencing index of each product to be tested to obtain the prior qualification of each product to be tested.
[0075] Furthermore, the specific analysis process of each prior qualified product is as follows:
[0076] The prior qualification degree of each product to be tested is compared with the set prior qualification threshold, where the prior qualification threshold is extracted from the product information database. If the prior qualification degree of a product to be tested is higher than the set prior qualification threshold, the product to be tested is recorded as a prior qualified product. The number of products to be tested whose prior qualification degree is higher than the set prior qualification threshold is counted and recorded as each prior qualified product.
[0077] Inspection compliance value assessment: Through the sequential sampling inspection model, each prior qualified product is subjected to sequential sampling inspection operations to obtain each sampled product, and the inspection status of each sampled product is analyzed to evaluate the inspection compliance value of each sampled product.
[0078] Specifically, the specific analysis process of obtaining each sampled product is as follows:
[0079] Randomly select sampling samples from each prior qualified product, and perform sequential sampling inspection on the sampling samples through a sequential sampling inspection model. The specific operation process is: first, the number of samples is not fixed. According to actual needs and judgment criteria, one or a certain number of products are selected from the batch for inspection each time; after each sampling and inspection of the product, the accumulated inspection results are compared with the judgment criteria of that time to obtain the inspection results of the sampling samples, and a judgment of acceptance, rejection or continued inspection is made based on the inspection results. The standard qualification of the sampling samples is evaluated based on the number of acceptance, rejection or continued inspection judgments. If the standard qualification of the sampling samples is higher than the standard qualification threshold, where the standard qualification threshold is the standard specified by the industry to which the tested product belongs, then this sampling sample is recorded as each sampling product; otherwise, continue to randomly select sampling samples and perform sequential sampling inspection operations.
[0080] Furthermore, the specific analysis process of the inspection status influence coefficient of each sampled product is as follows:
[0081] By analyzing the inspection status of each sampled product, the inspection status information of each sampled product is obtained, and the inspection weight and inspection diameter of each sampled product are extracted from the inspection status information, wherein the inspection weight and inspection diameter of each sampled product can be obtained by a mass sensor and a length sensor.
[0082] From the product information database, the inspection limit weight and the inspection limit diameter are extracted.
[0083] The inspection weight of each sampled product is subtracted from the inspection limit weight to obtain the median value of the inspection weight influence of each sampled product. This value is multiplied by the correction factor corresponding to the set median value of the inspection weight influence to obtain the inspection weight influence index of each sampled product.
[0084] The inspection diameter of each sampled product is subtracted from the inspection boundary diameter to obtain the median value of the inspection diameter influence of each sampled product. This value is multiplied by the correction factor corresponding to the set median value of the inspection diameter influence to obtain the inspection diameter influence index of each sampled product.
[0085] The correction factors corresponding to the intermediate values of the inspection weight influence and the correction factors corresponding to the intermediate values of the inspection diameter influence are directly extracted from the calibration certificates of the mass sensor and the length sensor; in this embodiment, the correction factors corresponding to the intermediate values of the inspection weight influence range from 0.47 to 0.56; the correction factors corresponding to the intermediate values of the inspection diameter influence range from 0.49 to 0.6.
[0086] Add the inspection weight influence index of each sampled product and the inspection diameter influence index of each sampled product to obtain the inspection status influence coefficient T of each sampled product.j .
[0087] Specifically, the inspection compliance value of each sampled product is analyzed in the following specific process:
[0088] According to the inspection status information of each sampled product, the compressive strength of each sampled product within the sampling test period is extracted, wherein the compressive strength can be directly measured by a compression testing machine, and the difference is processed with the set compression limit strength, wherein the compression limit strength can be directly extracted from the product information library, and the median value of the compression strength of each sampled product within the sampling test period is obtained, and the correction factor corresponding to the median value of the compression strength defined in the product information library is multiplied to obtain the compression strength influence degree coefficient S of each sampled product. j .
