Product quality problem solution automatic generation method and device, medium and product

Abstract

The invention discloses a product quality problem solution automatic generation method and device, a medium and a product, and relates to the field of data processing. The method comprises the steps that a fault code and product type information of a current product quality problem are determined as a target fault code and target product type information, the fault code comprises multiple pieces of fault identification information, and the fault identification information comprises failure part information, failure position information and failure mode information; determining a first recommended solution set from a historical database based on the target fault code and the target product type information; when the data size of the first recommended solution set is insufficient, a second recommended solution set is determined from a historical database based on the fault identification information, and the data size of the second recommended solution set is larger than that of the first recommended solution set; and finally, generating a quality problem recommendation solution based on the second recommendation solution set. The problem that the formulating efficiency of the product problem solution is not high can be relieved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular, to a method, device, medium and product for automatically generating a solution to product quality problems. Background Art

[0002] In industrial manufacturing, it is often inevitable that product quality problems occur. When a product quality problem occurs, according to the traditional method, when formulating a solution to a quality problem, the person in charge of the quality problem needs to manually write a solution according to the responsible module of the quality problem, the severity level of the problem, and the specific problem content.

[0003] However, in many cases, the person in charge may not necessarily have experience in dealing with a specific problem, and there is no clear solution guidance for the handling method and the departments involved. In this case, the solution formulated has great limitations and inaccuracies, and there is a high probability of being rejected in the subsequent solution review. Moreover, there is often a situation of repeated rejection and repeated formulation of solutions, which greatly affects the efficiency of solving quality problems. Summary of the Invention

[0004] In view of the above technical problems and deficiencies, the object of the present invention is to provide a method, device, medium and product for automatically generating a solution to product quality problems, which can alleviate the problem of low efficiency in formulating solutions to product problems.

[0005] To achieve the above object, in a first aspect, the present invention provides a method for automatically generating a solution to product quality problems, including: determining the fault code of the current product quality problem and the product model information as the target fault code and the target product model information, where the fault code includes multiple fault identification information, and the fault identification information includes failed part information, failed location information, and failure mode information; based on the target fault code and the target product model information, determining a first recommended solution set from a historical database, where the historical database includes multiple problem-solving records, and the problem-solving records include the model information of the problem product, the fault code, and the corresponding problem-solving method; when the data volume of the problem-solving records in the first recommended solution set is less than a preset threshold, determining a second recommended solution set from the historical database based on the fault identification information corresponding to the target fault code, where the data volume of the problem-solving records in the second recommended solution set is greater than the data volume of the problem-solving records in the first recommended solution set; generating a recommended solution to the quality problem according to the target fault code, the target product model information, and the problem-solving records in the second recommended solution set.

[0006] Through precise matching of the target fault code and product model information, the present invention screens out a highly relevant first recommended solution set from the historical database, preferentially recommends the solutions that have been verified to be effective with high frequency under the same vehicle model, directly reuses mature experience, and reduces the cost of manual trial and error. Secondly, when the data in the first recommended set is insufficient, it extends to the solution records of the same type of problems across vehicle models and components based on the failure mode, and supplements the recommended data through multi-level sorting to ensure the diversity and applicability of the solutions. In addition, when generating the final recommended solution, the system dynamically adapts the general operation steps in the historical solutions to the specific parts and positions of the current fault, which not only retains the effectiveness of the historical experience but also avoids logical conflicts caused by parameter differences. This improves the generation efficiency of the solutions and the accuracy of problem-solving.

[0007] Optionally, in some embodiments, a quality problem recommended solution is generated according to the target fault code, the target product model information, and the problem-solving records in the second recommended solution set, including: determining the solution text replacement placeholders for each target problem-solving record according to the target fault code, the target product model information, and the multiple target problem-solving records in the second recommended solution set, where the solution text replacement placeholders include a product model placeholder, a failed part placeholder, a failure position placeholder, and a failure mode placeholder; generating multiple solution information according to the solution text replacement placeholders corresponding to each target problem-solving record, and the solution information is used to represent the text content of the quality problem recommended solution.

[0008] By adopting the above technical solution, through the definition of placeholders in the solution text (such as product model, failed part, etc.), the precise adaptation of the historical maintenance solution to the current fault scenario is realized. This method dynamically binds the general solution template with the specific fault parameters (part model, position), avoids manual repeated writing of similar cases, and improves the solution generation efficiency; at the same time, the placeholder mechanism ensures that the recommended content strictly matches the parsing result of the fault code, preventing misoperations caused by parameter misalignment. For example, replacing "replace [failed part]" with "replace the automatic transmission sealing ring" makes the suggestion directly executable.

[0009] Optionally, in some embodiments, after generating multiple solution instruction information according to the solution text replacement placeholders corresponding to each target problem-solving record, it further includes: displaying the multiple solution information to the user; and determining the final quality problem recommended solution from the multiple solution information in response to the user's solution confirmation instruction.

[0010] Adopting the above technical solution, a user confirmation link is introduced, and multiple solution options are presented for manual selection to balance automated recommendation and human experience judgment. This measure avoids potential biases in single-algorithm decision-making (such as the misrecommendation of frequently occurring but inapplicable solutions), and improves the reliability of solutions through human-machine collaboration. At the same time, user decision-making data can be used to reverse-optimize the recommendation model. For example, solutions that are frequently adopted will receive higher weights in subsequent rankings, forming a data-driven continuous improvement cycle.

[0011] Optionally, in some embodiments, after determining the recommended solution for the final quality problem from multiple solution information in response to the user's solution confirmation instruction, it further includes: filling the recommended solution for the final quality problem into the solution maintenance page of the current product quality problem to obtain a problem solution recommendation page; in response to the user's editing operation on the problem solution recommendation page, adjusting the content on the problem solution recommendation page.

[0012] Adopting the above technical solution, the final solution is embedded in the standardized maintenance page and editing is supported, realizing visual management and flexible adjustment of the solution. The editing function allows maintenance personnel to fine-tune the solution according to on-site actual conditions (such as spare part inventory, tool limitations) (such as modifying torque parameters or supplementing inspection steps), enhancing the implementation adaptability of the solution; the maintenance page serves as the only information source, ensuring version control and traceability of revision records, and avoiding execution chaos caused by multiple versions of the solution.

[0013] Optionally, in some embodiments, after adjusting the content on the problem solution recommendation page in response to the user's editing operation on the problem solution recommendation page, it further includes: in response to the user's approval operation on the problem solution recommendation page, generating a new problem solution record based on the approved problem solution recommendation page; storing the new problem solution record in the historical database.

[0014] Adopting the above technical solution, the problem solution records passed through the review are saved back to the historical database to build a self-enhancing knowledge accumulation system. The new records not only contain the original fault data but also incorporate user-corrected content (such as optimized operation steps), enabling subsequent recommendations to absorb practical experience. Especially for rare faults or new equipment, the system can quickly expand the coverage of solutions and reduce the impact of the cold start problem on the recommendation quality.

[0015] Optionally, in some embodiments, based on the target fault code and the target product model information, a first recommended solution set is determined from the historical database, including: querying all historical problem-solving records containing the same fault code from the historical database according to the target fault code and the target product model information to obtain an initial data set; screening the initial data set based on the target product model information, retaining the historical records matching the target product model information to obtain a same-model data set; sorting the same-model data set according to the total number of times the fault codes in each problem-solving record appear in the historical database to obtain a sorting result; selecting the first N problem-solving records in the sorting result to obtain the first recommended solution set.

[0016] By adopting the above technical solution, the generation logic of the first recommended set is clarified, and the solutions for the same model and high-frequency faults are preferentially screened. Through the rigid conditions of "the same product model + complete match of the fault code", it is ensured that the initial recommendation highly conforms to the current device characteristics (such as the exclusive sealing design of a certain vehicle model's transmission), reducing the compatibility risk of cross-model adaptation; sorting by historical frequency implies the assumption of "high-frequency fault = mature solution", improving the usability of the initial recommendation and shortening the user's decision-making time.

[0017] Optionally, in some embodiments, based on the fault identification information corresponding to the target fault code, a second recommended solution set is determined from the historical database, including: retrieving problem-solving records containing the same failure mode from the historical database based on the target failure mode information in the target fault code to obtain an extended data set; screening out the problem-solving records matching the target product model information from the extended data set as the first-priority data, and the un-screened problem-solving records as the second-priority data; sorting the second-priority data according to the historical occurrence frequency of the target failure mode information to obtain the sorted second-priority data; intercepting the corresponding number of problem-solving records from the first-priority data and the sorted second-priority data according to the difference between the preset threshold and the current data volume of the first recommended solution set to obtain the second recommended solution set.

[0018] By adopting the above technical solution, the recommendation scope is expanded when the data is insufficient, and the coverage problem of long-tail faults is solved through failure mode matching and cross-model priority sorting. For example, when there are insufficient cases of oil leakage in a certain new model of vehicle, introduce the same-mode fault solutions of other vehicle models (such as oil leakage in a truck fuel tank), and then recommend them according to the model relevance and the universality of the failure mode, which not only ensures the relevance of the solutions (same model first), but also uses cross-scenario experience (high-frequency failure mode solutions) to make up for the data gap and enhance the robustness of the system under sparse data.

