A method, system, computer device, and storage medium for avoiding quality problems.

Abstract

This application relates to a method, system, computer device, and storage medium for avoiding quality problems. The method includes obtaining a parts list of a target project; obtaining problem data of quality problems corresponding to the parts from a problem database based on the parts list; obtaining experience data corresponding to the problem data from an experience database based on the problem data, the experience data including at least preventive measures and detection measures for quality problems; implementing the preventive measures of the experience data on the target project, and determining whether the quality problems corresponding to the parts recur based on the detection measures; when the quality problems corresponding to the parts recur, feeding back a result of avoidance failure to the problem database; when the quality problems corresponding to the parts do not recur, feeding back a result of avoidance success to the experience database. Using the method of this application can improve the problem that historical experience cannot be effectively utilized in the prior art.

Description

Technical Field

[0001] This application relates to the field of quality control technology, and in particular to a method, system, computer device, and storage medium for avoiding quality problems. Background Technology

[0002] During the design, development, and manufacturing of automobiles, a large number of quality problems are accumulated, resulting in a wealth of lessons learned. By applying these lessons, knowledge can be accumulated, and historical quality problems that have occurred in previous projects can be avoided.

[0003] Some companies have developed digital systems to record historical quality issues, managing these issues by entering them into database software. This has solved the problems of data recording and storage, but they have not designed modules to transform these issues into knowledge and experience. Engineers are unaware of which quality issues have been transformed and accumulated into knowledge, and therefore cannot effectively utilize historical experience and knowledge, nor can they know how effective historical experience and knowledge are in projects. Summary of the Invention

[0004] Based on this, a method, system, computer device, and storage medium for avoiding quality problems are provided to improve the problem that historical experience cannot be effectively utilized in the prior art.

[0005] On the one hand, a method for avoiding quality problems is provided, including:

[0006] Obtain the parts list of the target project, and based on the parts list, obtain the quality problem data corresponding to the parts from the problem database, wherein the problem database is established based on the collected text information of the quality problems;

[0007] Based on the problem data, experience data corresponding to the problem data is obtained from the experience base. The experience data is obtained by transforming the problem data and includes at least preventive and detection measures for quality problems.

[0008] The target project is implemented based on preventive measures according to the empirical data, and the detection measures are used to determine whether the quality problems corresponding to the parts recur.

[0009] When a quality problem corresponding to a part occurs repeatedly, the result of failure to avoid it is fed back to the problem database; when a quality problem corresponding to a part does not occur repeatedly, the result of successful avoidance is fed back to the experience database.

[0010] In one embodiment, the problem database is built based on the collected text information of quality problems, including:

[0011] When new problem data is collected, the new problem data is deduplicated based on the existing problem data in the problem database, including:

[0012] Based on the text information of the newly added question data, the importance score of each word in the newly added question data is obtained;

[0013] Multiple keywords are determined based on the importance scores;

[0014] Based on the frequency of occurrence of the keywords in the newly added question data, a first calculation vector is obtained;

[0015] A second calculation vector is obtained based on the frequency of occurrence of the keywords in the original problem data;

[0016] Based on the first and second calculation vectors, a cosine similarity is obtained, and based on the cosine similarity, it is determined whether the newly added question data is duplicated with the original question data.

[0017] When the newly added problem data does not duplicate the existing problem data, the newly added problem data is stored in the problem database.

[0018] In one embodiment, obtaining the importance score of each word in the newly added question data based on the text content of the newly added question data includes:

[0019] Obtain the term frequency and inverse document frequency of each word in the newly added question data;

[0020] The importance score is determined based on the term frequency and inverse document frequency.

[0021] In one embodiment, after determining multiple keywords based on the importance score, the process includes:

[0022] The keywords are arranged in a fixed order to obtain a text vector composed of keyword elements;

[0023] The step of obtaining the first calculation vector includes: converting the keyword elements in the text vector into a vector in which all numeric elements are first values, as the first calculation vector;

[0024] The process of obtaining the second calculation vector includes: traversing the frequency of each keyword element in the text vector in the original question data; when the frequency of the keyword element in the original question data is zero, it is converted into a numeric element with a second value; otherwise, it is converted into a numeric element with a first value, so as to obtain the second calculation vector.

