Product quality monitoring method, device, equipment and medium based on multi-stage decision-making

Abstract

The present application relates to a product quality monitoring method, device, equipment and medium based on multi-stage decision-making. The method includes: calculating the minimum number of sampling inspections for sampling inspections according to a statistical hypothesis testing model; performing sampling inspections on spare parts using the minimum number of sampling inspections, determining whether to accept spare parts from suppliers using a sampling plan and producing electronic products based on the accepted spare parts, and establishing a multi-stage decision model; solving the multi-stage decision model, considering the defective rate and sampling inspection error in the sampling inspection process to perform sensitivity analysis on the obtained complex production process decision plan, optimizing the sampling inspection process using a Bayesian inference method based on the analysis results, determining an optimized complex production process decision plan based on the optimized sampling plan; and using the optimized complex production process decision plan to monitor product quality. The use of this method can reduce production costs and improve product quality.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a product quality monitoring method, device, equipment and medium based on multi-stage decision-making. Background Art

[0002] With the rapid development of the electronic product manufacturing industry, the market's requirements for product quality are increasing, and manufacturers are facing increasingly stringent quality control challenges. In this context, the quality of spare parts directly affects the performance and reliability of the final product. At the same time, in the production process, the production cost of products is getting higher and higher, and the quality control is not in place, resulting in poor quality. As the production process of products is becoming more and more complicated, it is necessary to propose a method that can reduce production costs and improve product quality. Summary of the invention

[0003] Based on this, it is necessary to provide a product quality monitoring method, device, equipment and medium based on multi-stage decision-making that can reduce production costs and improve product quality in response to the above technical problems.

[0004] A product quality monitoring method based on multi-stage decision making, the method comprising:

[0005] Calculate the minimum number of sampling tests for sampling the spare parts provided by the supplier under a given confidence level and nominal value based on the statistical hypothesis test model; perform sampling tests on the spare parts using the minimum number of sampling tests to obtain a sampling plan;

[0006] Use sampling plans to determine whether to accept parts from suppliers and produce electronic products based on the accepted parts. Consider whether to test parts, semi-finished products, and finished products at each production stage, and how to deal with finished products and defective products to establish a multi-stage decision model;

[0007] Dynamic programming and Monte Carlo methods are used to solve multi-stage decision models and obtain decision plans for complex production processes;

[0008] Considering the defective rate and sampling inspection error in the sampling inspection process, a sensitivity analysis is conducted on the decision-making plan for complex production processes. Based on the analysis results, the sampling inspection process is optimized using the Bayesian inference method to obtain the optimized sampling plan; the optimized decision-making plan for the complex production process is determined based on the optimized sampling plan; and the optimized decision-making plan for the complex production process is used to monitor product quality.

[0009] In one embodiment, sampling inspection is performed on the spare parts provided by the supplier according to the statistical hypothesis testing model to obtain the minimum number of sampling inspections, including:

[0010] According to the statistical hypothesis testing model, the spare parts provided by the supplier are sampled and tested, and the minimum number of sampling tests is obtained:

[0011] ;

[0012] in, The supplier's nominal defective rate, represents the increment of the sample rate, represents the test statistic, Indicates the significance level.

[0013] In one embodiment, a multi-stage decision model is established by considering whether to inspect spare parts, semi-finished products, and finished products at each production stage, and how to deal with finished products and defective products, including:

[0014] At each production stage, consider whether to test spare parts, semi-finished products, and finished products, as well as how to deal with finished products and defective products to establish a multi-stage decision model:

[0015] Objective function:

[0016] ;

[0017] ;

[0018] ;

[0019] ;

[0020] ;

[0021] ;

[0022] ;

[0023] Constraints:

[0024] ;

