A product quality traceability analysis method, an electronic device, and a storage medium

Abstract

The application relates to the computer technical field, in particular to a product quality traceability analysis method, an electronic device and a storage medium, the method comprises the following steps: inputting a plurality of quality monitoring index data of a target product into a fault monitoring model, obtaining each fault problem corresponding to the target product and an initial score of each fault problem, and further obtaining a first fault cause analysis text corresponding to the target product according to the obtained result, respectively obtaining the correlation degree of each fault problem and a remaining fault problem, optimizing the initial score of each fault problem according to the correlation degree, inputting the optimization result into a given large language model, obtaining a second fault cause analysis text, and finally determining a fault traceability production link corresponding to the target product based on the first fault cause analysis text and the second fault cause analysis text; the application can improve the accuracy and reliability of fault cause prediction, and then improve the efficiency and accuracy of product quality traceability.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a product quality traceability analysis method, electronic device, and storage medium. Background Technology

[0002] Currently, traditional manufacturing processes typically involve multiple production stages, and different components may even involve different manufacturing companies. Due to the numerous production nodes and levels distributed throughout the production cycle, quality inspections are usually conducted at the end of the production cycle or before the product leaves the factory. When a fault occurs, it is difficult to trace the source in a timely manner, meaning it is impossible to pinpoint the production stage where the fault occurred. In practice, the experience of the staff is often used to guess which stage the fault occurred in. However, this is limited by the technical expertise of the inspection personnel. For example, staff at different stages may not fully understand the production mechanisms of other stages, leading to errors in traceability and affecting its efficiency. Furthermore, since the manufacturing process continues, more defective products are generated. Therefore, timely identification of the source of the fault is crucial for improving the company's production and product quality. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention provides a product quality traceability analysis method, electronic device, and storage medium, which can improve the accuracy and reliability of fault cause prediction, thereby improving the efficiency and accuracy of product quality traceability.

[0004] According to a first aspect of the present invention, a product quality traceability analysis method is provided, comprising the following steps:

[0005] Acquire several quality monitoring index data of the target product, and input these data into the trained fault monitoring model to obtain each fault problem and its initial score for the target product.

[0006] Each fault problem corresponding to the target product and its initial score are input into the pre-trained given large language model to obtain the first fault cause analysis text corresponding to the target product.

[0007] The correlation between each fault problem and the remaining fault problems is obtained separately; the remaining fault problems refer to other fault problems among all fault problems corresponding to the target product.

[0008] The initial score of each fault problem is optimized based on the correlation between each fault problem and the remaining fault problems to obtain the final score of each fault problem. The final score of each fault problem is then input into a given large language model to obtain the second fault cause analysis text.

[0009] Based on the first and second fault cause analysis texts, the probability of occurrence of each fault problem corresponding to the target product is obtained, and the production link for fault tracing corresponding to the target product is determined according to the probability of occurrence of each fault problem.

[0010] According to a second aspect of the present invention, a non-transitory computer-readable storage medium is provided, wherein at least one instruction or at least one program is stored therein, the at least one instruction or the at least one program being loaded and executed by a processor to implement the above-described product quality traceability analysis method.

[0011] According to a third aspect of the present invention, an electronic device is provided, including a processor and the aforementioned non-transitory computer-readable storage medium.

[0012] The present invention has at least the following beneficial effects:

[0013] The product quality traceability analysis method of this invention first inputs several quality monitoring index data of the target product into a fault monitoring model to obtain each fault problem and its initial score. The results are then input into a large language model to obtain the first fault cause analysis text for the target product. Next, the correlation degree between each fault problem and the remaining fault problems is obtained. Based on the correlation degree, the initial score of each fault problem is optimized, and the optimized final score is input into a given large language model to obtain the second fault cause analysis text. By introducing the correlation degree between different fault problems and improving the scores of correlated fault problems, the accuracy of fault cause prediction is improved, making the second fault cause analysis text more reliable. Finally, based on the first and second fault cause analysis texts, the production link corresponding to the fault traceability of the target product is determined. By comprehensively considering the two fault cause analysis texts, the accuracy and reliability of fault cause prediction can be improved, thereby increasing the efficiency and accuracy of product quality traceability. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 A flowchart of a product quality traceability analysis method provided in an embodiment of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] This embodiment provides a product quality traceability analysis method, such as... Figure 1 As shown, the method includes the following steps:

[0018] S100: Obtain several quality monitoring index data of the target product, and input the several quality monitoring index data into the trained fault monitoring model to obtain each fault problem and the initial score of each fault problem corresponding to the target product; it can be understood that the initial score of the fault problem can be the confidence level of the fault problem output by the fault monitoring model.