[0019] In a second aspect, an embodiment of the present invention provides an electronic device, comprising: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the electronic device to execute the method described in the first aspect or the second aspect, and any possible implementation method of the first aspect or the second aspect.

[0020] In a third aspect, the present invention provides a computer-readable storage medium comprising instructions, which, when executed on the electronic device, causes the electronic device to execute the method described in the first aspect or the second aspect, and any possible implementation of the first aspect or the second aspect.

[0021] In a fourth aspect, the present invention provides a computer program product comprising instructions, which, when the computer program product is run on the electronic device, enables the electronic device to execute the method described in the first aspect or the second aspect, and any possible implementation of the first aspect or the second aspect.

[0022] It is understandable that the electronic device provided in the second aspect, the storage medium provided in the third aspect, and the computer program product provided in the fourth aspect are all used to execute the method provided in the present invention. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, which will not be repeated here.

[0023] One or more technical solutions provided by the present invention have at least the following technical effects or advantages: 1. Effectively solve the problem of data sparsity and improve the coverage and practicality of the solution. Through the hierarchical progressive recommendation strategy (first accurately match the full fault solution of the same model, and then expand to the similar failure mode solution across models), the system can flexibly deal with the problem of insufficient data in new products or rare fault scenarios. For example, when a rare oil leakage occurs in a new model passenger car, the system can call the general processing logic of truck engine oil leakage, combined with placeholder technology to replace the difference parameters (such as part model, installation location), which not only breaks through the limitations of historical data, but also avoids the recommendation failure caused by no matching results, so that the coverage of the solution can be improved.

[0024] 2. Significantly improve the efficiency and automation level of quality management. The technical solution connects the fault feature analysis, intelligent matching of historical data, dynamic template generation and other links to form an end-to-end automated solution production chain. The process of manually searching the case library and writing adaptation solutions in the traditional mode has been compressed into automatic output in minutes, and the efficiency has been improved by more than 80%. For example, the system can parse the failed part code in the fault code in real time, automatically associate the engine model and fuel system parameters, and generate a maintenance plan including part replacement steps and torque standards, which greatly shortens the company's fault response cycle.

[0025] 3. Enhance the technical logic consistency and transferability of cross-scenario solutions. Through the structured placeholder technology and the failure mode similarity judgment rule, when the system expands cross-model solutions, it can accurately retain the core maintenance logic (such as the seal installation process, pressure test method), and only replace the model-related variables (such as part size, interface type). This "logic solidification, parameter variable" design not only avoids the risk of technical misuse in manual adaptation, but also enables the large-scale reuse of historical experience. For example, for the circuit short-circuit problems of different vehicle models, the solutions output by the system all include standard steps such as insulation detection and wire harness replacement, and only adjust the wire diameter specification and fuse parameters according to the model to ensure the stability of maintenance quality and reduce the cross-model maintenance cost of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The drawings here are incorporated into the specification and form a part of this specification, showing embodiments consistent with the present invention, and are used together with the specification to explain the principles of the present invention. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts. In the drawings: Figure 1 is a flowchart of a method for automatically generating a solution to product quality problems according to an embodiment of the present invention; Figure 2 is a schematic diagram of replacing placeholders in a solution text according to an embodiment of the present invention; Figure 3 is a schematic diagram of the architecture of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] The terms used in the following embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. As used in the specification of the present invention, the singular forms "a", "an", "the above", "the", and "this" are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in the present invention refers to any or all possible combinations including one or more of the listed items.

[0028] Hereinafter, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of the present invention, unless otherwise specified, the meaning of "a plurality" is two or more.

[0029] It should also be noted that, unless otherwise clearly specified and limited, in the embodiments of the present invention, terms such as "set" and "connect" should be understood in a broad sense. For example, "connect" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be the communication inside two components; it can be a wired communication connection or a wireless communication connection. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The embodiments of the present invention will be specifically described below.

[0030] In the traditional method, the process of the quality problem responsible person formulating a solution usually follows the following manually dominated path: When a quality problem occurs in a product, the responsible person first obtains a description of the problem phenomenon through on-site feedback or inspection reports (such as "equipment oil leakage" "part fracture"), and then, based on personal experience or simple classification rules, manually classifies the problem into the responsible module (such as "transmission system" "sealing component"), and divides the severity level of the problem based on subjective judgment (such as "urgent" "general"). Then, the responsible person needs to search for the handling records of similar problems one by one in the scattered historical work order records, Excel sheets or paper documents. For example, the responsible person manually screens potential reference cases in the unstructured text description through keywords (such as "oil leakage" "sealing"). If some relevant records are found, the responsible person needs to analyze the solution measures in each case (such as "replace the seal ring model" "adjust the assembly torque") one by one, and combine the information such as the equipment model and supplier batch of the current problem, and manually modify the wording to adapt to the new scenario. If there is no direct reference case, the responsible person may obtain scattered suggestions through cross-departmental meetings or oral inquiries of experienced colleagues, and then integrate these fragmented information into a preliminary solution. Due to the lack of unified structured storage of historical data (such as not classified according to dimensions such as faulty components, modes, and environments), and the solution descriptions are mostly natural language texts, it is difficult for the responsible person to quickly locate effective information, and the final solution often has problems such as incomplete measures (such as not associating specific part numbers), incomplete coverage of responsible departments (such as missing the quality inspection link), or logical contradictions (such as repair steps conflicting with equipment parameters), resulting in frequent rejections during the review stage, forcing the responsible person to repeatedly revise or even re-formulate the solution.

[0031] This process highly depends on individual experience and manual cross-verification, and cannot systematically reuse implicit knowledge, becoming a double bottleneck for efficiency and accuracy.

[0032] Therefore, the embodiments of the present invention provide an automatic generation method for product quality problem solutions, which significantly improves the generation efficiency and accuracy of quality problem solutions by introducing a structured data processing and intelligent retrieval mechanism, and effectively solves the core defects such as strong experience dependence, low utilization rate of historical data, and poor solution adaptability in traditional manual methods.

[0033] First, based on the structured parsing of fault codes (failed parts, failure locations, failure modes), the fault characteristics originally scattered in natural language descriptions are transformed into multi-dimensional tags recognizable by machines, completely eliminating the information misalignment caused by semantic ambiguity or expression differences in traditional methods. The system accurately filters out solutions for the same model and the same fault from the historical database by automatically matching the target fault code with the product model, forming the first recommendation set, ensuring a high degree of consistency between the recommended results and the current problem in terms of equipment model and fault mechanism, and avoiding omissions caused by cognitive limitations or keyword selection biases in manual retrieval.

[0034] Secondly, when the data volume of the first recommendation set is insufficient, the system automatically expands the retrieval across product models with the failure mode as the common feature, mines solutions with the same failure mechanism but different application scenarios from historical data, breaks through the bottleneck of scarce data for the same model in traditional methods through a dynamic supplementation mechanism, and realizes the multi-dimensional reuse of implicit experience. For example, for the first-time seal failure problem of a new model product, the system can automatically associate historical solutions caused by similar seal structure defects in other models, providing a cross-platform reference basis for the person in charge. In addition, by presetting a threshold to control the scale of the recommendation set, the system filters redundant information while ensuring the diversity of solutions, solving the problems of information overload or data fragmentation in manual processing.

[0035] The finally generated recommended solution deeply integrates the current fault characteristics with the historical solution logic, and outputs a standardized solution through automatic template replacement (such as adapting product model parameters, failed part specifications), significantly reducing logical contradictions or key parameter errors in manual writing. More importantly, the system forms a closed loop of "data accumulation - intelligent recommendation - review feedback", continuously optimizing the recommendation rules and data quality, while the isolated and static experience library in traditional methods cannot achieve such self-iterative upgrading.

[0036] Through the above technical integration, the embodiment of the present invention shortens the solution formulation cycle, improves the review passing rate, and at the same time promotes the transformation of enterprise quality knowledge from individual experience to systematic assets.

[0037] The following combines Figure 1 to specifically illustrate the method for automatically generating a solution to the product quality problem of this embodiment, which can be applied to a system for automatically generating a solution to the product quality problem (hereinafter referred to as the system). The method includes the following steps: Step 101, determine the fault code and product model information of the current product quality problem as the target fault code and target product model information.

[0038] Among them, the fault code includes multiple fault identification information, and the fault identification information includes failed part information, failed location information, and failure mode information. The failed part information refers to the specific part name or number that has failed in the quality problem (such as bearings, seals); the failed location information describes the installation location of the part in the equipment or product (such as the 3rd section of the drive shaft, the housing interface); the failure mode information characterizes the physical or functional abnormal type of the fault occurrence (such as oil leakage, fracture, poor contact).

[0039] For example, assume that the current product quality problem is oil leakage in flatbed truck No. 6, and the fault code is 171T311101011000G02M15, where 171T311101011000 represents the failed part (automatic transmission with retarder assembly), G02 represents the failed location (transmission assembly), and M15 is the failure mode (oil leakage).

[0040] Then, the target product model information is flatbed truck No. 6, the target fault code is 171T311101011000G02M15, the corresponding failed part information is 171T311101011000, the failed location information is G02, and the failure mode information is M15.

[0041] Step 102, based on the target fault code and the target product model information, determine the first recommended solution set from the historical database.