[0025] In one embodiment, the experience data is obtained by transforming the problem data, including:

[0026] Based on the text information of the problem data, determine whether the problem data meets the empirical conditions, including determining that the empirical conditions are met when the text information of the problem data contains a cause analysis field and an improvement measure field;

[0027] The experience data is obtained by using the textual information of the problem data that meets the empirical conditions to establish the experience database.

[0028] In one embodiment, obtaining the experience data to establish the experience base includes:

[0029] Determine the problem category corresponding to the problem data, where the problem category includes structural component quality problems or software and hardware quality problems;

[0030] When the problem data corresponds to a structural component quality problem, the empirical data is obtained based on FMEA tool analysis;

[0031] When the problem data corresponds to a software or hardware quality issue, the empirical data is obtained based on TRIZ causal chain analysis.

[0032] In one embodiment, the problem category also includes management problems;

[0033] The process of obtaining the experience data to establish the experience base also includes:

[0034] When the problem data corresponds to a management problem category, the experience data is obtained based on the analysis of a timeline chart of management problems.

[0035] On the other hand, a quality problem avoidance system is provided, the system comprising:

[0036] The problem database module is used to store and output problem data for the quality problems corresponding to the parts. The problem database module is established based on the collected text information of the quality problems.

[0037] The experience base module is used to store and output the experience data corresponding to the problem data, and the experience data includes at least the preventive measures and detection measures for quality problems.

[0038] A specialized avoidance module is used to obtain a parts list for a target project, and based on the parts list, obtain problem data on the quality problems corresponding to the parts from a problem database module, and based on the problem data, obtain experience data corresponding to the problem data from an experience database module; implement preventive measures for the target project based on the experience data, and determine whether the quality problems corresponding to the parts recur based on the detection measures; when the quality problems corresponding to the parts recur, the avoidance failure result is fed back to the problem database; when the quality problems corresponding to the parts do not recur, the avoidance success result is fed back to the experience database.

[0039] In another aspect, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method.

[0040] A computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the method.

[0041] The aforementioned quality problem avoidance method, system, computer equipment, and storage medium store problem data and experience data for quality problems in a problem database and an experience database, respectively. When a new project is initiated, a parts list for the target project is obtained. Based on the parts list, problem data for the corresponding quality problems of the parts is retrieved from the problem database. Then, based on the problem data, experience data corresponding to the problem data is retrieved from the experience database. Preventive measures based on the experience data are implemented for the target project. Based on the detection measures of the experience data, it is determined whether the quality problem corresponding to the part will recur. If it recurs, the result of avoidance failure is fed back to the problem database; if it does not recur, the result of avoidance success is fed back to the experience database. Through this method, implementers can perform closed-loop management of quality problems. When implementing a new project, they can find the experience data corresponding to the quality problem based on the quality problem, thereby implementing preventive measures to avoid the recurrence of the quality problem, effectively utilizing historical experience and knowledge. Furthermore, detection measures can be used to detect the avoidance results and feed the avoidance results back to the problem database or experience database, indicating whether the experience is effective for subsequent projects. Attached Figure Description

[0042] Figure 1 This is a flowchart illustrating a quality problem avoidance method in one embodiment;

[0043] Figure 2 This is a flowchart illustrating the problem database management process in one embodiment;

[0044] Figure 3 This is a flowchart illustrating the experience base management process in one embodiment;

[0045] Figure 4 A structural block diagram of a quality problem avoidance system in one embodiment;

[0046] Figure 5 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0047] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0048] Quality issues in automobile design, development, and manufacturing severely impact vehicle quality and safety. By establishing a comprehensive quality management system, strengthening quality management during the design and development phases, enhancing quality management during the manufacturing phase, improving quality inspection and monitoring, and fostering a quality culture, product quality can be effectively improved, costs reduced, customer satisfaction increased, and market competitiveness enhanced. Furthermore, analytical methods such as fault tree analysis, failure mode and effects analysis, and Six Sigma management can identify potential problems in products or processes, allowing for preventative corrective measures and ultimately improving product or process quality and reliability.

[0049] Some companies manually input historical quality issues into spreadsheets using office software such as Excel, but this is difficult to manage and maintain. Therefore, a digital system for recording historical quality issues was developed, which manages the quality issues by entering them into the database software, thus solving the problem of data recording and storage.