[0025] in, Indicates the purchase cost, represents the detection cost, represents the assembly cost, represents the switching cost, represents the dismantling cost, m Indicates the number of processes, n Indicates the number of spare parts types. represents the number of semi-finished products in the jth process, Indicates spare parts i The defective rate of , i = 1, 2, ..., n; Indicates spare parts i The purchase price of Indicates spare parts i The cost of testing; Indicates j The assembly cost of the process, j =1, 2, …, m ; Indicates j The first process a The defective rate of a finished product (or semi-finished product), a=1, 2, ..., , represents the jth process a The testing cost of finished products (or semi-finished products); Indicates the market price of the finished product; It indicates the replacement loss of unqualified finished products; Indicates j The first process a Dismantling cost of unqualified finished products (or semi-finished products), j=1, 2, …, m; represents the maximum acceptable defective rate that can flow into the society; K represents the number of recursive cycles, Indicated in k Does the next cycle detect the i spare parts, k=1, 2…, K; Indicated in k Does the next cycle detect the j The first step of the process a finished products, a=1, 2, ..., , :Indicates in k Does the second cycle dismantle the j The first step of the process a Unqualified finished products; Indicates k The defective rate of finished products entering the market in the second cycle.

[0026] In one embodiment, dynamic programming and Monte Carlo methods are used to solve the multi-stage decision model to obtain a complex production process decision plan, including:

[0027] Using the dynamic programming method, starting from the final product, the optimal decision for each stage is reversely calculated and recorded in order to construct the optimal strategy. At the same time, the Monte Carlo method is used to simulate the production process to estimate the expected profit of various decisions, and the plan with the highest profit is selected as the decision plan for the complex production process.

[0028] In one embodiment, a sensitivity analysis is performed on a complex production process decision plan by considering the defective rate and sampling detection error in the sampling detection process, and the sampling detection process is optimized using a Bayesian inference method based on the analysis results, including:

[0029] Considering the defective rate and sampling detection error in the sampling detection process, a sensitivity analysis is conducted on the decision-making plan of the complex production process, and the maximum profit of the decision-making plan of the complex production process under the confidence interval of the defective rate is calculated. It is determined that the defective rate detection error has a great influence on the calculation result of the maximum profit and has high sensitivity. The Bayesian inference method is used to continuously update the estimate of the defective rate based on historical detection data and current sampling detection data, and the optimal balance point between detection accuracy and detection cost is found to optimize the sampling detection process.

[0030] In one embodiment, the defect rate confidence interval is:

[0031] ;

[0032] in, represents the estimated defective rate obtained by sampling, Represents a normal distribution Quantile, n is the sample size, i.e. the number of spare parts types.

[0033] A product quality monitoring device based on multi-stage decision-making, the device comprising:

[0034] The sampling inspection module is used to calculate the minimum number of sampling inspections for sampling inspections of spare parts provided by suppliers under given confidence levels and nominal values ​​according to the statistical hypothesis testing model; the spare parts are sampled and inspected using the minimum number of sampling inspections to obtain a sampling plan;

[0035] A multi-stage decision model building module is used to use a sampling plan to determine whether to accept spare parts from suppliers and produce electronic products based on the accepted spare parts. At each production stage, consider whether to test spare parts, semi-finished products, and finished products, as well as how to deal with finished products and defective products to establish a multi-stage decision model;

[0036] The multi-stage decision model solving module is used to solve the multi-stage decision model using dynamic programming and Monte Carlo methods to obtain the decision plan for complex production processes;

[0037] The product quality monitoring module is used to conduct sensitivity analysis on complex production process decision plans by considering the defective rate and sampling inspection error in the sampling inspection process, optimize the sampling inspection process based on the analysis results using the Bayesian inference method to obtain an optimized sampling plan; determine the optimized complex production process decision plan based on the optimized sampling plan; and use the optimized complex production process decision plan to monitor product quality.