[0042] Among them, the historical database includes multiple problem-solving records. The problem-solving records include the model information of the problem product, the fault code, and the corresponding problem-solving methods, such as "replace the seal ring of the transmission assembly and perform a pressure test to verify the sealing performance".

[0043] Specifically, the system uses the complete fault code as the exact matching condition to screen all problem-solving records with the same fault code from the historical database (hereinafter referred to as records for short). During this process, the system preferentially retains the records that exactly match the target product model information (such as flatbed truck No. 6) and counts the occurrence frequency of this fault code under the same platform model. If there are multiple records with the same fault code for the same model (for example, flatbed truck No. 7 + fault code 171T311101011000G02M15 appears 10 times), they are sorted in descending order of the occurrence frequency, and the first 10 are intercepted as the first recommended solution set; if there are the same fault code data for different models (such as flatbed truck No. 8 appears 8 times), they are sorted according to the model priority (target model > other models) and frequency and supplemented to a maximum of 10. For example, the 10 high-frequency records of flatbed truck No. 7 will be fully included in the recommendation, while the 8 records of flatbed truck No. 8 may be partially or completely excluded due to the low frequency and non-matching model.

[0044] The specific implementation process of this step is as follows: First, perform exact match screening. Using the complete fault code as a condition, query all records with the same fault code in the historical database. For example, if the target fault code is 171T311101011000G02M15 (oil leakage in the automatic transmission with retarder assembly), only extract the records with exactly the same fault code (such as the data with the same fault code in Flatbed Truck No. 7 and Flatbed Truck No. 8). Then, conduct vehicle model priority classification, and preferentially retain the records that are exactly the same as the target product model (Flatbed Truck No. 6) in the matching results.

[0045] If there are records with the same fault code in the target vehicle model (target product model), they will be preferentially displayed; if there are no records in the target vehicle model, the matching data of other vehicle models will be included. For example, if the target vehicle model is Flatbed Truck No. 6 but there is no matching data, and there are records with the same fault code in Flatbed Truck No. 7 and No. 8 (appearing 10 times and 8 times respectively), then proceed to the next screening step.

[0046] Subsequently, perform frequency sorting and truncation. Arrange the historical records of the same fault code in descending order according to the total frequency of the problem occurrence under the same vehicle model. For example, the problem-solving records of Flatbed Truck No. 7 + fault code 171T311101011000G02M15 appear 10 times, and the records of Flatbed Truck No. 8 + the same fault code appear 8 times. Then, the 10 records of Flatbed Truck No. 7 have a higher priority. The system will sort the records from high to low according to the number of records and truncate the top 10 as the first recommended solution set.

[0047] If the frequency records of the same vehicle model are sufficient (such as 10 records of Flatbed Truck No. 7), they will be directly included in the recommended set in full quantity, and the data of other vehicle models (such as 8 records of Flatbed Truck No. 8) will be excluded due to lower frequency and non-matching vehicle models; if the data of the same vehicle model is insufficient (such as only 3 records), then supplement the records with higher frequencies in other vehicle models, but the total number does not exceed 10.

[0048] When the data volume of the problem-solving records in the first recommended solution set is less than the preset threshold (for example, 10 records), the first recommended solution set will be recommended to the user for reference.

[0049] Among them, the first recommended solution set is generated based on the principles of giving priority to the target vehicle model and high frequencies. For example, if there is no data for the target vehicle model (Flatbed Truck No. 6), and there are 10 records and 8 records for Flatbed Truck No. 7 and No. 8 respectively, then the first recommended solution set will include all 10 high-frequency records of Flatbed Truck No. 7, and the data of Flatbed Truck No. 8 will not be included due to insufficient frequency and vehicle model priority.

[0050] In this embodiment, the historical database is constructed as follows: First, a large number of records of solving past product quality problems are collected. These records should include the model information of the problem products, fault codes, and the corresponding problem-solving methods. Then, the database structure is designed to ensure that the above information can be stored and support accurate matching queries by fault codes and product models. Next, the collected data is entered into the database, and indexes are established to improve the query efficiency. At the same time, the database needs to be updated regularly, adding new problem-solving records, and performing data cleaning and verification to ensure the accuracy and integrity of the data. In addition, data classification and tagging can also be considered to expand the query scope through fuzzy matching or partial matching when the data volume is insufficient. Finally, a data access and management mechanism is established to ensure the security and availability of the database.

[0051] Step 103, when the data volume of the problem-solving records in the first recommended solution set is less than the preset threshold, based on the fault identification information corresponding to the target fault code, a second recommended solution set is determined from the historical database.

[0052] Among them, the data volume of the problem-solving records in the second recommended solution set is greater than that in the first recommended solution set. The data volume of the second recommended solution set is larger because the screening condition of the first recommended solution set is a strict match based on the complete fault code + the same model, resulting in insufficient data volume after screening. Therefore, the screening condition is relaxed to only match the failure mode + cross-model compatibility, so as to include more historical records (such as solutions for different vehicle models, different components but the same failure mode), and the quantity is supplemented by frequency sorting, making the data volume of the problem-solving records in the second recommended solution set more sufficient.

[0053] When the data volume of the first recommended set is insufficient, the system extracts the failure mode in the target fault code (such as M15 corresponding to "oil leakage"), relaxes the screening condition, and retrieves all records with the same failure mode from the historical database (without restricting the platform vehicle model and complete fault code). For example, if the target failure mode is "oil leakage", records including different vehicle models (flatbed vehicle No. 6, No. 7, No. 8) or different failure parts / positions but the same failure mode of "oil leakage" are all included in the candidates. Subsequently, the system performs multi-level sorting according to the vehicle model matching degree and the frequency of the failure mode: first, retain the records that are the same as the target vehicle model (flatbed vehicle No. 6), and then sort the records of other vehicle models in descending order of the frequency of the failure mode. Finally, the data is supplemented to the preset quantity (such as 10 pieces) according to the sorting result, for example, supplemented until the total quantity reaches the standard. If it is still insufficient, the supplement is stopped, and only the existing data is used as the recommendation.

[0054] In some embodiments, the specific implementation process of this step is as follows: When the amount of data in the first recommendation set is less than the preset threshold (such as less than 10 items), the system expands and filters the second recommendation set from the historical database based on the failure mode identifier in the target fault code (such as M15 corresponding to "oil leakage").

[0055] First, extract the failure mode, parse the target fault code, extract the last 3 digits of the failure mode code (such as M15) and map it to the specific problem type (such as oil leakage). For example, the failure mode of the target fault code 171T311101011000G02M15 is "oil leakage".

[0056] Then, fuzzy matching screening is performed, with failure mode as the condition, and the screening rules are relaxed to retrieve all records with the same failure mode in the historical database, without limiting the platform model and complete fault code. For example, if the failure mode is "oil leakage", records with different models (flatbed car No. 6, No. 7, No. 8), different failed parts (such as chassis assembly, transmission assembly) or different failure locations (such as fuel tank, seal) but the same failure mode of "oil leakage" are included as candidates.

[0057] Next, multi-level priority sorting is performed. The first priority is the records of the same model, and the records that are the same as the target model (Flatbed Car No. 6) are retained first. For example, if there are records of oil leakage in other parts of Flatbed Car No. 6 in the history library (such as engine oil leakage), even if the fault code is different (due to failure location or part difference), it will be extracted first. The second priority is the descending order of failure mode frequency. For records of non-target models, they are sorted from high to low according to the frequency of occurrence of the failure mode in the history library. For example, Flatbed Car No. 8 has 3 oil leakage problems and Flatbed Car No. 7 has 2 oil leakage problems. Then the 3 records of Flatbed Car No. 8 are ranked higher. Finally, data supplementation and interception are performed. After sorting according to the above priority rules, the system intercepts data from high to low until the preset number (such as 10) is supplemented.

[0058] For example, if the first recommendation set has only 3 data items, the second recommendation set needs to be supplemented with 7 items: Assume that the following data exists in the history library: ① 1 record of oil leakage of flatbed car No. 6 (car model matches, highest priority); ③ 3 records of oil leakage of flatbed car No. 8 (frequency 3 times, second priority); ② 2 records of oil leakage of flatbed car No. 7 (frequency 2 times, lowest priority).

[0059] The system includes them in the order of sorting ①>③>②. When making up to 7 items, it will include 1 item of ①, 3 items of ③, and 2 items of ②, and the remaining 1 vacancy will not be filled. Finally, the second recommended set contains solutions for the same type of failure modes across vehicle models and components. For example, it may include records such as "Engine oil leakage of flatbed truck No. 6: Replace the cylinder gasket" and "Fuel tank oil leakage of flatbed truck No. 8: Tighten the interface". Although these solutions are not exactly the same as the target fault code, they can be used as effective supplements because the failure modes are the same and they have been prioritized. If it is still less than 10 items after making up, the system only outputs the current data and does not expand further.

[0060] Step 104, generate a recommended solution for the quality problem according to the target fault code, the target product model information, and the problem-solving records in the second recommended solution set.

[0061] Specifically, first parse the target fault code to clarify the name of the failed part (such as the automatic transmission with retarder assembly), the name of the failure location (such as the transmission assembly), the name of the failure mode (such as oil leakage), and the target product model (such as flatbed truck No. 6).