[0050] However, the existing approach does not take into account historical experience. When developing new projects, engineers and other implementers cannot quickly and comprehensively identify the problems that the new project will face, nor can they effectively find historical experience that the new project can utilize, nor can they quickly identify the effectiveness of historical experience.

[0051] This application provides a method for avoiding quality problems, which uses data from a problem database and an experience database to specifically avoid problems in new projects.

[0052] In one embodiment, the method is as follows Figure 1 As shown, it includes the following steps:

[0053] Step 110: Obtain the parts list of the target project, and based on the parts list, obtain the problem data of the quality problems corresponding to the parts from the problem database.

[0054] When a new vehicle model project is launched, the first step is to identify the parts involved in the project and create a parts list. Based on the parts list, applicable quality issues are then identified from the issue database.

[0055] The issue database is configured to collect quality issues from both inside and outside the enterprise. For example, it can collect quality issues from internal channels (such as production and after-sales service) and public channels of other companies in the same industry. The collected quality issues are stored in the database in the form of text information, such as fields for problem description, cause analysis and improvement measures. When the issue database outputs issue data, it outputs text information containing the above fields.

[0056] Step 120: Based on the problem data, obtain the experience data corresponding to the problem data from the experience base.

[0057] The experience base is used to transform quality problems into technical or management knowledge, and to apply the knowledge to product design standards, test specifications, drawings, 3D models, software functional requirements lists, checklists, special characteristic checklists, control plans, FMEA (Failure Mode and Effects Analysis), work instructions and process documents, etc., to form preventive measures and detection measures.

[0058] Among them, experience data is obtained by transforming problem data. For example, for quality problems in the problem database, problem analysis and knowledge transformation are carried out, and analytical methods and tools are used to analyze and improve the causes of problem failures, thereby obtaining corresponding preventive and detection measures.

[0059] Once implementers obtain the problem data for quality issues, they can filter out the relevant experience data from the experience database and develop preventative and detection measures for the quality issues.

[0060] Step 130: Implement the target project based on the preventive measures according to the empirical data, and determine whether the quality problem corresponding to the part recurs based on the detection measures.

[0061] During the implementation of the target project, preventive measures are implemented to avoid quality problems. At the same time, detection measures are used to monitor the effectiveness of quality problem avoidance, detect whether the same quality problems will recur, and output the detection results.

[0062] Step 140: Feedback detection results, including feeding back failure to avoidance to the problem database when the quality problem corresponding to the part occurs repeatedly; and feeding back successful avoidance to the experience database when the quality problem corresponding to the part does not occur repeatedly.

[0063] For example, if the same quality problem recurs during the implementation of the target project, it indicates that the historical experience data has not been effective in improving the target project, so it is marked as a recurring problem in the problem database; if the same quality problem is successfully avoided, the historical experience data is marked as effective experience, and the number of projects to which it was applied is increased by one, thereby providing a prompt to subsequent project implementers, or recording new experience gained during the implementation of the target project into the experience database.

[0064] In the above process, implementers can obtain problem data corresponding to each part from the problem database based on the parts list, such as the problem description, cause analysis and improvement measures for the quality problems involved in each part, and then obtain experience data on quality problems from the experience database based on the problem data, such as preventive measures and detection measures. Therefore, implementers can effectively use historical experience to improve the implementation process of the current target project, thereby avoiding the recurrence of quality problems. On the other hand, they can also monitor the avoidance results through detection measures, and then record the successful avoidance results in the experience database, or the unsuccessful avoidance results in the problem database, so as to facilitate subsequent implementers to track the application of experience.

[0065] The issue database is used to record a large number of quality issues. However, during the design and production process, the same quality issue may be raised by different people and repeatedly recorded, causing problems such as database bloat and data pollution. Implementers have to manually identify duplicate issues when using it.

[0066] In one embodiment of this application, a deduplication process for problem data is proposed for the management of the problem database. When new problem data is collected, the new problem data is deduplicated based on the existing problem data in the problem database.

[0067] The newly added issue data includes internal issues collected from internal channels within the company, or external issues collected from public channels (such as the media) of other companies in the same industry.

[0068] Information on internal or external problems is compiled into text according to a prescribed format and entered into the problem database. The prescribed format includes fields such as problem description, cause analysis, and improvement measures for describing internal or external problems.