[0038] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0039] Calculate the minimum number of sampling tests for sampling the spare parts provided by the supplier under a given confidence level and nominal value based on the statistical hypothesis test model; perform sampling tests on the spare parts using the minimum number of sampling tests to obtain a sampling plan;

[0040] Use sampling plans to determine whether to accept parts from suppliers and produce electronic products based on the accepted parts. Consider whether to test parts, semi-finished products, and finished products at each production stage, and how to deal with finished products and defective products to establish a multi-stage decision model;

[0041] Dynamic programming and Monte Carlo methods are used to solve multi-stage decision models and obtain decision plans for complex production processes;

[0042] Considering the defective rate and sampling inspection error in the sampling inspection process, a sensitivity analysis is conducted on the decision-making plan for complex production processes. Based on the analysis results, the sampling inspection process is optimized using the Bayesian inference method to obtain the optimized sampling plan; the optimized decision-making plan for the complex production process is determined based on the optimized sampling plan; and the optimized decision-making plan for the complex production process is used to monitor product quality.

[0043] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0044] Calculate the minimum number of sampling tests for sampling the spare parts provided by the supplier under a given confidence level and nominal value based on the statistical hypothesis test model; perform sampling tests on the spare parts using the minimum number of sampling tests to obtain a sampling plan;

[0045] Use sampling plans to determine whether to accept parts from suppliers and produce electronic products based on the accepted parts. Consider whether to test parts, semi-finished products, and finished products at each production stage, and how to deal with finished products and defective products to establish a multi-stage decision model;

[0046] Dynamic programming and Monte Carlo methods are used to solve multi-stage decision models and obtain decision plans for complex production processes;

[0047] Considering the defective rate and sampling inspection error in the sampling inspection process, a sensitivity analysis is conducted on the decision-making plan for complex production processes. Based on the analysis results, the sampling inspection process is optimized using the Bayesian inference method to obtain the optimized sampling plan; the optimized decision-making plan for the complex production process is determined based on the optimized sampling plan; and the optimized decision-making plan for the complex production process is used to monitor product quality.

[0048] The above-mentioned product quality monitoring method, device, equipment and medium based on multi-stage decision-making, this application calculates the minimum number of sampling tests through a statistical hypothesis test model, and the manufacturer can determine the minimum sample size required under the conditions of a given confidence level and nominal value. Using the minimum sample size for testing can save testing costs and time, while ensuring the reliability of the test results at a certain statistical confidence level, preventing unqualified parts from entering the production process, and improving the quality of the final product. The parts provided by the supplier are tested using the minimum number of sampling tests, and a sampling plan is formed based on the test results to determine whether to accept the parts and put them into production. Through scientific sampling testing, unqualified parts can be identified and screened more efficiently, reducing the risk of defective products entering the production line. This not only saves resources for testing all parts one by one, but also ensures that the quality of the parts invested in the production process is passed, improving overall production efficiency and product qualification rate. Then consider whether to conduct testing at each production stage, and reasonably handle unqualified and qualified products, and establish a multi-stage decision model. The production process is divided into multiple stages, and targeted testing and decision-making are carried out at each stage, which can accurately control the quality problems at each stage and reduce the risk of unqualified final products. At the same time, unnecessary comprehensive testing can be avoided, saving testing costs and time. Dynamic programming is used to solve multi-stage decision-making problems through step-by-step optimization. Dynamic programming and Monte Carlo methods can evaluate the cost and quality results under different strategies in a complex production environment, thereby obtaining the optimal decision-making plan for a complex production process. By solving the optimal solution, the cost-effectiveness of each decision can be maximized, and the defective rate and rework cost can be minimized. Finally, the variables that have a significant impact on cost and quality can be identified through sensitivity analysis to ensure that the optimization strategy is applicable to different scenarios. Bayesian inference reduces the probability of misjudgment and further reduces the defective rate by continuously updating and optimizing the detection plan. This can make the sampling detection process more efficient, while improving the accuracy of detection and reducing the risks and additional costs caused by errors. The optimized production decision plan can manage production resources more accurately, reduce waste and duplication of work, and reduce production costs. At the same time, product quality monitoring can ensure that problems in the production process are discovered and solved in a timely manner, thereby improving the reliability and market competitiveness of the product. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A schematic flow chart of a product quality monitoring method based on multi-stage decision-making in one embodiment;