[0019] Specifically, the quality monitoring index data refers to the data obtained by testing the target product according to a set of preset quality monitoring indicators. For example, when the target product is rubber, the testing items include physical performance testing, chemical performance testing, electrical performance testing, and environmental performance testing. Among them, the quality monitoring indicators involved in physical performance testing include hardness, tensile strength, and tear strength; the quality monitoring indicators involved in chemical performance testing include chemical corrosion resistance, heat resistance, and oxidation resistance; the quality monitoring indicators involved in electrical performance testing include insulation resistance, dielectric constant, and dielectric loss; and the quality monitoring indicators involved in environmental performance testing include the content of toxic and harmful substances in the rubber. When the target product is rubber tires, in actual operation, it also includes appearance inspection, uniformity inspection, dynamic balance testing, and X-ray inspection of the rubber tires.

[0020] Furthermore, each fault corresponds to a quality monitoring indicator. For example, if the hardness is unqualified, a hardness fault is considered to exist; if the tensile strength is unqualified, a tensile fault is considered to exist.

[0021] Specifically, the fault monitoring model is a neural network prediction model, which is used to predict the fault problems of the target product based on several quality monitoring index data of the target product. Those skilled in the art know the specific training process of the neural network prediction model, so it will not be described in detail here.

[0022] S200: Input each fault problem corresponding to the target product and its initial score into the pre-trained given large language model to obtain the first fault cause analysis text corresponding to the target product. In specific implementation, the large language model is trained based on the scores of historical fault products and the fault cause analysis results to obtain the pre-trained given large language model.

[0023] The above-mentioned faults of the target product can be obtained based on several quality monitoring indicators. The first fault cause analysis text is obtained based on several faults and the score of each fault. The predicted causes of the faults are then obtained from the first fault cause analysis text, which is helpful for tracing the source of the faults in the production process.

[0024] S300, obtain the correlation degree between each fault problem and the remaining fault problems respectively; the remaining fault problems refer to other fault problems among all fault problems corresponding to the target product; it can be understood as: for any fault problem, obtain the correlation degree between each fault problem other than the stated fault problem and the stated fault problem.

[0025] Specifically, the correlation between each fault and the remaining faults is obtained through the following steps:

[0026] S301, Obtain the historical fault product dataset; the historical fault product dataset includes several historical fault products and several fault issues corresponding to each historical fault product.

[0027] S302, take any two fault problems as the first fault problem and the second fault problem respectively, and determine the historical fault products that have both the first fault problem and the second fault problem according to the several fault problems corresponding to each historical fault product, and take them all as the target fault products.

[0028] S303, when the number of target faulty products accounts for less than the proportion of all historical faulty products acquired, it is determined that the first fault problem and the second fault problem are unrelated; wherein, those skilled in the art set the preset proportion threshold according to actual needs, for example, 0.3, that is, when the number of target faulty products accounts for less than the proportion of all historical faulty products acquired, it is considered that the first fault problem and the second fault problem are problems that occur simultaneously by chance and are unrelated.

[0029] S304, when the proportion of the number of target faulty products to the total number of all historical faulty products is not less than a preset proportion threshold, the proportion of the number of target faulty products to the total number of all historical faulty products is determined as the degree of correlation between the first fault problem and the second fault problem.

[0030] As mentioned above, for several fault problems, considering that multiple fault problems may occur in the same production stage in the actual production process, such as hardness and tear strength, when the hardness is low, the tear strength may also be relatively low, so the correlation between the two is high. It is also possible that the semi-finished products in the previous production stage will affect the production effect of the next production stage, and their correlation is also high. Therefore, the correlation degree of different fault problems is introduced. Based on this, the score of fault problems is optimized, which is conducive to predicting more accurate and reliable fault causes.