[0069] The problem database merges the three fields of problem description, cause analysis, and improvement measures for quality problems. The merged text is used as the text information for new problem data. Based on the text information of the new problem data, the importance score of each word in the new problem data is obtained.

[0070] In a piece of text information, different words have different degrees of importance for the information expression of the text. By evaluating the importance of each word, the word that can best represent the new question data is screened out as a keyword. For example, a series of preprocessing operations are first performed, such as removing special symbols, splitting the text into words, and excluding common stop words such as "de", "zai", etc., to reduce noise. The implementation of preprocessing operations can be based on existing methods, and will not be repeated herein.

[0071] The importance score of each word is calculated based on the TF-IDF (term frequency-inverse document frequency) algorithm. The TF-IDF algorithm is a statistical method used for information retrieval and text mining. It consists of two parts: TF (term frequency) and IDF (inverse document frequency). Term frequency represents the number of times a word appears in the text information of new question data, which reflects the importance of the word in the new question data. Inverse document frequency refers to the reciprocal of the frequency of a word appearing in all text information (including new question data and original question data), and the logarithm of the obtained reciprocal value is taken, which reflects the universality or rarity of the word.

[0072] According to the TF-IDF algorithm, calculate the term frequency TF and inverse document frequency IDF of each word in the new question data, then calculate the TF-IDF value based on the term frequency TF and inverse document frequency IDF, and take the TF-IDF value as the importance score, where TF-IDF value = TF * IDF.

[0073] Sort the TF-IDF values in descending order, and select words with high TF-IDF values as keywords. In this embodiment, the 10 words with the highest TF-IDF values are extracted as keywords.

[0074] After determining 10 keywords, it is necessary to identify based on the keywords whether the quality problem that needs to be added already exists in the problem database. In this embodiment, deduplication is performed on the newly added quality problem based on cosine similarity. For example:

[0075] Calculate the occurrence frequency of each keyword in the new question data, and combine the occurrence frequencies corresponding to each keyword to obtain a first calculation vector;

[0076] According to the occurrence frequency of each keyword in each original question data, combine the occurrence frequencies corresponding to each keyword to obtain a second calculation vector corresponding to each original question data;

[0077] The cosine similarity between the first calculated vector and the second calculated vector corresponding to each original problem data is calculated. The closer the cosine similarity is to 1, the more similar the two text information are. When the cosine similarity exceeds a certain threshold, the newly added quality problem is considered a duplicate problem and is not included in the problem database. Otherwise, the newly added problem data is stored in the problem database.

[0078] In a further implementation, the presence or absence of keywords in the original question data is used as the basis for judgment. A first and second calculated vector are obtained by converting text vectors into numerical vectors. For example:

[0079] 1. After determining multiple keywords based on their importance scores, arrange the keywords in a fixed order (e.g., in descending order of TF-IDF values) to obtain a text vector (K1, K2, ..., K10) composed of keyword elements K1 to K10;

[0080] 2. Since all keywords have appeared in the new question data, the keyword elements K1 to K10 in the text vector are converted into a vector in which all the numeric elements are the first value, which is used as the first calculation vector. The first value is 1 for example. Therefore, the first calculation vector is (1, 1, ..., 1).

[0081] 3. Iterate through the frequency of each keyword element in the original question data within the text vector. When the frequency of a keyword element in the original question data is zero (i.e., the keyword has not appeared in the current original question data), it is converted into a numeric element with a second value, exemplified by 0. Otherwise, it is converted into a numeric element with a first value to obtain the second calculated vector. Therefore, the second calculated vector is a vector containing either 0 or 1, for example:

[0082] The second computational vector corresponding to the original problem data 1 is: (1, 0, ..., 0);

[0083] The second calculation vector corresponding to the original problem data 2 is: (0, 1, ..., 1);

[0084] ...

[0085] The second computational vector corresponding to the original problem data n is: (0, 0, ..., 1).

[0086] Calculate the cosine similarity between the first computation vector (1, 1, ..., 1) and each second computation vector. When the cosine similarity exceeds a certain value, the newly added quality problem is considered a duplicate problem.

[0087] Based on whether the keyword appears in the original question data, a second calculation vector consisting of 0 or 1 is obtained, and cosine similarity is calculated using this vector, which reduces the computational complexity and thus reduces computational cost.

[0088] Based on the above deduplication process, duplicate issues can be effectively identified and removed, greatly avoiding the output of invalid information to implementers.