[0062] Subsequently, extract the historical problem-solving records one by one from the second recommended solution set, and adapt the content of the solution text for each record: identify the general description segments related to the failed part, failure location, and failure mode in the solution through text recognition technology (such as "Replace [component]" and "Detect [problem type]"), and replace the general terms in these segments (such as "[component]" and "[problem type]") with the specific names of the current failure (such as "transmission assembly" and "oil leakage"). For example, if the solution in a certain historical record is "Replace the oil tank and perform an oil leakage detection", the system identifies "oil tank" as a similar attribute of the failure location name (belonging to the "sealing component" category) through a predefined component-problem type mapping table, and then replaces it with the current failure location name "transmission assembly" to form an adapted solution "Replace the transmission assembly and perform an oil leakage detection".

[0063] Finally, output all the adapted solutions in the original sorting of the second recommended set (the same vehicle model first, the failure mode frequency in descending order), and filter out the logically conflicting content (such as the original solution contains operation steps irrelevant to the current failure), ensuring that the recommended content strictly matches the failure mode (oil leakage) and component type (transmission-related) of the target failure, and finally generate a directly applicable recommended solution for the quality problem.

[0064] This embodiment adopts the above method steps, and improves the efficiency and accuracy of solution generation by intelligently matching the key features of historical data and current quality problems. First, by accurately matching the target fault code (including the failed part, failed location, and failure mode) and product model information, a highly relevant first recommended solution set is screened out from the historical database. Solutions that have been verified effectively with high frequency under the same vehicle model are preferentially recommended, directly reusing mature experience and reducing the cost of manual trial and error. Second, when the data in the first recommended set is insufficient, based on the failure mode (such as oil leakage), it is extended to the solution records of the same type of problems across vehicle models and components (such as engine oil leakage in flatbed vehicle No. 6 and fuel tank oil leakage in flatbed vehicle No. 8). The recommended data is supplemented through multi-level sorting (prioritizing the same vehicle model and descending order of failure mode frequency) to ensure the diversity and applicability of the solutions. In addition, when the system generates the final recommended solution, the general operation steps in the historical solutions (such as "replace the sealing component" and "perform oil leakage detection") are dynamically adapted to the specific parts and locations of the current fault (such as the automatic transmission with retarder assembly), which not only retains the effectiveness of historical experience but also avoids logical conflicts caused by parameter differences.

[0065] This embodiment can not only effectively shorten the time for formulating problem solutions, but also improve the first-pass rate of solution review through data-driven priority rules. At the same time, the accumulated solutions continuously feed back to the historical database, forming a closed-loop optimization, building a standardized and reusable quality problem knowledge base for the enterprise, and long-term reducing the quality control cost and promoting product reliability improvement.

[0066] In some embodiments, the method of this embodiment specifically includes the following steps: S201, determine the fault code of the current product quality problem and the product model information as the target fault code and the target product model information.

[0067] This step can refer to the description of the foregoing embodiment and will not be elaborated here.

[0068] S202, according to the target fault code and the target product model information, query all historical problem solution records containing the same fault code from the historical database to obtain the initial data set.

[0069] Specifically, the system performs precise query operations in the historical database based on the target fault code (such as 171T311101011000G02M15) and the target product model information (such as flatbed car No. 6), extracts all historical problem solving records containing the same complete fault code, and forms an initial data set. For example, if the target fault code is "171T311101011000G02M15" (automatic transmission with retarder assembly oil leakage), only records with completely consistent fault code fields are retained (such as records with the same fault code in flatbed car No. 7 and flatbed car No. 8), and excludes data with other codes that do not match.

[0070] This step is achieved through precise matching of the database (such as SQL WHERE fault code = target value), ensuring that the initial data set is completely consistent with the technical characteristics of the current fault, providing a highly relevant data foundation for subsequent recommendations.

[0071] S203, screening the initial data set based on the target product model information, retaining the historical records matching the target product model information, and obtaining a data set of the same model.

[0072] Based on the initial data set, the system further selects historical records that fully match the target product model information (such as flatbed car No. 6) to generate a data set of the same model. For example, if the initial data set contains records of flatbed cars No. 6, No. 7, and No. 8, only the record of flatbed car No. 6 is retained, and the data of other models are eliminated.

[0073] This step is implemented through field matching (such as WHERE model information = target model), with the purpose of giving priority to reusing solutions under the same model, because the design, component configuration and problem mode of products of the same model are more similar, and the adaptability and success rate of the solution are higher. If the data of the same model is empty (such as no matching record for flatbed car No. 6), skip this step and directly enter the cross-model data processing.

[0074] S204, sorting the fault codes of each problem solving record in the same model data set according to the total number of occurrences in the historical database to obtain a sorting result.

[0075] For each record in the same model data set, the system can count the total number of occurrences of the corresponding fault code in the historical database and sort them from high to low. For example, if the record corresponding to the fault code 171T311101011000G02M15 of flatbed vehicle No. 6 has appeared 8 times in the historical database, and another record of the same model has appeared 5 times, then the record with 8 occurrences will be ranked higher. This sorting rule is based on the assumption that "high frequency = high reliability", and it is believed that solutions that appear repeatedly are more likely to be effective after multiple verifications.

[0076] In terms of technical implementation, sorting is completed through a pre-computed frequency statistics table or dynamic aggregation queries (such as COUNT and ORDER BY in SQL) to ensure real-time and accurate results.

[0077] S205, Select the top N problem-solving records from the sorting results to obtain the first recommended solution set.

[0078] Specifically, the system intercepts the top N (e.g., N = 10) problem-solving records according to the above sorting results to generate the first recommended solution set. For example, if the sorted same-type data set contains 15 records, then select the top 10 with the highest frequency; if the data volume is less than N (e.g., only 5), then all are included in the first recommended solution set to ensure that the data volume in the first recommended solution set is sufficient as much as possible.

[0079] This step is implemented through paging queries (such as LIMIT N in SQL). The finally output recommended set satisfies both the same-type matching and high-frequency priority principles, taking into account the adaptability and reliability of the solutions, and providing high-quality candidate data for subsequent manual review or automatic processing.

[0080] When the data volume of the problem-solving records in the first recommended solution set is greater than or equal to the preset threshold, the system pushes the first recommended solution set to the user for the user to refer to and use.

[0081] When the data volume of the problem-solving records in the first recommended solution set is less than the preset threshold, go to step 206.

[0082] S206, Based on the target failure mode information in the target fault code, retrieve problem-solving records containing the same failure mode from the historical database to obtain an extended data set.

[0083] Specifically, the system extracts the failure mode information in the target fault code (e.g., M15 corresponds to "oil leakage") and retrieves all records containing the same failure mode (such as "oil leakage") in the historical database to form an extended data set.

[0084] This step is achieved through fuzzy matching: ignoring the failed parts and location information in the fault code (such as the first 19 characters), and only using the failure mode code (the last 3 characters) as the screening condition. For example, if the target failure mode is oil leakage (M15), the extended data set contains all records with the same failure mode of oil leakage for different vehicle models (such as flatbed vehicle No. 6, No. 7, No. 8) and different failed parts (such as transmission, fuel tank), even if their complete fault codes are different (such as 171T311101012000A03M15).

[0085] S207, filtering out problem solving records that match the target product model information from the extended data set as first priority data, and problem solving records that are not filtered out as second priority data.

[0086] Specifically, from the extended data set, the system prioritizes filtering out records that fully match the target product model information (such as flatbed vehicle No. 6) as first-priority data (such as records of oil leakage in the engine of flatbed vehicle No. 6), and the remaining records are classified as second-priority data (such as oil leakage in the fuel tank of flatbed vehicle No. 7 and oil leakage in the sealing ring of flatbed vehicle No. 8).

[0087] This step is achieved through field matching (such as WHERE model information = target model) to ensure that solutions with the same model are recommended first because their component configuration and design are closer to the current problem and have higher adaptability. If the first priority data is empty (such as flatbed car No. 6 has no oil leakage record), it will directly enter the second priority processing.

[0088] S208 , sorting the second priority data according to the historical occurrence frequency of the target failure mode information to obtain sorted second priority data.

[0089] For the second priority data (records of non-target models), the system can count the total frequency of each record's failure mode in the historical database and sort them from high to low by frequency. For example, if the oil leakage problem of flatbed car No. 8 occurs 3 times and that of flatbed car No. 7 occurs 2 times, then the 3 records of flatbed car No. 8 are ranked higher.

[0090] This ranking is based on the principle of "high frequency = high universality", and believes that solutions that appear frequently across models may have wider effectiveness. In terms of technical implementation, the ranking is completed through pre-calculated frequency statistics tables or dynamic aggregation queries (such as SQL COUNT and ORDER BY).

[0091] S209 , according to the difference between the preset threshold and the current data volume of the first recommended solution set, extract a corresponding number of problem solving records from the first priority data and the sorted second priority data to obtain a second recommended solution set.