[0050] Figure 2 It is a structural block diagram of a product quality monitoring device based on multi-stage decision-making in one embodiment;

[0051] Figure 3 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0053] In one embodiment, Figure 1 As shown, a product quality monitoring method based on multi-stage decision-making is provided, including the following steps:

[0054] Step 102, calculating the minimum number of sampling tests for sampling the spare parts provided by the supplier under a given confidence level and nominal value based on the statistical hypothesis testing model; sampling the spare parts using the minimum number of sampling tests to obtain a sampling plan.

[0055] Common distribution types of defective spare parts include normal distribution, binomial distribution, hypergeometric distribution, etc. Therefore, a smaller sample experiment can be used to first assume the distribution type, and then perform the corresponding fitting test to determine the distribution type of each defective spare part. Based on the statistical hypothesis test model (two-sided), the minimum number of tests under a given reliability and nominal value is determined. The minimum number of sampling tests is used to sample and test spare parts, and a sampling plan is formed based on the sampling results of each time.

[0056] Through the statistical hypothesis testing model, the minimum sample size for sampling and testing supplier parts can be calculated under the set confidence level and nominal value. This process helps manufacturers minimize testing costs and time while ensuring statistical validity. Through hypothesis testing, the risk of defective products entering the production line can be identified and reduced, ensuring that only parts that meet quality standards are accepted.

[0057] Step 104, using a sampling plan to determine whether to accept the supplier's spare parts and produce electronic products based on the accepted spare parts, considering whether to test the spare parts, semi-finished products, and finished products at each production stage, and how to deal with finished products and defective products to establish a multi-stage decision model.

[0058] Consider whether to conduct testing at each production stage (such as spare parts, semi-finished products, and finished products), and reasonably handle defective and qualified products, and establish a multi-stage decision-making model.

[0059] This model divides the production process into multiple stages, and conducts targeted testing and decision-making at each stage, which can accurately control quality issues at each stage and reduce the risk of unqualified final products. At the same time, it can avoid unnecessary comprehensive testing and save testing costs and time.

[0060] Step 106, using dynamic programming and Monte Carlo methods to solve the multi-stage decision model to obtain a decision plan for the complex production process.

[0061] Dynamic programming solves multi-stage decision-making problems through step-by-step optimization, and the Monte Carlo method can perform simulation calculations under uncertain conditions to obtain the optimal decision-making plan for complex production processes.

[0062] Dynamic programming and Monte Carlo methods can evaluate the cost and quality results of different strategies in a complex production environment. By solving the optimal solution, it can ensure that the cost-effectiveness of each decision is maximized and the defective rate and rework costs are minimized. Monte Carlo simulation can cope with uncertainty, so that the best decision can be made under changing production conditions.

[0063] Step 108, conduct sensitivity analysis on the complex production process decision plan by considering the defective rate and sampling inspection error in the sampling inspection process, optimize the sampling inspection process according to the analysis results using the Bayesian inference method to obtain an optimized sampling plan; determine the optimized complex production process decision plan according to the optimized sampling plan; and use the optimized complex production process decision plan to monitor product quality.

[0064] Sensitivity analysis evaluates the impact of defective rates and detection errors in the sampling and testing process on decision-making results, while Bayesian inference dynamically optimizes the sampling and testing process based on historical data and prior knowledge. Sensitivity analysis helps identify variables that have a significant impact on cost and quality, ensuring that optimization strategies are applicable to different scenarios. Bayesian inference reduces the probability of misjudgment and further reduces the defective rate by continuously updating and optimizing the testing plan. This can make the sampling and testing process more efficient, while improving the accuracy of testing and reducing the risks and additional costs caused by errors. Based on the optimized sampling inspection and multi-stage decision-making model, the decision-making plan for the entire production process is further optimized, and product quality is continuously monitored. The optimized production decision plan can manage production resources more accurately, reduce waste and duplication of work, and reduce production costs. At the same time, product quality monitoring can ensure that problems in the production process are discovered and resolved in a timely manner, thereby improving product reliability and market competitiveness.