[0089] like Figure 2 As shown, a flowchart of the issue database management process is provided. After collecting issue information, the information is organized to obtain new issue data. The new issue data is intelligently deduplicated using a natural language processing model (i.e., the TF-IDF algorithm). If it is determined to be a duplicate issue, the issue is silenced and not included in the issue database. Further, issues that the system determines are not duplicate issues undergo database review and are submitted to engineers for judgment. Valuable issues are included in the issue database, while worthless issues are silenced. Silent issues can be reactivated if their value changes or if they are misjudged.

[0090] The experience base stores information such as failure modes, causes, and effects from past projects, products, or processes. This allows organizations to accumulate valuable knowledge and experience, avoid repeating the same mistakes, and provide a reference for new projects, such as... Figure 3 As shown, one embodiment provides the following experience base management process:

[0091] 1. Empirical Condition Judgment: The problem data in the problem database contains a wide variety of content, and some of the problem data may not meet the empirical conditions. In this embodiment, the text information of the problem data is used to determine whether it contains the cause analysis field and the improvement measure field. If they exist, the problem data can be judged to meet the empirical conditions, and the problem analysis and knowledge transformation will be initiated. Otherwise, no processing will be performed.

[0092] 2. Problem analysis and knowledge transformation: Based on the textual information of problem data that meets the conditions for empirical transformation, use analysis tools to obtain empirical data to build an experience database.

[0093] Analytical tools such as FMEA, TRIZ (Inventive Problem Solving Theory), and management tools analyze and improve the causes of problem failures from both technical and management dimensions. Different analytical tools can provide targeted analysis for different dimensions and different objects.

[0094] Technical Dimension: First, determine the problem category corresponding to the problematic data. When the problem category is structural component quality issues, use FMEA tools to analyze and transform the knowledge from structural component quality issues. FMEA tools assess and analyze potential failure modes in the design, manufacturing, and assembly processes to determine the causes and effects of failures and formulate corresponding preventive and corrective measures. When the problem category is software and hardware quality issues, use TRIZ (Torrent Analysis) tools to analyze and transform the knowledge from software and hardware quality issues. TRIZ analyzes the causal relationships of software and hardware quality issues to identify the root causes and propose solutions. The knowledge accumulated by both tools, including failure chains, preventive measures, and detection measures, can be simultaneously transferred to other production systems to avoid duplication of work.

[0095] Management dimension: Utilize timeline charts and other methods to analyze the causes and escape routes of problems in the process, create or update management process documents, formulate preventive and detection measures, and fundamentally solve problems from a management perspective.

[0096] 3. Team Review. The team reviews the effectiveness of the root cause analysis and knowledge accumulation in both technical and management dimensions. If the review is passed, the knowledge is added to the experience base. If it fails, the team returns to the previous step.

[0097] It should be understood that, although Figures 1-3 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order in which these steps are executed, and they can be performed in other orders. Furthermore, Figures 1-3 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0098] In one embodiment, such as Figure 4 As shown, a quality problem avoidance system is provided, including: a problem database module 210, a problem database module 210, and a special avoidance module 230, wherein:

[0099] Problem database module 210 is used to store and output problem data of quality problems corresponding to parts. The problem database module 210 is established based on the collected text information of quality problems.

[0100] The experience base module 220 is used to store and output the experience data corresponding to the problem data, and the experience data includes at least the preventive measures and detection measures for quality problems.

[0101] The special avoidance module 230 is used to obtain a parts list of the target project, obtain problem data of quality problems corresponding to the parts from the problem database module 210 based on the parts list, and obtain experience data corresponding to the problem data from the experience database module 220 based on the problem data; implement the target project with preventive measures based on the experience data, and determine whether the quality problems corresponding to the parts recur based on the detection measures; when the quality problems corresponding to the parts recur, the avoidance failure result is fed back to the problem database; when the quality problems corresponding to the parts do not recur, the avoidance success result is fed back to the experience database.

[0102] The aforementioned quality problem avoidance system can perform closed-loop management of quality problems. When implementing new projects, it can find corresponding experience data based on quality problems, thereby implementing preventive measures to avoid the recurrence of quality problems. It can also use detection measures to detect the avoidance results and feed the avoidance results back to the problem database or experience database to indicate whether the experience data is effective for subsequent projects.