[0092] Specifically, the system intercepts data in order of priority based on the difference between the preset threshold (e.g., the total number of recommendations is 10) and the current amount of data in the first recommendation set (e.g., the first recommendation set already has 3 items, and 7 items need to be supplemented): first include all the first priority data (e.g., 1 oil leakage record of flatbed car No. 6), and if it is still insufficient, supplement the remaining number in descending order of frequency from the sorted second priority data (e.g., take 3 oil leakage records of flatbed car No. 8 and 2 oil leakage records of flatbed car No. 7, and supplement a total of 6 items, and the total recommendation set is 3+1+6=10 at this time). If the amount of data is still insufficient, only the acquired records are retained.

[0093] This step is implemented through conditional judgments (such as IF…ELSE) and paging queries (such as LIMIT), ultimately ensuring that the data volume of the second recommended set is greater than that of the first recommended set, and the recommended content meets both vehicle model adaptability and failure mode experience reusability.

[0094] S210. According to the target fault code, target product model information, and multiple target problem-solving records in the second recommended solution set, determine the solution text replacement placeholders for each target problem-solving record.

[0095] Among them, the solution text replacement placeholders include product model placeholders, failed part placeholders, failure location placeholders, and failure mode placeholders.

[0096] The system performs placeholder replacement on each historical solution text in the second recommended set according to the target fault code and target product model information. First, parse the fault code to extract the specific names of the failed part, location, and mode (for example, the failed part is "automatic transmission with retarder assembly"), and at the same time determine the current product model (such as flatbed truck No. 6); then locate the general placeholder fields (such as "[failure location]", "[failure mode]") in the historical solution text and replace them with the corresponding information of the target fault (such as "transmission assembly", "oil leakage"). For example, convert "Replace the seal ring of [failure location]" to "Replace the seal ring of the transmission assembly". This process is achieved through predefined replacement rules and string operations to ensure that the solution content matches the current problem precisely.

[0097] Exemplarily, as Figure 2 shown, the solution text is "The oil tank of the chassis assembly of flatbed truck No. 8 leaks oil. It is recommended to replace the oil tank and perform sealing adjustment". The system replaces the placeholders according to the following rules: Placeholder A (product model placeholder): Replace the platform vehicle model "flatbed truck No. 8" with placeholder A; Placeholder B (failed part placeholder): Replace the failed part name "oil tank" with placeholder B; Placeholder C (failure location placeholder): Replace the failure location name "chassis assembly" with placeholder C; Placeholder D (failure mode placeholder): Infer the failure mode as "oil leakage" according to the context and replace it with placeholder D.

[0098] S211. Generate multiple solution information according to the solution text replacement placeholders corresponding to each target problem-solving record. The solution information is used to represent the text content of the recommended solution for quality problems.

[0099] After completing the placeholder replacement, the system sorts the processed multiple solution texts according to the priority of the second recommended set (same model type first, failure mode frequency in descending order) and integrates them into the final recommended solution list.

[0100] For example, the generated solutions "Adjust the bolts of the transmission assembly to solve the oil leakage" and "Replace the seal of the automatic transmission with retarder assembly" will be output in the order of the original data, and at the same time, the content conflicting with the current fault logic (such as the operation steps involving irrelevant failure modes) will be automatically filtered. The generated solution list is directly associated with the target fault characteristics, reducing manual secondary modification and realizing the standardized and rapid output of solutions.

[0101] Exemplarily, following the example of the previous step, after completing the placeholder replacement, the solution information can be "ABCD, it is recommended to replace C and perform a sealing adjustment".

[0102] This embodiment realizes the generalization and structured reuse of the solution template. Through placeholder standardization (A = platform vehicle model, B = failed part, C = failure location, D = failure mode), the system abstracts the specific fault description (such as "The oil tank of the chassis assembly of flatbed truck No. 8 leaks oil") into a symbol combination with clear logical relationships (ABCD), enabling the same solution template to be dynamically adapted to different scenarios. For example, when dealing with "The seal of the transmission assembly of flatbed truck No. 6 leaks oil", only need to replace the placeholders with A (flatbed truck No. 6), B (seal), C (transmission assembly), D (oil leakage), and the adapted solution "The seal of the transmission assembly of flatbed truck No. 6 leaks oil, it is recommended to replace the transmission assembly and perform a sealing adjustment" can be generated. This design not only retains the core operation logic of the historical solutions (replace the component at the failure location + sealing adjustment), but also isolates the specific parameters through placeholders, avoiding the non-reusability of the template due to differences in part models and positions, thus significantly improving the cross-scenario applicability of the solution and reducing the cost of manually writing repetitive content.

[0103] S212, display multiple solution information to the user.

[0104] Specifically, the system sorts the processed multiple solution information according to the priority and displays it to the user. For example, the solution text "The oil tank of the chassis assembly of flatbed truck No. 8 leaks oil, it is recommended to replace the oil tank and perform a sealing adjustment", after placeholder replacement, generates the general template "ABCD, it is recommended to replace C and perform a sealing adjustment".

[0105] When dealing with the current quality problem (such as oil leakage from the sealing ring of the transmission assembly of flatbed truck No. 6), the system replaces the placeholders A / B / C / D in the solution information with the target information (A = flatbed truck No. 6, B = sealing ring, C = transmission assembly, D = oil leakage) to generate an adapted solution: "There is oil leakage from the sealing ring of the transmission assembly of flatbed truck No. 6. It is recommended to replace the sealing ring and adjust the sealing performance."

[0106] At the same time, the system extracts templates of other similar failure modes (such as oil leakage) from the historical database (for example, "Adjust the C connection bolts to solve D"), and after replacement, generates the second solution: "There is oil leakage from the sealing ring of the transmission assembly of flatbed truck No. 6. Adjust the connection bolts of the transmission assembly to solve the oil leakage problem."

[0107] Finally, the system displays multiple solutions to the user in the form of a list according to the rules of giving priority to the same model type and sorting in descending order of the frequency of failure modes.

[0108] S213, in response to the user's solution confirmation instruction, determine the recommended solution for the final quality problem from multiple solution information.

[0109] Specifically, after the user selects or confirms one of the solutions through the interface, the system performs the binding of the final solution according to the user's instruction. For example, if the user selects the first solution "Replace the sealing ring and adjust the sealing performance", the system associates the current fault code (171T311101011000G02M15), the product model type (flatbed truck No. 6) with the operation steps of this solution (replace the sealing ring, sealing performance detection) to generate a standardized solution document containing a complete fault analysis (failed parts, location, mode) and specific operation procedures. At the same time, the system adds this confirmation record to the historical database and updates the frequency statistics of the fault code to provide data support for subsequent recommendations.

[0110] S214, fill the recommended solution for the final quality problem into the solution maintenance page of the current product quality problem to obtain the problem solution recommendation page.

[0111] Specifically, the system fills the structured template of the final recommended solution into the solution maintenance page of the product quality problem. For example, the original problem "There is oil leakage from the oil tank of the chassis assembly of flatbed truck No. 8. It is recommended to replace the oil tank and adjust the sealing performance" generates a general template "There is oil leakage of B at C in the D of A. It is recommended to replace B and adjust the sealing performance" (A = placeholder for platform vehicle model, B = placeholder for failed part, C = placeholder for failure location, D = placeholder for failure mode) after placeholder replacement.

[0112] At this time, based on the analysis result of the current quality problem (such as oil leakage from the seal ring of the transmission assembly of flatbed truck No. 6), the system dynamically binds the placeholders to specific parameters: A is replaced by "flatbed truck No. 6", B is replaced by "seal ring", C is replaced by "transmission assembly", and D is replaced by "oil leakage", generating an adapted solution text "The oil leakage of flatbed truck No. 6 appears at the seal ring of the transmission assembly. It is recommended to replace the seal ring and adjust the sealing performance", and automatically fills this content into the edit box on the solution maintenance page for the user to view and confirm.

[0113] S215, in response to the user's editing operation on the problem solution recommendation page, adjusts the content on the problem solution recommendation page.

[0114] Specifically, when the user performs an editing operation on the recommended content on the solution maintenance page (such as modifying the text to "Replace the seal ring of the transmission assembly and perform three pressure tests"), the system listens to the user's input actions in real time, synchronizes the modified content to a memory variable through front-end event capture (such as the onChange event), and triggers the re-rendering of the page DOM elements to display the updated text.

[0115] If the user adjusts the format (such as adding a paragraph separator or highlighting keywords), the system parses the operation instructions through a rich text editor component (such as Quill.js) and persists the format tags (such as HTML tags or CSS styles). All editing operations will not affect the logical binding relationship of the original placeholders (such as A / B / C / D are still associated with the fault code analysis result), only perform personalized adjustments on the presentation layer text. The solution finally confirmed by the user will include the edited content and the original fault parameters, forming a traceable complete record.

[0116] When the user performs an editing operation on the problem solution recommendation page (such as modifying the text description, adjusting the step order, or supplementing operation details), the system captures the user's input behavior in real time through the event listening module on the front-end interface, synchronizes the modified content to a temporary data structure in memory (such as a JSON object), and dynamically updates the page DOM elements to immediately display the adjusted effect; if the user clicks Save, the system associates the edited complete solution text with the fault code and product model information of the current quality problem, generates a new version record containing the original recommended solution and the user's corrected content, and at the same time retains the modification log (such as modification time, operator, content difference), and transmits the updated data back to the server through the API interface for persistent storage in the "user corrected solution" independent partition of the historical database, ensuring that the system recommendation and the manually optimized version can be distinguished during subsequent queries or statistics, while keeping the analysis result of the original fault parameters (failed parts, locations, modes) unchanged, and only performing adaptive adjustments on the operation description layer of the solution.