[0089] The main function of the correction factor corresponding to the above-mentioned intermediate value of the compression strength is to compensate or reduce the influence of the system error on the compression strength influence coefficient. When there is a systematic error in the result of the intermediate value of the compression strength, multiplying it by the correction factor can make the corrected compression strength influence coefficient closer to the true value, thereby improving the accuracy and reliability of the compression strength influence coefficient; in this embodiment, the correction factor corresponding to the intermediate value of the compression strength ranges from 0.56 to 0.65.
[0090] The inspection compliance value of each sampled product is comprehensively analyzed. In a specific embodiment, it can be obtained from the industry's own industry standards or technical specifications, which will include the inspection compliance value of each sampled product.
[0091] The inspection compliance values of the above-mentioned sampled products are obtained by comprehensive analysis of the inspection status influence coefficient of each sampled product and the compression strength influence coefficient of each sampled product in this embodiment, so as to judge the inspection compliance value of each sampled product. In this embodiment, a more accurate calculation method is adopted to obtain the compliance value, and the specific formula is:
[0092]
[0093] where γ j It is expressed as the inspection compliance value of the jth sampled product. In this embodiment, when the inspection state influence coefficient of the sampled product is too high, it means that the inspection state of the sampled product is not good at this time. The sampled product may be affected by external forces, such as compression force, which will cause the compression strength influence coefficient of the sampled product to increase, so that the inspection compliance value of the sampled product decreases. Therefore, this expression is simplified to weaken the negative impact of the inspection state influence coefficient of the sampled product and the compression strength influence coefficient of the sampled product on the inspection compliance value of the sampled product.
[0094] j represents the serial number of each sampled product, j=1,2,3,...,n, and n represents the number of sampled products.
[0095] T j It is expressed as the inspection status influence coefficient of the j-th sampling product, which represents the numerical value of the inspection status influence degree of each sampling product obtained by comprehensive analysis of the inspection weight and inspection diameter of each sampling product.
[0096] S j It is expressed as the compression strength influence coefficient of the j-th sampling product, which means the value of the compression strength influence degree of each sampling product obtained by comprehensive analysis of the compression strength and compression boundary strength of each sampling product within the sampling test period.
[0097] θ1 represents the weight factor corresponding to the predefined inspection status influence coefficient, and θ2 represents the weight factor corresponding to the predefined compression strength influence coefficient, where the weight factor corresponding to the inspection status influence coefficient and the weight factor corresponding to the compression strength influence coefficient are both analyzed and integrated with the historical inspection status influence coefficient and the compression strength influence coefficient, and the specific values of the weight factors are determined by experts in related fields based on their own experience, and e is a natural constant.
[0098] In this embodiment, the weight factor corresponding to the inspection state influence coefficient ranges from 0.6 to 0.7, and the weight factor corresponding to the inspection state influence coefficient is set to 0.66. The weight factor corresponding to the compression strength influence coefficient ranges from 0.58 to 0.68, and the weight factor corresponding to the inspection state influence coefficient is set to 0.6.
[0099] In this embodiment, the inspection compliance values of the above-mentioned sampled products are as shown in Table 2:
[0100] Table 2 Inspection compliance values of each sampled product
[0101]
[0102] In this embodiment, as the inspection status influence coefficient of the sampling product and the compression strength influence coefficient of the sampling product decrease, the inspection compliance value of the sampling product increases accordingly. It can be seen that the inspection status influence coefficient of the sampling product and the compression strength influence coefficient of the sampling product are negatively correlated with the inspection compliance value of the sampling product. Therefore, if the inspection compliance value of the sampling product is to be increased, it is necessary to reduce the negative impact brought about by the inspection status influence coefficient of the sampling product and the compression strength influence coefficient of the sampling product to ensure that the inspection compliance value of the sampling product meets the specification standard.