[0103] In one embodiment, when the problem database module 210 collects new problem data, it performs deduplication on the new problem data based on the existing problem data in the problem database module 210, including:

[0104] Based on the text information of the newly added question data, the importance score of each word in the newly added question data is obtained;

[0105] Multiple keywords are determined based on the importance scores;

[0106] Based on the frequency of occurrence of the keywords in the newly added question data, a first calculation vector is obtained;

[0107] A second calculation vector is obtained based on the frequency of occurrence of the keywords in the original problem data;

[0108] Based on the first and second calculation vectors, a cosine similarity is obtained, and based on the cosine similarity, it is determined whether the newly added question data is duplicated with the original question data.

[0109] When the newly added problem data does not duplicate the existing problem data, the newly added problem data is stored in the problem database module 210.

[0110] In one embodiment, the question base module 210 calculates and obtains the word frequency and inverse document frequency of each word in the newly added question data; and determines the importance score based on the word frequency and inverse document frequency.

[0111] In one embodiment, the question base module 210 arranges the keywords in a fixed order to obtain a text vector composed of keyword elements; converts the keyword elements in the text vector into a vector in which all numeric elements are of the first value as a first calculation vector; iterates through the frequency of occurrence of each keyword element in the text vector in the original question data, and when the frequency of occurrence of the keyword element in the original question data is zero, it converts it into a numeric element of the second value, otherwise it converts it into a numeric element of the first value to obtain a second calculation vector.

[0112] The experience base module 220 determines whether the problem data meets the empirical conditions based on the text information of the problem data, including determining that the empirical conditions are met when the text information of the problem data contains a cause analysis field and an improvement measure field; and obtains the experience data based on the text information of the problem data that meets the empirical conditions to establish the experience base module 220.

[0113] When the experience base module 220 is established, it determines the problem category corresponding to the problem data. The problem category includes structural component quality problems or software and hardware quality problems.

[0114] When the problem data corresponds to the problem category of structural component quality problems, empirical data on structural component quality problems are obtained based on FMEA tool analysis.

[0115] When the problem data corresponds to a software or hardware quality problem, empirical data on software or hardware quality problems is obtained based on TRIZ causal chain analysis.

[0116] The problem categories also include management problems;

[0117] When the problem data corresponds to a management problem category, the experience data is obtained based on the analysis of a timeline chart of management problems.

[0118] For specific limitations regarding the quality problem avoidance system, please refer to the limitations on quality problem avoidance methods mentioned above, which will not be repeated here. Each module in the aforementioned quality problem avoidance system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0119] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 5As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database stores problem data and experience data. The network interface communicates with external terminals via a network connection. When executed by the processor, the computer program implements a quality problem avoidance method.

[0120] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0121] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps:

[0122] Obtain the parts list of the target project, and based on the parts list, obtain the quality problem data corresponding to the parts from the problem database, wherein the problem database is established based on the collected text information of the quality problems;

[0123] Based on the problem data, experience data corresponding to the problem data is obtained from the experience base. The experience data is obtained by transforming the problem data and includes at least preventive and detection measures for quality problems.

[0124] The target project is implemented based on preventive measures according to the empirical data, and the detection measures are used to determine whether the quality problems corresponding to the parts recur.

[0125] When a quality problem corresponding to a part occurs repeatedly, the result of failure to avoid it is fed back to the problem database; when a quality problem corresponding to a part does not occur repeatedly, the result of successful avoidance is fed back to the experience database.

[0126] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0127] The problem database is built based on the collected text information of quality problems, including:

[0128] When new problem data is collected, the new problem data is deduplicated based on the existing problem data in the problem database, including:

[0129] Based on the text information of the newly added question data, the importance score of each word in the newly added question data is obtained;

[0130] Multiple keywords are determined based on the importance scores;

[0131] Based on the frequency of occurrence of the keywords in the newly added question data, a first calculation vector is obtained;

[0132] A second calculation vector is obtained based on the frequency of occurrence of the keywords in the original problem data;

[0133] Based on the first and second calculation vectors, a cosine similarity is obtained, and based on the cosine similarity, it is determined whether the newly added question data is duplicated with the original question data.

[0134] When the newly added problem data does not duplicate the existing problem data, the newly added problem data is stored in the problem database.