[0065] In the above-mentioned product quality monitoring method based on multi-stage decision-making, this application calculates the minimum number of sampling tests through a statistical hypothesis test model, and the manufacturer can determine the minimum sample size required under the conditions of a given confidence level and nominal value. Using the minimum sample size for testing can save testing costs and time, while ensuring the reliability of the test results at a certain statistical confidence level, preventing unqualified parts from entering the production process, and improving the quality of the final product. The parts provided by the supplier are tested using the minimum number of sampling tests, and a sampling plan is formed based on the test results to determine whether to accept the parts and put them into production. Through scientific sampling testing, unqualified parts can be identified and screened more efficiently, reducing the risk of defective products entering the production line. This not only saves resources for testing all parts one by one, but also ensures that the quality of the parts invested in the production process is passed, improving overall production efficiency and product qualification rate. Then consider whether to conduct testing at each production stage, and reasonably handle unqualified and qualified products, and establish a multi-stage decision model. The production process is divided into multiple stages, and targeted testing and decision-making are carried out at each stage, which can accurately control the quality problems at each stage and reduce the risk of unqualified final products. At the same time, unnecessary comprehensive testing can be avoided, saving testing costs and time. Dynamic programming is used to solve multi-stage decision-making problems through step-by-step optimization. Dynamic programming and Monte Carlo methods can evaluate the cost and quality results under different strategies in a complex production environment, thereby obtaining the optimal decision-making plan for a complex production process. By solving the optimal solution, the cost-effectiveness of each decision can be maximized, and the defective rate and rework cost can be minimized. Finally, the variables that have a significant impact on cost and quality can be identified through sensitivity analysis to ensure that the optimization strategy is applicable to different scenarios. Bayesian inference reduces the probability of misjudgment and further reduces the defective rate by continuously updating and optimizing the detection plan. This can make the sampling detection process more efficient, while improving the accuracy of detection and reducing the risks and additional costs caused by errors. The optimized production decision plan can manage production resources more accurately, reduce waste and duplication of work, and reduce production costs. At the same time, product quality monitoring can ensure that problems in the production process are discovered and solved in a timely manner, thereby improving the reliability and market competitiveness of the product.

[0066] In one embodiment, sampling inspection is performed on the spare parts provided by the supplier according to the statistical hypothesis testing model to obtain the minimum number of sampling inspections, including:

[0067] According to the statistical hypothesis testing model, the spare parts provided by the supplier are sampled and tested, and the minimum number of sampling tests is obtained:

[0068] ;

[0069] in, The supplier's nominal defective rate, represents the increment of the sample rate, represents the test statistic, Indicates the significance level.

[0070] In a specific embodiment, it is assumed that , N=1000, the calculation results are as follows.

[0071] (1) Scenario 1: If the defective rate of spare parts exceeds 10% at a confidence level of 95%, the parts will be rejected.

[0072] ①Normal distribution (binomial distribution):

[0073] ;

[0074] ②Hypergeometric distribution: .

[0075] If 139 spare parts are sampled from the population for testing, and the result is 17 defective, then the defective rate is 0.1223. Substituting into the formula The test statistic Z=2.351 is calculated. According to the standard normal distribution table, it is greater than the critical value of 1.96. Therefore, the original hypothesis can be rejected and it is determined that the defective rate exceeds 10%. The company should reject this batch of parts.

[0076] Verification by confidence interval: According to the formula , substituting the data into the calculation, the 95% confidence interval is [0.0412, 0.1588]. Since the upper limit of the confidence interval is 0.1588, which is greater than the nominal value of 0.10, it means that the defective rate exceeds the nominal value, so this batch of spare parts should be rejected.