[0117] S216. Based on the problem solution recommendation page after passing the review by the user, generate a new problem-solving record.

[0118] Specifically, when the user clicks the "Approved" button, it can be considered that the user has approved the problem solution recommendation page. The system extracts the complete content in the problem solution recommendation page (including the target fault code, product model, the solution text edited by the user, timestamp, operator information), and structures and encapsulates the different parts from the original recommended solution (such as the modified operation steps "Perform three pressure tests after replacing the sealing ring") to generate a new problem-solving record.

[0119] For example, the system combines fields such as the current fault code (171T311101011000G02M15), product model (Tablet No. 6 vehicle), final solution text ("Replace the sealing ring of the transmission assembly and perform three pressure tests"), review status (Approved), version number (V2.1), etc. into a JSON data object, and at the same time generates a unique record identifier (such as a SHA-256 value) through a hashing algorithm to ensure data integrity and traceability.

[0120] S217. Store the new problem-solving record in the historical database.

[0121] Specifically, the system persistently stores the newly generated problem-solving record in a specified data table (such as quality_solution_records) of the historical database through a database API interface (such as a RESTful POST request).

[0122] When storing, the system performs the following operations: 1) Data cleaning: Verify the fault code format (16 + 3 + 3 digits) and the product model naming specification (such as "Tablet X vehicle"). 2) Field mapping: Align the keys of the JSON object with the column names of the database table (such as solution_text maps to the solutions.content field). 3) Index optimization: Create a composite index on the fault code (fault_code) and product model (product_model) fields to accelerate subsequent queries. 4) Version control: If a historical record already exists for the same fault code, add a new record and associate it with the version chain (point to the original record through the parent_id field).

[0123] Meanwhile, the system writes the user operation logs (such as "User ID123 was approved on September 15, 2023, 14:30") into an independent operation audit table (audit_logs) to achieve full-link tracking of data modification.

[0124] The implementation process of the method in this embodiment is described below in combination with specific application examples as follows: (1) The key positioning content of quality problems is the platform vehicle model (product model) + fault code (the fault code has a fixed splicing rule, failure part information (16 bits) + failure location information (3 bits) + failure mode information (3 bits)). The fault code is determined offline, clearly indicating which component has what problem at what location.

[0125] For example: For a certain quality problem, the platform vehicle model is Flatbed No. 6, and the fault code is 171T311101011000G02M15. Then the fault code can be split into the following fault identification information according to the rule: Failure part: 171T311101011000 (Automatic transmission with retarder assembly); Failure location: G02 (Transmission assembly); Failure mode: M15 (Oil leakage).

[0126] Note: The failure parts, failure locations, and failure modes mentioned later are intercepted according to this rule and will not be described repeatedly later.

[0127] The fault code is basic data pre-maintained in the system and has an independent ledger page. Relevant data such as the corresponding code and name can be queried according to the splicing content.

[0128] Based on the above platform vehicle model and fault code information, it can be initially confirmed at this time that the problem is: Oil leakage in the transmission assembly of the automatic transmission with retarder assembly of Flatbed No. 6 vehicle.

[0129] (2) Since the same problem may occur in different platform vehicle models, that is, the same fault code for different platform vehicle models. At this time, the system will use the fault code of the current quality problem as a condition to query the historical quality problem data (problem solution records) with the same fault code in the historical database, and use the solutions of the queried historical quality problem data as recommended solutions for backup (the first recommended solution set) and top them (the maximum number of recommended solution backup data is 10).

[0130] Note: If the same fault code of the same platform vehicle model has multiple quality problems, the system will sort them according to the occurrence frequency. The higher the occurrence frequency, the more likely it is to be used as recommended solution backup data. For example: The flatbed No. 7 vehicle + fault code 171T311101011000G02M15 has occurred 10 times. The flatbed No. 8 vehicle + fault code 171T311101011000G02M15 has occurred 8 times. Then, the 10 pieces of data of the flatbed No. 7 vehicle + fault code 171T311101011000G02M15 will be used as the recommended solution for subsequent operations, and other data will not be recommended.

[0131] (3) After the previous operation, if the number of alternative data in the first recommended solution set is less than 10, the system obtains the failure mode of the current problem according to the fault code splicing rule. The system queries the historical quality problem data of the same failure mode, sorts the obtained data in the order of platform vehicle model > frequency of occurrence of the failure mode, uses the obtained problem-solving record data as the recommended data, and supplements the alternative data of the recommended solution to 10 pieces.

[0132] For example: The platform vehicle model of the current quality problem is the flatbed No. 6 vehicle, and the failure mode is oil leakage. The data obtained by the system using the failure mode of oil leakage as the query condition includes the following data: ① Historical quality problem data of the flatbed No. 6 vehicle with the failure mode of oil leakage (due to different failure positions, the failure code is different from the current quality problem, so it was not matched in the previous operation), with a quantity of 1 piece.

[0133] ② Historical quality problem data of the flatbed No. 7 vehicle with the failure mode of oil leakage, with a quantity of 2 pieces.

[0134] ③ Historical quality problem data of the flatbed No. 8 vehicle with the failure mode of oil leakage, with a quantity of 3 pieces.

[0135] Because the platform vehicle model of ① is the same as that of the current problem, the data included in ① is sorted as the optimal recommendation. Then, the frequency of occurrence of ③ is higher than that of ②, so the data priority of ③ is higher than that of ②. According to the priority rule, the final obtained priority order is ①>③>②. The system supplements the data with higher ranking into the recommended solution alternatives (up to 10 pieces) according to the sorting to obtain the second recommended solution set.

[0136] Note: If the number of alternative data in the second recommended solution set is still less than 10 after this operation, no further supplementation will be performed in the subsequent steps, and only the currently obtained data will be used as the recommended data for subsequent operations.

[0137] (4)The system processes the content of the corresponding solutions for the finally obtained recommended solution data (up to 10 items), splits the obtained failed part code / name, failed location code / name, failure mode code / name, and the platform model corresponding to the problem according to the failure code corresponding to each item of data, intercepts and deletes them in the corresponding solution content, and replaces them with the platform model, failed part code / name, failed location code / name, and failure mode code / name information of the current quality problem.

[0138] For example: a. The system obtains the data corresponding to the current quality problem (the quality problem for which a recommended solution needs to be provided): Platform model: Flatbed No. 6 vehicle; Failure code: 171T311101011000G02M15 (according to the failure code splicing rule and failure code information query, the following data is obtained); Failed part code: 171T311101011000, failed part name: Automatic transmission with retarder assembly; Failed location code: G02, failed location name: Transmission assembly; Failure mode code: M15, failure mode name: Oil leakage.

[0139] b. One piece of data corresponding to the quality problem in the recommended solution data obtained by the system is: Platform model: Flatbed No. 8 vehicle; Solution content: Oil leakage from the oil tank of the chassis assembly of the flatbed No. 8 vehicle. It is recommended to replace the oil tank and adjust the sealing performance; Failure code: 171T311101012000A03M15; according to the failure code splicing rule and failure code information query, the following data is obtained: Failed part code: 171T311101012000, failed part name: Chassis assembly; Failed location code: A03, failed location name: Oil tank; Failure mode code: M15, failure mode name: Oil leakage; As Figure 2 shown: The system intercepts and deletes the failed part, failed location, and failure mode information included in the solution content.

[0140] Obtained content: ABCD, it is recommended to replace C and adjust the sealing performance. (ABCD is the placeholder for the corresponding content described Figure 2 in) c. Replace the placeholder in the content obtained in step b with the current quality problem data obtained in step a as follows: A (placeholder for platform vehicle model) → Flatbed Truck No. 6; B (placeholder for failed part) → Automatic transmission with retarder assembly; C (placeholder for failure location) → Transmission assembly; D (placeholder for failure mode) → Oil leakage; Obtain the content of the final solution: The automatic transmission with retarder assembly of the transmission assembly of Flatbed Truck No. 6 has oil leakage. It is recommended to replace the transmission assembly and adjust the sealing performance.

[0141] (5) When the user fills in the solution, the system will intelligently push the 10 recommended data obtained after all the above operations (application pop-up window). The user can view the content of the solution obtained after the system's intelligent processing and all the information of the corresponding quality problem data for each recommended data. After determining the appropriate recommended solution, by clicking the apply button, the content of the solution obtained after the system processes the selected data can be filled into the solution maintenance page of the current problem at one key, and the content of the filled solution can be manually adjusted and supplemented.

[0142] (6) When the solution to this problem passes the review, the system will record the problem and the corresponding solution in the system's historical database. After the accumulation of data in the later stage, the format of the solution content will be more templatized. When the system intelligently recommends solutions for new problems, the data screening will be faster, the recommended solutions will be more accurate, and the solution content will be more standardized.