[0135] In one embodiment, when the processor executes the computer program, it further implements the following steps: obtaining the importance score of each word in the newly added question data based on the text content of the newly added question data includes:

[0136] Obtain the term frequency and inverse document frequency of each word in the newly added question data;

[0137] The importance score is determined based on the term frequency and inverse document frequency.

[0138] In one embodiment, when the processor executes the computer program, it further implements the following steps: after determining multiple keywords based on the importance scores, the steps include:

[0139] The keywords are arranged in a fixed order to obtain a text vector composed of keyword elements;

[0140] The step of obtaining the first calculation vector includes: converting the keyword elements in the text vector into a vector in which all numeric elements are first values, as the first calculation vector;

[0141] The process of obtaining the second calculation vector includes: traversing the frequency of each keyword element in the text vector in the original question data; when the frequency of the keyword element in the original question data is zero, it is converted into a numeric element with a second value; otherwise, it is converted into a numeric element with a first value, so as to obtain the second calculation vector.

[0142] In one embodiment, when the processor executes the computer program, it further performs the following steps: the empirical data is obtained by transforming the problem data, including:

[0143] Based on the text information of the problem data, determine whether the problem data meets the empirical conditions, including determining that the empirical conditions are met when the text information of the problem data contains a cause analysis field and an improvement measure field;

[0144] The experience data is obtained by using the textual information of the problem data that meets the empirical conditions to establish the experience database.

[0145] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0146] The process of obtaining the experience data to establish the experience base includes:

[0147] Determine the problem category corresponding to the problem data, where the problem category includes structural component quality problems or software and hardware quality problems;

[0148] When the problem data corresponds to the problem category of structural component quality problems, empirical data on structural component quality problems are obtained based on FMEA tool analysis.

[0149] When the problem data corresponds to a software or hardware quality problem, empirical data on software or hardware quality problems is obtained based on TRIZ causal chain analysis.

[0150] The problem categories also include management problems;

[0151] When the problem data corresponds to a management problem category, the experience data is obtained based on the analysis of a timeline chart of management problems.

[0152] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0153] Obtain the parts list of the target project, and based on the parts list, obtain the quality problem data corresponding to the parts from the problem database, wherein the problem database is established based on the collected text information of the quality problems;

[0154] Based on the problem data, experience data corresponding to the problem data is obtained from the experience base. The experience data is obtained by transforming the problem data and includes at least preventive and detection measures for quality problems.

[0155] The target project is implemented based on preventive measures according to the empirical data, and the detection measures are used to determine whether the quality problems corresponding to the parts recur.

[0156] When a quality problem corresponding to a part occurs repeatedly, the result of failure to avoid it is fed back to the problem database; when a quality problem corresponding to a part does not occur repeatedly, the result of successful avoidance is fed back to the experience database.

[0157] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0158] The problem database is built based on the collected text information of quality problems, including:

[0159] When new problem data is collected, the new problem data is deduplicated based on the existing problem data in the problem database, including:

[0160] Based on the text information of the newly added question data, the importance score of each word in the newly added question data is obtained;

[0161] Multiple keywords are determined based on the importance scores;

[0162] Based on the frequency of occurrence of the keywords in the newly added question data, a first calculation vector is obtained;

[0163] A second calculation vector is obtained based on the frequency of occurrence of the keywords in the original problem data;

[0164] Based on the first and second calculation vectors, a cosine similarity is obtained, and based on the cosine similarity, it is determined whether the newly added question data is duplicated with the original question data.

[0165] When the newly added problem data does not duplicate the existing problem data, the newly added problem data is stored in the problem database.

[0166] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0167] The step of obtaining the importance score of each word in the newly added question data based on the text content of the newly added question data includes:

[0168] Obtain the term frequency and inverse document frequency of each word in the newly added question data;

[0169] The importance score is determined based on the term frequency and inverse document frequency.

[0170] In one embodiment, when the computer program is executed by a processor, it further implements the following steps: after determining multiple keywords based on the importance scores, the steps include:

[0171] The keywords are arranged in a fixed order to obtain a text vector composed of keyword elements;

[0172] The step of obtaining the first calculation vector includes: converting the keyword elements in the text vector into a vector in which all numeric elements are first values, as the first calculation vector;

[0173] The process of obtaining the second calculation vector includes: traversing the frequency of each keyword element in the text vector in the original question data; when the frequency of the keyword element in the original question data is zero, it is converted into a numeric element with a second value; otherwise, it is converted into a numeric element with a first value, so as to obtain the second calculation vector.