[0077] (2) Case 2: With a confidence level of 90%, if the defective rate of spare parts does not exceed 10%, they are accepted.

[0078] ①Normal distribution (binomial distribution):

[0079]

[0080] ②Hypergeometric distribution: .

[0081] If 98 samples of spare parts are drawn from the population for testing, and the result is 6 defective products, then the defective rate is 0.0612. Substituting into the formula The test statistic Z=-1.89 is calculated. According to the standard normal distribution table, it is less than the critical value of 1.645. Therefore, the original hypothesis cannot be rejected. It is believed that the defective rate is not significantly greater than 10%, and the company should accept this batch of parts.

[0082] Verification by confidence interval: According to the formula , substituting the data into the calculation, the 90% confidence interval is [0.0745, 0.0954]. Since the upper limit of the confidence interval 0.0845 is less than the nominal value 0.10, this batch of spare parts should be accepted.

[0083] Generally speaking, if the machining process is caused by the combined effect of many independent random errors, and there is no error factor with a significant advantage, the defective rate of processed parts will be normally distributed. When the distribution type is unknown and cannot be fitted and verified temporarily, the conclusion of the normal distribution can be used as a reference.

[0084] In one embodiment, a multi-stage decision model is established by considering whether to inspect spare parts, semi-finished products, and finished products at each production stage, and how to deal with finished products and defective products, including:

[0085] At each production stage, consider whether to test spare parts, semi-finished products, and finished products, as well as how to deal with finished products and defective products to establish a multi-stage decision model.

[0086] Objective function:

[0087] ;

[0088] ;

[0089] ;

[0090] ;

[0091] ;

[0092] ;

[0093] ;

[0094] Constraints:

[0095] ;

[0096] in, Indicates the purchase cost, represents the detection cost, represents the assembly cost, represents the switching cost, represents the dismantling cost, m Indicates the number of processes, n Indicates the number of spare parts types. represents the number of semi-finished products in the jth process, Indicates spare parts i The defective rate of , i = 1, 2, ..., n; Indicates spare parts i The purchase price of Indicates spare parts i The cost of testing; Indicates j The assembly cost of the process, j =1, 2, …, m ; Indicates j The first process a The defective rate of a finished product (or semi-finished product), a=1, 2, ..., , represents the jth process a The testing cost of finished products (or semi-finished products); Indicates the market price of the finished product; It indicates the replacement loss of unqualified finished products; Indicates j The first process a Dismantling cost of unqualified finished products (or semi-finished products), j=1, 2, …, m; represents the maximum acceptable defective rate that can flow into the society; K represents the number of recursive cycles, Indicated in k Does the next cycle detect the i spare parts, k=1, 2…, K; Indicated in k Does the next cycle detect the j The first step of the process a finished products, a=1, 2, ..., , :Indicates in k Does the second cycle dismantle the j The first step of the process a Unqualified finished products; Indicates k The defective rate of finished products entering the market in the second cycle.

[0097] In one embodiment, dynamic programming and Monte Carlo methods are used to solve the multi-stage decision model to obtain a complex production process decision plan, including:

[0098] Using the dynamic programming method, starting from the final product, the optimal decision for each stage is reversely calculated and recorded in order to construct the optimal strategy. At the same time, the Monte Carlo method is used to simulate the production process to estimate the expected profit of various decisions, and the plan with the highest profit is selected as the decision plan for the complex production process.

[0099] In a specific embodiment, after the relevant parameters are valued, the model is solved using dynamic programming and Monte Carlo methods, and the results are shown in Table 1.

[0100] Table 1

[0101]

[0102] Note: The optimal decision solution only shows the decision results of the first cycle:

[0103] .