[0103] Sampling inspection efficiency optimization: perform difference processing on the inspection compliance value of each sampled product and the preset inspection compliance value to obtain the inspection compliance deviation value of each sampled product, and match the inspection efficiency optimization plan of each sampled product to achieve the optimization state of sequential sampling inspection efficiency of each sampled product.
[0104] Furthermore, the specific analysis process of achieving the optimization state of sequential sampling inspection efficiency of each sampled product is as follows:
[0105] The inspection compliance value of each sampled product is subtracted from the preset inspection compliance value, where the preset inspection compliance value is obtained by the inspector based on multiple inspection compliance values and his own inspection summary, to obtain the inspection compliance deviation value of each sampled product, and match it with the inspection efficiency optimization plan corresponding to each inspection compliance deviation value interval defined in the product information library. The specific matching process is: setting the inspection compliance deviation value interval, determining the interval to which the inspection compliance deviation value of each sampled product belongs, assigning the inspection efficiency optimization plan corresponding to the interval to each sampled product corresponding to the inspection compliance deviation value, and obtaining the inspection efficiency optimization plan for each sampled product, thereby achieving the sequential sampling inspection efficiency optimization state for each sampled product.
[0106] The specific optimization process is as follows: first, analyze the reasons that lead to the low efficiency of sequential sampling inspection, which may be insufficient equipment performance, cumbersome inspection process, human operational errors, etc.; based on the cause analysis, formulate corresponding optimization plans, such as introducing advanced inspection equipment and technology, simplifying the inspection process, and strengthening employee training; establish a monitoring mechanism to monitor the optimized inspection process in real time to ensure that various measures are effectively implemented, regularly evaluate the optimization effect, understand whether the set goals are achieved, and make adjustments and improvements based on the evaluation results, so as to achieve the optimization state of sequential sampling inspection efficiency for each sampled product.
[0107] Reference Figure 2 As shown, the second aspect of the present invention provides a system for a sequential sampling inspection efficiency optimization method based on prior information, including: a priori qualified product statistics module, an inspection compliance value evaluation module and a sampling inspection efficiency optimization module.
[0108] The second aspect of the present invention provides a system for optimizing the efficiency of sequential sampling inspections based on prior information, and also includes a product information library, which is used to store length thresholds, width thresholds, reference values for the maximum angle of apparent density change, length deviation limit values, width deviation limit values, thermal conductivity corresponding to each temperature deviation value interval, tensile reference strength, thermal conductivity reference value, inspection limit weight, inspection limit diameter, correction factors corresponding to the intermediate value of compressive strength, and inspection efficiency optimization plans corresponding to each inspection compliance deviation value interval.
[0109] The a priori qualified product statistics module is connected to the inspection compliance value evaluation module, the inspection compliance value evaluation module is connected to the sampling inspection efficiency optimization module, and the a priori qualified product statistics module, the inspection compliance value evaluation module and the sampling inspection efficiency optimization module are all connected to the product information database.
[0110] The a priori qualified product statistics module is used to obtain a priori information of each product to be tested, evaluate the a priori qualified degree of each product to be tested, and count the products to be tested whose a priori qualified degree is higher than the set a priori qualified threshold value, which are recorded as each a priori qualified product.
[0111] The inspection compliance value evaluation module is used to perform sequential sampling inspection operations on each prior qualified product through a sequential sampling inspection model to obtain each sampled product, analyze the inspection status of each sampled product, and evaluate the inspection compliance value of each sampled product.
[0112] The sampling inspection efficiency optimization module is used to perform difference processing on the inspection compliance value of each sampled product and the preset inspection compliance preset value to obtain the inspection compliance deviation value of each sampled product, and match the inspection efficiency optimization plan of each sampled product, so as to achieve the sequential sampling inspection efficiency optimization state of each sampled product.
[0113] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.