[0031] S400 optimizes the initial score of each fault problem based on the correlation between each fault problem and the remaining fault problems to obtain the final score of each fault problem. The final score of each fault problem is then input into a given large language model to obtain the second fault cause analysis text. This can be understood as follows: both the first and second fault cause analysis texts include the determined final fault cause and the cause of each fault cause. For example, product faults caused by non-compliance of process parameters during manufacturing or product faults caused by overheating or vibration after prolonged operation of equipment can help staff analyze and trace the fault.

[0032] Furthermore, the S400 steps include the following steps:

[0033] S401, for any selected fault problem, obtain the degree of correlation between the selected fault problem and each of the remaining fault problems; this can be understood as: the remaining fault problems here are any fault problems corresponding to the target product other than the selected fault problem. Since the degree of correlation between each fault problem and other fault problems has been obtained in step S300, the degree of correlation between the selected fault problem and each of the remaining fault problems can be determined.

[0034] S402, based on the correlation between the selected fault problem and each of the remaining fault problems, calculate the final score corresponding to the selected fault problem, wherein the final score corresponding to the selected fault problem meets the following conditions:

[0035] S=S0×(1+∑ n j＝1 K j / η), where S is the final score corresponding to the selected fault problem, S0 is the initial score corresponding to the selected fault problem, and K j The degree of correlation between the selected fault problem and the j-th fault problem among the remaining fault problems, where n is the number of fault problems among the remaining fault problems, and η is a preset score narrowing reference threshold; this can be understood as, ∑ n j＝1 K jThe value of / η should be kept between 0 and 1 to prevent the final score from being excessively amplified.

[0036] As mentioned above, when two related fault problems occur simultaneously, it indicates that the prediction accuracy of these two fault problems is relatively high. Therefore, the scores of these two fault problems should be increased accordingly. By improving the scores of related fault problems, it is beneficial to improve the accuracy of fault cause prediction and make the obtained second fault cause analysis text more reliable.

[0037] S500, based on the first and second fault cause analysis texts, obtains the probability of occurrence of each fault problem corresponding to the target product, and determines the corresponding fault tracing production link for the target product according to the probability of occurrence of each fault problem. For example, for rubber tires, the production links include: adding rubber preparations to natural rubber, fiber cord production, bead and steel cord production, ply calendering, ply cutting, belt calendering, belt cutting, bead forming, inner liner lamination, and tire vulcanization, etc.

[0038] In a specific embodiment, the probability of occurrence of any fault problem corresponding to the target product is obtained through the following steps:

[0039] S501, based on the order of appearance of several fault issues in the first fault cause analysis text, assign a first accurate prediction probability to each fault issue in the first fault cause analysis text according to a preset probability allocation rule. In specific implementation, since fault issues with high prediction accuracy will be mentioned first in the text, a larger accurate prediction probability should be assigned to fault issues that appear earlier in the order.

[0040] Furthermore, the preset probability allocation rule means that as the fault problems appear in sequence, the allocated accurate prediction probabilities decrease, and the sum of the allocated accurate prediction probabilities is 1. In specific implementations, probabilities can also be allocated according to a pre-set probability allocation table. That is, the accurate prediction probability of each fault problem is pre-set for different numbers of fault problems. For example, when there are 3 fault problems, the probabilities are set to 0.5, 0.3, and 0.2 respectively; when there are 2 fault problems, the probabilities are set to 0.7 and 0.3 respectively.

[0041] S502, based on the order of occurrence of several fault issues in the second fault cause analysis text, assign a second accurate prediction probability to each fault issue in the second fault cause analysis text according to a preset probability allocation rule.

[0042] Specifically, the method for assigning the second accurate prediction probability to fault problems is the same as the method for assigning the first accurate prediction probability to fault problems, and will not be repeated here.

[0043] S503, based on the first accurate prediction probability of each fault problem in the first fault cause analysis text, the second accurate prediction probability of each fault problem in the second fault cause analysis text, and the preset text weights corresponding to the first and second fault cause analysis texts respectively, the occurrence probability P of any fault problem corresponding to the target product is calculated; the occurrence probability P of any fault problem corresponding to the target product meets the following condition:

[0044] P = P1 × w1 + P2 × w2, where P1 is the first accurate prediction probability for any fault problem, P2 is the second accurate prediction probability for any fault problem, w1 is the preset text weight corresponding to the first fault cause analysis text, and w2 is the preset text weight corresponding to the second fault cause analysis text.