[0143] In some embodiments, after generating the recommended solution for the quality problem, this embodiment can also convert the recommended solution for the quality problem into an AR (Augmented Reality) operation guide, and superimpose virtual instructions (such as part disassembly sequence, torque parameters, and dangerous area markings) through smart glasses or mobile devices. The specific implementation process is as follows: Deploy the SLAM (Simultaneous Localization and Mapping) spatial positioning engine: Integrate the visual-inertial SLAM algorithm in an AR device (such as HoloLens), and use the camera and IMU data to construct a three-dimensional point cloud map of the maintenance environment in real time, accurately tracking the position and orientation of the device in physical space (accuracy ±2mm). Among them, the visual-inertial SLAM algorithm is a real-time positioning and mapping technology that fuses camera visual data and inertial measurement unit (IMU) information. It captures environmental feature points (such as device contours, bolt positions) through the camera and calculates their relative movements. At the same time, it uses the angular velocity and acceleration data of the IMU to infer the instantaneous pose changes of the device, and then aligns the two types of data in space and time through the Kalman filter or graph optimization algorithm to make up for the limitations of a single sensor (such as tracking loss of the camera during rapid movement or in low light conditions, and cumulative error drift of the IMU). Finally, it realizes high-precision three-dimensional spatial positioning (error less than 2 millimeters) of the device in the maintenance scenario and the construction of a dynamic environment map, ensuring that AR virtual guides (such as disassembly arrows, torque prompts) can be accurately superimposed on the real surface of the physical device, and maintaining a stable and reliable virtual-real fusion effect even under complex working conditions (such as personnel movement, tool occlusion).

[0144] Load the device CAD model library: Convert the CAD engineering model of the target part (such as a transmission) into lightweight mesh data (such as glTF format), associate metadata (part name, disassembly and assembly sequence, torque parameters), and pre-load it into the local storage of the AR device.

[0145] Real-time CAD-physical space registration: Based on the point cloud map generated by SLAM, align the key feature points (such as bolt holes, edge contours) of the CAD model and the physical part through the ICP (Iterative Closest Point) algorithm to achieve millimeter-level fitting of the virtual model and the physical object.

[0146] Generate dynamic AR overlay guides: Integrate the real-time camera image and the registered CAD model, and overlay dynamic three-dimensional arrows (indicating the disassembly direction), thermal color temperature markings (such as overloaded components showing red), and floating text prompts (such as "Bolt M8: 25N·m ±5%") on the surface of the physical device.

[0147] Embed the voice command interaction module: Integrate a voice recognition engine (such as Whisper), and maintenance personnel can describe operation anomalies (such as "bolt jamming") in natural language, and the system will parse and trigger corresponding prompts in real time (such as "switch to impact wrench mode").

[0148] Record maintenance deviation events: When a deviation is detected by voice description or sensor data (such as unqualified torque), the current AR screen (including superimposed guidance) is automatically captured, the timestamp and operation context are recorded, and a structured deviation log is generated (such as "Step 3: Bolt torque 18 N·m < 25 N·m | Reason: Stuck | Voice note: It is recommended to lubricate the thread").

[0149] Dynamically adjust the guidance logic: Based on the deviation log, a reinforcement learning model is trained. When the same type of deviation occurs repeatedly (such as a certain bolt getting stuck frequently), a pre-operation guidance (such as "Spray rust inhibitor in advance") is automatically inserted before the corresponding step, and the risk area is highlighted and prompted.

[0150] In this embodiment, through the intuitive guidance of virtual-real fusion, the high efficiency, standardization, and safety of the maintenance process are realized: The animation of the part disassembly sequence and the torque parameter annotation (such as "Bolt M8: 25 N·m") superimposed on AR can accurately guide the maintenance personnel to operate according to the standard process, avoiding misassembly or parameter deviation caused by lack of experience; The real-time dangerous area annotation (such as the red mask of high-temperature components) and tactile feedback (such as the vibration of the equipment when the torque is exceeded) can actively warn of potential risks and reduce the probability of work-related injuries; At the same time, the AR guidance transforms the abstract text plan into spatially visual interactive steps (such as arrow-guided tool paths), enabling the novice maintenance efficiency to be increased by more than 60%, significantly shortening the fault handling cycle and equipment downtime losses, and ultimately achieving multiple goals of controllable maintenance quality, cost optimization, and personnel safety. In some embodiments, the dynamic adjustment of the guidance logic specifically includes: Collect maintenance deviation logs: The system records the deviation events (such as "bolt stuck") during the maintenance process in real time, including operation steps, sensor data (torque value), voice notes, and AR screen screenshots, and generates a structured deviation log database.

[0152] Identify high-frequency similar deviation patterns: Analyze the deviation logs through clustering algorithms (such as DBSCAN), count the occurrence frequency and context correlation of the same type of deviation (such as a certain bolt getting stuck) (such as the frequency increases by 50% when the environmental humidity > 80%), and mark the high-frequency risk nodes.

[0153] Dynamically insert pre-operation guidance: When the maintenance process triggers a high-frequency risk node, call the adaptive maintenance pre-operation reinforcement learning model to output pre-operation instructions (such as "Before Step 3: Spray rust inhibitor"), and highlight the associated components (such as the bolt flashing red) and display the operation animation in the AR view.

[0154] Closed-loop verification and optimization: After performing the pre-operation, the system tracks the actual deviation occurrence rate (such as the number of stuck times), and feeds the result back to the reinforcement learning model to update the policy weights, forming a self-evolving loop of "risk prediction - intervention - effect verification".

[0155] Among them, the construction and training process of the adaptive maintenance pre-operation reinforcement learning model includes the following stages: (1) Data preparation stage: Extract structured deviation logs from the historical database. Each log record contains the maintenance stage (such as disassembly, installation), equipment sensor data (torque, temperature), environmental parameters (humidity, dust), pre-operation actions (such as spraying rust remover), and result indicators (deviation occurrence, time consumption, cost).

[0156] Perform NLP processing on unstructured data (maintenance personnel's voice remarks), extract keywords (such as "rust", "loose") and encode them into binary features (such as "rust = 1"), and construct a training set and a test set after cleaning invalid or conflicting data.

[0157] (2) State and action space design stage: The state space consists of four-dimensional features: 1) One-Hot encoding of the maintenance stage (for example, 10 stages correspond to 10 dimensions); 2) Equipment working conditions (normalized sensor data such as temperature, vibration); 3) Environmental parameters (humidity, dust concentration); 4) Historical frequency of similar deviations (statistical value of a sliding window). The action space defines 20 pre-operation instructions, such as A0 (no operation), A1 (spraying rust remover), A2 (preheating the tool to 80 °C), A3 (cleaning the contact surface), etc. Each action corresponds to a unique operation code and execution parameters.

[0158] (3) Reward function design stage: Design a multi-objective weighted reward function R: basic deviation reward R 偏差 (deviation not occurring +1.0, occurring -0.5), efficiency reward R 效率 (+0.2 for each minute saved, -0.1 for overtime), cost penalty R 成本 (the consumed cost is converted into a penalty value according to a certain ratio), exploration incentive R 探索 (+0.3 for the first execution of an action).

[0159] Reward function R = 0.6R 偏差 + 0.2R 效率 + 0.1R 成本 + 0.1R 探索 , and balance short-term benefits and long-term policy stability by dynamically adjusting the weights.

[0160] (4) Algorithm implementation stage: Adopt a Double Deep Q-Network (DDQN) architecture: a main network (input 64-dimensional state, output 20 action Q-values) and a target network (parameters synchronized with a delay). The hidden layer is a fully connected layer with 128→64 nodes (ReLU activation), and the output layer has a linear activation. The experience replay pool stores 100,000 samples, and sampling is performed according to the TD error priority (the extraction probability of samples with high error +30%) to prevent overfitting caused by data correlation. The optimizer uses Adam (learning rate 0.001), and the loss function is the mean squared error (MSE).

[0161] (5) Training process stage: During offline pre-training, 512 historical data are extracted in each batch, and the Q-values of the main network and the target network are calculated. target Qtarge = r + γ⋅max TargetNet(s'). Where: Qtarge represents the target Q-value, which represents the expected long-term return of the current state-action pair and is used to calculate the training loss; r represents the immediate reward, that is, the direct reward obtained after performing the action (if the deviation does not occur, +1.0); γ represents the discount factor, which can take a value of 0.95 and is used to balance the importance of the current reward and future potential rewards (the weight of future rewards decays to 95% of the current); TargetNet(s') represents the Q-value prediction of the target network for all possible actions in the next state s'; max represents selecting the maximum Q-value in the next state s', which represents the expected return of the optimal action in the future.

[0162] Update the parameters of the main network, and synchronize the target network every 1000 steps. In the online fine-tuning stage, new maintenance data is received in real time. If the TD error > 0.5, incremental training is triggered: add the new data to the experience pool and sample 256 pieces according to the new-old ratio of 1:3, and perform gradient descent; retrain the whole amount once a week to cope with equipment aging or process changes.

[0163] (6) Policy verification and deployment stage: Verify the model effect on an independent test set (30% historical data), requiring a deviation reduction rate ≥ 40% and a cost increase rate ≤ 15%, and the action interpretability is evaluated by experts (≥ 8 / 10 points). After verification, the model is encapsulated as a REST API, which receives real-time state vectors per second (such as the current maintenance stage, humidity 80%, bolt torque 18 N·m), and returns the top-3 pre-operation suggestions (such as A1 confidence 82%, A3 confidence 75%). The AR system automatically inserts guidelines according to the confidence threshold (> 70%) (such as highlighting the bolt and showing the animation of spraying rust remover), otherwise prompts manual decision-making. The execution result is fed back to the database to drive the continuous optimization of the model.