[0104] In one embodiment, a sensitivity analysis is performed on a complex production process decision plan by considering the defective rate and sampling detection error in the sampling detection process, and the sampling detection process is optimized using a Bayesian inference method based on the analysis results, including:

[0105] Considering the defective rate and sampling detection error in the sampling detection process, a sensitivity analysis is conducted on the decision-making plan of the complex production process, and the maximum profit of the decision-making plan of the complex production process under the confidence interval of the defective rate is calculated. It is determined that the defective rate detection error has a great influence on the calculation result of the maximum profit and has high sensitivity. The Bayesian inference method is used to continuously update the estimate of the defective rate based on historical detection data and current sampling detection data, and the optimal balance point between detection accuracy and detection cost is found to optimize the sampling detection process.

[0106] In one embodiment, the defect rate confidence interval is:

[0107] ;

[0108] in, represents the estimated defective rate obtained by sampling, Represents a normal distribution Quantile, n is the sample size, i.e. the number of spare parts types.

[0109] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0110] In one embodiment, Figure 2As shown, a product quality monitoring device based on multi-stage decision-making is provided, including: a sampling detection module 202, a multi-stage decision model building module 204, a multi-stage decision model solving module 206 and a product quality monitoring module 208, wherein:

[0111] The sampling inspection module 202 is used to calculate the minimum number of sampling inspections for sampling inspections of spare parts provided by suppliers under a given confidence level and nominal value according to a statistical hypothesis testing model; and to perform sampling inspections on spare parts using the minimum number of sampling inspections to obtain a sampling plan;

[0112] A multi-stage decision model building module 204 is used to determine whether to accept the spare parts of the supplier using a sampling plan and produce electronic products based on the accepted spare parts, and to consider whether to test the spare parts, semi-finished products, and finished products at each production stage, and how to deal with finished products and unqualified products to establish a multi-stage decision model;

[0113] The multi-stage decision model solving module 206 is used to solve the multi-stage decision model using dynamic programming and Monte Carlo methods to obtain a complex production process decision solution;

[0114] The product quality monitoring module 208 is used to perform sensitivity analysis on the complex production process decision plan by considering the defective rate and sampling detection error in the sampling detection process, and optimize the sampling detection process according to the analysis results using the Bayesian inference method to obtain an optimized sampling plan; determine the optimized complex production process decision plan according to the optimized sampling plan; and use the optimized complex production process decision plan to monitor product quality.

[0115] For the specific definition of a product quality monitoring device based on multi-stage decision-making, please refer to the definition of a product quality monitoring method based on multi-stage decision-making above, which will not be repeated here. Each module in the above-mentioned product quality monitoring device based on multi-stage decision-making can be implemented in whole or in part through software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0116] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 3As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a product quality monitoring method based on multi-stage decision-making is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0117] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0118] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0119] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0120] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A product quality monitoring method based on multi-stage decision making, characterized in that: The method comprises: Calculating the minimum number of sampling tests for sampling the spare parts provided by the supplier under a given confidence level and nominal value according to a statistical hypothesis testing model; sampling the spare parts using the minimum number of sampling tests to obtain a sampling plan; Using the sampling plan, determine whether to accept the spare parts from the supplier and produce the electronic products based on the accepted spare parts, and consider whether to test the spare parts, semi-finished products, and finished products at each production stage, as well as how to deal with finished products and defective products to establish a multi-stage decision model; The multi-stage decision model is solved by using dynamic programming and Monte Carlo methods to obtain a decision plan for a complex production process; A sensitivity analysis is performed on the complex production process decision plan by considering the defective rate and sampling inspection error in the sampling inspection process, and the sampling inspection process is optimized using the Bayesian inference method based on the analysis results to obtain an optimized sampling plan; an optimized complex production process decision plan is determined based on the optimized sampling plan; and product quality monitoring is performed using the optimized complex production process decision plan.

2. The method according to claim 1, characterized in that According to the statistical hypothesis testing model, the spare parts provided by the supplier are sampled and tested to obtain the minimum number of sampling tests, including: According to the statistical hypothesis testing model, the spare parts provided by the supplier are sampled and tested, and the minimum number of sampling tests is obtained as follows: ； in, The supplier's nominal defective rate, represents the increment of the sample rate, represents the test statistic, Indicates the significance level.