[0045] Specifically, determining the production stage for tracing the source of the fault in the target product based on the probability of occurrence of each fault includes the following steps:

[0046] S510, when the probability of occurrence of any fault problem corresponding to the target product is greater than a preset probability threshold, the fault problem is determined as the target fault corresponding to the target product; in specific implementation, fault problems with a probability of occurrence of the corresponding fault problem not greater than the preset probability threshold can be used as alternative faults. When the production process corresponding to the target fault is found to be without problems after verification, the staff will then test the production process corresponding to the alternative fault.

[0047] S520, determine the production process corresponding to each target fault according to the preset relationship mapping table; the relationship mapping table includes several preset fault problems and the production process corresponding to each preset fault problem.

[0048] The above-mentioned comprehensive consideration of the first and second fault cause analysis texts can improve the predictive rationality of the probability of occurrence of each fault problem. On this basis, the mapped production links are made more reliable, thereby improving the accuracy of product quality traceability. Moreover, only a few quality inspection index data of the target product are required to obtain the predicted production links, thus improving the efficiency of product quality traceability.

[0049] Furthermore, the method also determines the probability of occurrence of each fault problem corresponding to the target product through the following steps:

[0050] S10: Input the initial score of each fault problem corresponding to the target product into the given large language model to obtain several third fault cause analysis texts.

[0051] S20: Based on the first fault cause analysis text, the second fault cause analysis text, and several third fault cause analysis texts, determine the probability of occurrence of each fault problem corresponding to the target product. The method of obtaining the probability of occurrence of the fault problem is the same as that of obtaining the probability of occurrence of the fault problem in step S500, and will not be repeated here.

[0052] As mentioned above, by increasing the number of fault cause analysis texts, it is equivalent to increasing the sample size of the detection data. Furthermore, the third fault cause analysis text corresponding to a fault problem may also involve other fault problems. Therefore, the probability of occurrence of each fault problem corresponding to the target product can be more reasonable and reliable, which is conducive to tracing the source of the fault problem of the target product in the production process.

[0053] In summary, the product quality traceability analysis method of this invention first inputs several quality monitoring index data of the target product into a fault monitoring model to obtain each fault problem and its initial score. The results are then input into a large language model to obtain the first fault cause analysis text for the target product. Next, the correlation degree between each fault problem and the remaining fault problems is obtained. Based on the correlation degree, the initial score of each fault problem is optimized, and the optimized final score is input into a given large language model to obtain the second fault cause analysis text. By introducing the correlation degree between different fault problems and improving the scores of correlated fault problems, the accuracy of fault cause prediction is improved, making the second fault cause analysis text more reliable. Finally, based on the first and second fault cause analysis texts, the production link corresponding to the fault traceability of the target product is determined. By comprehensively considering the two fault cause analysis texts, the accuracy and reliability of fault cause prediction can be improved, thereby increasing the efficiency and accuracy of product quality traceability.

[0054] Embodiments of the present invention also provide a non-transitory computer-readable storage medium that can be disposed in an electronic device to store at least one instruction or at least one program related to implementing a method in the method embodiments, wherein the at least one instruction or the at least one program is loaded and executed by the processor to implement the method provided in the above embodiments.

[0055] Embodiments of the present invention also provide an electronic device, including a processor and the aforementioned non-transitory computer-readable storage medium.

[0056] While specific embodiments of the invention have been described in detail by way of example, those skilled in the art should understand that the examples are for illustrative purposes only and not intended to limit the scope of the invention. It should also be understood that various modifications can be made to the embodiments without departing from the scope and spirit of the invention. The scope of the invention is defined by the appended claims.

Claims

1. A product quality traceability analysis method, characterized in that, The method includes the following steps: Acquire several quality monitoring index data of the target product, and input these data into the trained fault monitoring model to obtain each fault problem and its initial score for the target product. Each fault problem corresponding to the target product and its initial score are input into the trained given large language model to obtain the first fault cause analysis text corresponding to the target product. Each fault problem is correlated with the remaining fault problems; the remaining fault problems refer to other fault problems among all fault problems corresponding to the target product. The initial score of each fault problem is optimized based on the degree of correlation between each fault problem and the remaining fault problems to obtain the final score of each fault problem. The final score of each fault problem is then input into a given large language model to obtain the second fault cause analysis text. Based on the first and second fault cause analysis texts, the probability of occurrence of each fault problem corresponding to the target product is obtained, and the production link for fault tracing corresponding to the target product is determined according to the probability of occurrence of each fault problem.