Claims
1. A method for optimizing the efficiency of sequential sampling inspection based on prior information, characterized in that: include: Prior qualified product statistics: obtain the prior information of each product to be tested, evaluate the prior qualified degree of each product to be tested, and count the products to be tested whose prior qualified degree is higher than the set prior qualified threshold, and record them as prior qualified products; Inspection compliance value assessment: Through the sequential sampling inspection model, each prior qualified product is subjected to sequential sampling inspection operations to obtain each sampled product, analyze the inspection status of each sampled product, and evaluate the inspection compliance value of each sampled product; Sampling inspection efficiency optimization: perform difference processing on the inspection compliance value of each sampled product and the preset inspection compliance preset value to obtain the inspection compliance deviation value of each sampled product, and match the inspection efficiency optimization plan of each sampled product, so as to achieve the optimization state of sequential sampling inspection efficiency of each sampled product; The specific analysis process of the inspection compliance values of each sampled product is as follows: By analyzing the inspection status of each sampled product, the inspection status information of each sampled product is obtained, and the compression strength of each sampled product within the sampling test period is extracted, and the difference is processed with the set compression boundary strength to obtain the median value of the compression strength of each sampled product within the sampling test period. The correction factor corresponding to the median value of the compression strength defined in the product information library is multiplied to obtain the compression strength influence coefficient S of each sampled product. j , and analyze the coefficient of influence of the inspection status of each sampled product; The inspection compliance value of each sampled product is obtained comprehensively, and the specific formula is: where γ j It is the inspection compliance value of the jth sampled product, j is the number of each sampled product, j = 1, 2, 3, ..., n, n is the number of sampled products, T j It is expressed as the inspection status influence coefficient of the jth sampled product, S j It is represented as the compression strength influence coefficient of the jth sampled product, θ1 is represented as the weight factor corresponding to the predefined inspection state influence coefficient, θ2 is represented as the weight factor corresponding to the predefined compression strength influence coefficient, and e is a natural constant; The specific analysis process of the inspection status influence coefficient of each sampled product is as follows: According to the inspection status information of each sampled product, the inspection weight and inspection diameter of each sampled product are extracted from the inspection status information; Extract the inspection limit weight and inspection limit diameter from the product information database; Perform difference processing on the inspection weight of each sampled product and the inspection limit weight to obtain the inspection weight impact median value of each sampled product, and multiply it by the correction factor corresponding to the set inspection weight impact median value to obtain the inspection weight impact index of each sampled product; Perform difference processing on the inspection diameter of each sampled product and the inspection definition diameter to obtain the inspection diameter influence median value of each sampled product, and multiply it by the correction factor corresponding to the set inspection diameter influence median value to obtain the inspection diameter influence index of each sampled product; Add the inspection weight influence index of each sampled product and the inspection diameter influence index of each sampled product to obtain the inspection status influence coefficient T of each sampled product. j ; The specific analysis process of the prior qualification of each product to be tested is as follows: Analyze the physical property influence coefficient of each product to be tested, and multiply it with the weight factor corresponding to the set physical property influence coefficient to obtain the physical property influence index of each product to be tested; analyze the material property influence coefficient of each product to be tested, and multiply it with the weight factor corresponding to the set material property influence coefficient to obtain the material property influence index of each product to be tested; Add the physical property influencing index of each product to be tested and the material property influencing index of each product to be tested to obtain the prior qualification of each product to be tested; The specific analysis process of the influence degree coefficient of the physical properties of each product to be tested is as follows: Extracting physical property correlation information of each product to be tested and material property correlation information of each product to be tested based on prior information of each product to be tested; According to the physical property association information of each product to be tested, wherein the physical property association information includes the length, width and height of each product to be tested, the apparent volume of each product to be tested is obtained by a volume algorithm; Obtain the mass of each product to be tested, and perform comprehensive processing with the apparent volume of each product to be tested to obtain the apparent density of each product to be tested, construct an apparent density polyline of the product to be tested, and extract the