3. The method according to claim 1, characterized in that: At each production stage, consider whether to test spare parts, semi-finished products, and finished products, as well as how to deal with finished products and defective products to establish a multi-stage decision model, including: At each production stage, consider whether to test spare parts, semi-finished products, and finished products, as well as how to deal with finished products and defective products to establish a multi-stage decision model: Objective function: Constraints: in, Indicates the purchase cost, represents the detection cost, represents the assembly cost, represents the switching cost, represents the dismantling cost, m Indicates the number of processes, n Indicates the number of spare parts types. represents the number of semi-finished products in the jth process, Indicates spare parts i The purchase price of Indicates spare parts i The cost of testing; Indicates j The assembly cost of the process, j =1, 2, …, m ; Indicates the market price of the finished product; It indicates the replacement loss of unqualified finished products; represents the maximum acceptable defective rate that can flow into the society; K represents the number of recursive cycles, Indicated in k Does the next cycle detect the i spare parts, k=1, 2…, K; Indicated in k Does the next cycle detect the j The first step of the process a finished products, a=1, 2, ..., , :Indicates in k Does the second cycle dismantle the j The first step of the process a Unqualified finished products; Indicates k The defective rate of finished products entering the market in the second cycle.

4. The method according to claim 1, characterized in that: Dynamic programming and Monte Carlo methods are used to solve the multi-stage decision model to obtain a complex production process decision plan, including: Using dynamic programming methods, starting from the final product, the optimal decision for each stage is reversely calculated and recorded in order to construct the optimal strategy. At the same time, the Monte Carlo method is used to simulate the production process to estimate the expected profit of various decisions, and the plan with the highest profit is selected as the decision plan for the complex production process.

5. The method according to claim 1, characterized in that: Considering the defective rate and sampling detection error in the sampling detection process, sensitivity analysis is performed on the decision-making scheme of complex production processes. Based on the analysis results, the sampling detection process is optimized using the Bayesian inference method, including: Considering the defective rate and sampling detection error in the sampling detection process, a sensitivity analysis is conducted on the decision-making plan of the complex production process, and the maximum profit of the decision-making plan of the complex production process under the confidence interval of the defective rate is calculated. It is determined that the defective rate detection error has a great influence on the calculation result of the maximum profit and has high sensitivity. The Bayesian inference method is used to continuously update the estimate of the defective rate based on historical detection data and current sampling detection data, and the optimal balance point between detection accuracy and detection cost is found to optimize the sampling detection process.

6. The method according to claim 5, characterized in that The confidence interval of defective rate is: ； in, represents the estimated defective rate obtained by sampling, Represents a normal distribution Quantile, n is the sample size, i.e. the number of spare parts types.

7. A product quality monitoring device based on multi-stage decision making, characterized in that: The device comprises: A sampling inspection module is used to calculate the minimum number of sampling inspections for sampling inspections of spare parts provided by suppliers under a given confidence level and nominal value according to a statistical hypothesis testing model; and to perform sampling inspections on spare parts using the minimum number of sampling inspections to obtain a sampling plan; A multi-stage decision model building module is used to use the sampling plan to determine whether to accept the spare parts of the supplier and produce electronic products based on the accepted spare parts, and to consider whether to test the spare parts, semi-finished products, and finished products at each production stage, and how to deal with finished products and unqualified products to establish a multi-stage decision model; A multi-stage decision model solving module is used to solve the multi-stage decision model using dynamic programming and Monte Carlo methods to obtain a complex production process decision plan; The product quality monitoring module is used to perform sensitivity analysis on complex production process decision plans by considering the defective rate and sampling detection error in the sampling detection process, optimize the sampling detection process using the Bayesian inference method based on the analysis results, and obtain an optimized sampling plan; determine an optimized complex production process decision plan based on the optimized sampling plan; and use the optimized complex production process decision plan to monitor product quality.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images