2. The product quality traceability analysis method according to claim 1, characterized in that, The correlation between each fault problem and the remaining fault problems is determined through the following steps: Obtain a historical faulty product dataset; the historical faulty product dataset includes several historical faulty products and several fault issues corresponding to each historical faulty product; Any two fault problems are designated as the first fault problem and the second fault problem, respectively. Based on the several fault problems corresponding to each historical fault product, historical fault products that simultaneously have the first fault problem and the second fault problem are identified and all are designated as target fault products. When the proportion of the number of target faulty products to the total number of all historical faulty products is less than a preset proportion threshold, it is determined that the first fault problem and the second fault problem are unrelated. When the proportion of the number of target faulty products to the total number of all historical faulty products is not less than a preset proportion threshold, the proportion of the number of target faulty products to the total number of all historical faulty products is determined as the degree of correlation between the first fault problem and the second fault problem.

3. The product quality traceability analysis method according to claim 1, characterized in that, The optimization process for the initial score of each fault problem based on the correlation between each fault problem and the remaining fault problems includes the following steps: For any selected fault problem, obtain the degree of correlation between the selected fault problem and each of the remaining fault problems; Based on the correlation between the selected fault problem and each of the remaining fault problems, the final score corresponding to the selected fault problem is calculated, wherein the final score corresponding to the selected fault problem meets the following conditions: S=S0×(1+∑ n j＝1 K j / η), where S is the final score corresponding to the selected fault problem, S0 is the initial score corresponding to the selected fault problem, and K j The degree of correlation between the selected fault problem and the j-th fault problem among the remaining fault problems, where n is the number of fault problems among the remaining fault problems, and η is a preset score narrowing reference threshold.

4. The product quality traceability analysis method according to claim 1, characterized in that, The probability of any failure problem occurring for the target product can be obtained through the following steps: Based on the order of appearance of several fault issues in the first fault cause analysis text, a first accurate prediction probability is assigned to each fault issue in the first fault cause analysis text according to a preset probability allocation rule. Based on the order of appearance of several fault issues in the second fault cause analysis text, a second accurate prediction probability is assigned to each fault issue in the second fault cause analysis text according to a preset probability allocation rule. Based on the first accurate prediction probability of each fault problem in the first fault cause analysis text, the second accurate prediction probability of each fault problem in the second fault cause analysis text, and the preset text weights corresponding to the first and second fault cause analysis texts respectively, the occurrence probability P of any fault problem corresponding to the target product is calculated; the occurrence probability P of any fault problem corresponding to the target product meets the following condition: P = P1 × w1 + P2 × w2, where P1 is the first accurate prediction probability for any fault problem, P2 is the second accurate prediction probability for any fault problem, w1 is the preset text weight corresponding to the first fault cause analysis text, and w2 is the preset text weight corresponding to the second fault cause analysis text.

5. The product quality traceability analysis method according to claim 1, characterized in that, The process of determining the production stage corresponding to the fault in the target product based on the probability of occurrence of each fault includes the following steps: When the probability of any fault problem corresponding to the target product is greater than a preset probability threshold, the fault problem is identified as the target fault corresponding to the target product. The production process corresponding to each target fault is determined based on the preset relationship mapping table; The relationship mapping table includes several preset fault problems and the production process corresponding to each preset fault problem.

6. The product quality traceability analysis method according to claim 1, characterized in that, The method also determines the probability of occurrence of each fault problem corresponding to the target product through the following steps: The initial score for each fault problem corresponding to the target product is input into a given large language model to obtain several third fault cause analysis texts; Based on the first fault cause analysis text, the second fault cause analysis text, and several third fault cause analysis texts, the probability of occurrence of each fault problem corresponding to the target product is determined.

7. A non-transitory computer-readable storage medium, wherein the storage medium stores at least one instruction or at least one program segment, characterized in that, The at least one instruction or the at least one program segment is loaded and executed by the processor to implement the product quality traceability analysis method as described in any one of claims 1-6.

8. An electronic device, characterized in that, Includes a processor and the non-transitory computer-readable storage medium as described in claim 7.

Images