[0164] In some embodiments, after dynamically adjusting the guidance logic, the following steps are further included: Synchronous remote expert collaboration: The deviation log and the AR screen are uploaded to the cloud in real time. The expert views the synchronized perspective with overlaid guidance through the WebAR interface and marks the correction opinions (such as "Use a heat gun to soften the thread sealant"), and the local AR view is updated with prompts immediately.

[0165] Generate an intelligent maintenance report: Integrate the standard operation guidelines, deviation records, and expert correction content, generate an interactive report according to the timeline, the AR guidance animation can be clicked to play back the steps, and it can be exported in PDF / 3DPDF format for quality traceability.

[0166] Closed-loop optimization of CAD-scene mapping: According to the SLAM positioning data and manual correction records of each maintenance, continuously optimize the feature matching rules between the CAD model and the physical device (such as adaptively adjusting the ICP weight parameters), and improve the initialization speed and registration accuracy of subsequent AR guidance.

[0167] This embodiment realizes the precision and intelligence of maintenance guidance through augmented reality (AR) technology. Based on the high-precision spatial registration of the visual-inertial SLAM algorithm and the CAD model, virtual operation guidelines (such as part disassembly paths and torque parameter markings) are overlaid on the surface of the physical device in real time to guide the maintenance personnel to operate according to the standardized process; combined with voice interaction and sensor feedback (such as tactile vibration warning of over-limit torque), operation deviations (such as "The bolt does not reach 25 N·m") are captured and recorded in real time, and the guidance logic is dynamically adjusted (such as inserting a lubrication pre-operation step); at the same time, it supports remote experts to provide synchronous guidance for complex problems through AR markings, forming a closed-loop maintenance network of "local operation-cloud collaboration". Ultimately, the maintenance efficiency of novices is increased by 60%, the operation error rate is reduced by 45%, the cost of expert resource utilization is reduced by 70%, and the guidance strategy is self-optimized based on historical deviation data, continuously improving the maintenance quality and equipment reliability.

[0168] The method provided in the above embodiment can be executed by a product quality problem solution automatic generation system, and this system is composed of electronic devices. The following describes this electronic device in the embodiment of the present invention from the perspective of hardware processing. Please refer to Figure 3 , which is a schematic structural diagram of an entity device of the electronic device in the embodiment of the present invention.

[0169] It should be noted that Figure 3 The structure of the electronic device shown is only an example and should not bring any limitations to the functions and usage ranges of the embodiments of the present invention.

[0170] Such as Figure 3As shown, the electronic device includes a Central Processing Unit (CPU) 401, which can perform various appropriate actions and processes according to the program stored in the Read-Only Memory (ROM) 402 or the program loaded from the storage section 408 into the Random Access Memory (RAM) 403, such as executing the method described in the above embodiments. In the Random Access Memory (RAM) 403, various programs and data required for system operation are also stored. The Central Processing Unit (CPU) 401, the Read-Only Memory (ROM) 402, and the Random Access Memory (RAM) 403 are connected to each other via a bus 404. An Input / Output (I / O) interface 405 is also connected to the bus 404.

[0171] The following components are connected to the Input / Output (I / O) interface 405: an input section 406 including an audio input device, a button switch, etc.; an output section 407 including a display, an audio output device, an indicator light, etc.; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the Input / Output (I / O) interface 405 as needed. A removable medium 411, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 410 as needed so that the computer program read from it can be installed into the storage section 408 as needed.

[0172] Specifically, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network via the communication section 409, and / or installed from the removable medium 411. When the computer program is executed by the Central Processing Unit (CPU) 401, various functions defined in the present invention are executed.

[0173] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fibers, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0174] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. Among them, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the above-mentioned module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings.

[0175] Specifically, the electronic device of this embodiment includes a processor and a memory. The memory is coupled to one or more processors. The memory is used to store computer program code, and the computer program code includes computer instructions. One or more processors call the computer instructions to cause the electronic device to execute the method provided in the above-mentioned embodiment.

[0176] On the other hand, the present invention also provides a computer-readable storage medium, which may be included in the electronic device described in the above-mentioned embodiment; or it may exist separately and not be assembled into the electronic device. The above-mentioned storage medium carries one or more computer programs. When the above-mentioned one or more computer programs are executed by a processor of the electronic device, the electronic device is caused to implement the method provided in the above-mentioned embodiment.

[0177] As mentioned above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the various embodiments of the present invention.

[0178] As used in the foregoing embodiments, depending on the context, the term "when" may be construed to mean "if", "after", "in response to determining", or "in response to detecting". Similarly, depending on the context, the phrase "upon determining" or "if (the stated condition or event) is detected" may be construed to mean "if determined", "in response to determining", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0179] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the foregoing embodiments can be implemented by a computer program instructing relevant hardware. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the foregoing method embodiments. The foregoing storage medium includes various media that can store program codes, such as ROM, random access memory (RAM), magnetic disks, or optical discs.

Claims

1. A method for automatically generating a solution to product quality problems, characterized in that, Including: Determine the fault code of the current product quality problem and the product model information as the target fault code and the target product model information. The fault code includes multiple fault identification information, and the fault identification information includes failed part information, failed location information, and failure mode information; Based on the target fault code and the target product model information, determine the first recommended solution set from the historical database. The historical database includes multiple problem-solving records, and the problem-solving records include the model information of the problem product, the fault code, and the corresponding problem-solving method; When the data volume of the problem-solving records in the first recommended solution set is less than the preset threshold, based on the fault identification information corresponding to the target fault code, determine the second recommended solution set from the historical database. The data volume of the problem-solving records in the second recommended solution set is greater than that of the problem-solving records in the first recommended solution set; Generate a recommended solution for the quality problem according to the target fault code, the target product model information, and the problem-solving records in the second recommended solution set.

2. The method according to claim 1, wherein The generating a recommended solution for the quality problem according to the target fault code, the target product model information, and the problem-solving records in the second recommended solution set includes: Determine the solution text replacement placeholders of each of the target problem-solving records according to the target fault code, the target product model information, and the multiple target problem-solving records in the second recommended solution set. The solution text replacement placeholders include product model placeholder, failed part placeholder, failed location placeholder, and failure mode placeholder; Generate multiple solution information according to the solution text replacement placeholders corresponding to each of the target problem-solving records. The solution information is used to represent the text content of the recommended solution for the quality problem.

3. The method according to claim 2, wherein After generating multiple solution instruction information according to the solution text replacement placeholders corresponding to each of the target problem-solving records, it further includes: Display multiple pieces of the solution information to the user; In response to the solution confirmation instruction of the user, determine the final recommended solution for the quality problem from the multiple pieces of the solution information.

4. The method according to claim 3, wherein After determining the final recommended solution for the quality problem from the multiple pieces of the solution information in response to the solution confirmation instruction of the user, it further includes: Fill the final recommended solution for the quality problem into the solution maintenance page of the current product quality problem to obtain a problem solution recommendation page; In response to the editing operation of the user on the problem solution recommendation page, adjust the content in the problem solution recommendation page.

5. The method according to claim 4, wherein After adjusting the content in the problem solution recommendation page in response to the editing operation of the user on the problem solution recommendation page, it further includes: In response to the audit approval operation of the user on the problem solution recommendation page, generate a new problem-solving record based on the problem solution recommendation page after audit approval; Store the new problem-solving record in the historical database.

6. The method according to any one of claims 1-5, characterized in that, Based on the target fault code and the target product model information, determining a first recommended solution set from the historical database, including: Querying all historical problem-solving records containing the same fault code from the historical database according to the target fault code and the target product model information to obtain an initial data set; Filtering the initial data set based on the target product model information, retaining the historical records that match the target product model information to obtain a same-model data set; Sorting the same-model data set according to the total number of times the fault codes of each problem-solving record appear in the historical database to obtain a sorting result; Selecting the first N problem-solving records in the sorting result to obtain the first recommended solution set.

7. The method according to any one of claims 1-5, characterized in that, Based on the fault identification information corresponding to the target fault code, determining a second recommended solution set from the historical database, including: Retrieving problem-solving records containing the same failure mode from the historical database based on the target failure mode information in the target fault code to obtain an extended data set; Filtering out the problem-solving records that match the target product model information from the extended data set as the first-priority data, and the unfiltered problem-solving records as the second-priority data; Sorting the second-priority data according to the historical occurrence frequency of the target failure mode information to obtain the sorted second-priority data; Intercepting the corresponding number of problem-solving records from the first-priority data and the sorted second-priority data according to the difference between the preset threshold and the current data volume of the first recommended solution set to obtain the second recommended solution set.

8. An electronic device, characterized in that, Including one or more processors and a memory; The memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to cause the electronic device to execute the method according to any one of claims 1-7.

9. A computer-readable storage medium storing computer instructions, characterized in that, When the computer instructions run on the electronic device, causing the electronic device to execute the method according to any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product runs on the electronic device, causing the electronic device to execute the method according to any one of claims 1-7.

Images