maximum angle value of the apparent density change of the product to be tested; The length and width of each product to be detected are processed by difference with the length threshold and width threshold defined in the product information library to obtain the length deviation value and width deviation value of each product to be detected; Extract the maximum angle reference value of apparent density change, the length deviation limit value and the width deviation limit value of each product to be tested from the product information database, and comprehensively analyze the influence coefficient of the physical properties of each product to be tested; The material property influence coefficient of each product to be tested is specifically analyzed as follows: According to the material property correlation information of each product to be tested, the temperature value of each product to be tested at each testing time point is extracted from the material property correlation information, the temperature mean value of each product to be tested in the testing cycle is obtained by mean processing, and the tensile strength of each product to be tested in the testing cycle is obtained by Arrhenius model analysis; The temperature average of each product to be tested in the testing cycle is recorded as the temperature value of each product to be tested at the initial time point of the testing, and the difference is processed with the temperature value of each product to be tested at the end time point of the testing to obtain the temperature deviation value of each product to be tested in the testing cycle, and the temperature deviation value is matched with the thermal conductivity corresponding to each temperature deviation value interval defined in the product information library to obtain the thermal conductivity of each product to be tested in the testing cycle; From the product information database, the tensile reference strength and thermal conductivity reference values are extracted, and the influence coefficient of the material properties of each product to be tested is comprehensively analyzed.
2. The method for optimizing sequential sampling inspection efficiency based on prior information according to claim 1, characterized in that: The specific analysis process of achieving the optimization state of sequential sampling inspection efficiency of each sampled product is as follows: The inspection compliance value of each sampled product is subtracted from the preset inspection compliance value to obtain the inspection compliance deviation value of each sampled product, which is then matched with the inspection efficiency optimization plan corresponding to each inspection compliance deviation value interval defined in the product information library to obtain the inspection efficiency optimization plan for each sampled product, thereby achieving the sequential sampling inspection efficiency optimization state for each sampled product.
3. The method for optimizing sequential sampling inspection efficiency based on prior information according to claim 1, characterized in that: The specific analysis process of obtaining each sampled product is as follows: Randomly select sampling samples from each prior qualified product, and perform sequential sampling inspection operations on the sampling samples through the sequential sampling inspection model. According to the inspection results of the sampling samples, the standard qualification of the sampling samples is evaluated. If the standard qualification of the sampling samples is higher than the standard qualification threshold, the sampling samples are recorded as each sampling product; otherwise, continue to randomly select sampling samples and perform sequential sampling inspection operations.
4. The method for optimizing sequential sampling inspection efficiency based on prior information according to claim 1, characterized in that: The specific analysis process of each prior qualified product is as follows: The prior qualification degree of each product to be tested is compared with the set prior qualification threshold. If the prior qualification degree of a product to be tested is higher than the set prior qualification threshold, the product to be tested is recorded as a prior qualified product. Thus, the products to be tested whose prior qualification degree is higher than the set prior qualification threshold are statistically recorded as prior qualified products.
5. A system using the sequential sampling inspection efficiency optimization method based on prior information as claimed in any one of claims 1 to 4, characterized in that: include: The prior qualified product statistics module is used to obtain the prior information of each product to be tested, evaluate the prior qualified degree of each product to be tested, and count the products to be tested whose prior qualified degree is higher than the set prior qualified threshold value, which are recorded as prior qualified products, and then enter the inspection compliance value evaluation module; The inspection compliance value evaluation module is used to perform sequential sampling inspection operations on each prior qualified product through a sequential sampling inspection model to obtain each sampled product, analyze the inspection status of each sampled product, evaluate the inspection compliance value of each sampled product, and enter the sampling inspection efficiency optimization module; The sampling inspection efficiency optimization module is used to perform difference processing on the inspection compliance value of each sampled product and the preset inspection compliance preset value to obtain the inspection compliance deviation value of each sampled product, and match the inspection efficiency optimization plan of each sampled product, so as to achieve the sequential sampling inspection efficiency optimization state of each sampled product.