[0174] In one embodiment, when the computer program is executed by a processor, it further performs the following steps: the empirical data is obtained by transforming the problem data, including:

[0175] Based on the text information of the problem data, determine whether the problem data meets the empirical conditions, including determining that the empirical conditions are met when the text information of the problem data contains a cause analysis field and an improvement measure field;

[0176] The experience data is obtained by using the textual information of the problem data that meets the empirical conditions to establish the experience database.

[0177] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0178] The process of obtaining the experience data to establish the experience base includes:

[0179] Determine the problem category corresponding to the problem data, where the problem category includes structural component quality problems or software and hardware quality problems;

[0180] When the problem data corresponds to a structural component quality problem, the empirical data is obtained based on FMEA tool analysis;

[0181] When the problem data corresponds to a software or hardware quality issue, the empirical data is obtained based on TRIZ causal chain analysis.

[0182] The problem categories also include management problems;

[0183] When the problem data corresponds to a management problem category, the experience data is obtained based on the analysis of a timeline chart of management problems.

[0184] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0185] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0186] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for avoiding quality problems, characterized in that, include: Obtain the parts list for the target project, and based on the parts list, obtain the issue data of the quality issues corresponding to the parts from the issue database; Based on the problem data, experience data corresponding to the problem data is obtained from the experience base. The experience data is obtained by transforming the problem data and includes at least preventive and detection measures for quality problems. The target project is implemented based on preventive measures according to the empirical data, and the detection measures are used to determine whether the quality problems corresponding to the parts recur. When a quality problem corresponding to a part occurs repeatedly, the result of failure to avoid it is fed back to the problem database; when a quality problem corresponding to a part does not occur repeatedly, the result of successful avoidance is fed back to the experience database. The problem database is established based on the collected text information of quality problems and includes: When new problem data is collected, the new problem data is deduplicated based on the existing problem data in the problem database, including: Based on the text information of the newly added question data, the importance score of each word in the newly added question data is obtained; Multiple keywords are determined based on the importance scores; Based on the frequency of occurrence of the keywords in the newly added question data, a first calculation vector is obtained; A second calculation vector is obtained based on the frequency of occurrence of the keywords in the original problem data; Based on the first and second calculation vectors, a cosine similarity is obtained, and based on the cosine similarity, it is determined whether the newly added question data is duplicated with the original question data. When the newly added problem data does not duplicate the existing problem data, the newly added problem data is stored in the problem database.

2. The quality problem avoidance method according to claim 1, characterized in that, The step of obtaining the importance score of each word in the newly added question data based on the text content of the newly added question data includes: Obtain the term frequency and inverse document frequency of each word in the newly added question data; The importance score is determined based on the term frequency and inverse document frequency.

3. The quality problem avoidance method according to claim 1, characterized in that, After determining multiple keywords based on the importance score, the process includes: The keywords are arranged in a fixed order to obtain a text vector composed of keyword elements; The step of obtaining the first calculation vector includes: converting the keyword elements in the text vector into a vector in which all numeric elements are first values, as the first calculation vector; The process of obtaining the second calculation vector includes: traversing the frequency of each keyword element in the text vector in the original question data; when the frequency of the keyword element in the original question data is zero, it is converted into a numeric element with a second value; otherwise, it is converted into a numeric element with a first value, so as to obtain the second calculation vector.

4. The quality problem avoidance method according to claim 1, characterized in that, The empirical data is obtained by transforming the problem data, including: Based on the text information of the problem data, determine whether the problem data meets the empirical conditions, including determining that the empirical conditions are met when the text information of the problem data contains a cause analysis field and an improvement measure field; The experience data is obtained by using the textual information of the problem data that meets the empirical conditions to establish the experience database.

5. The quality problem avoidance method according to claim 4, characterized in that, The process of obtaining the experience data to establish the experience base includes: Determine the problem category corresponding to the problem data, where the problem category includes structural component quality problems or software and hardware quality problems; When the problem data corresponds to a structural component quality problem, the empirical data is obtained based on FMEA tool analysis; When the problem data corresponds to a software or hardware quality issue, the empirical data is obtained based on TRIZ causal chain analysis.

6. The quality problem avoidance method according to claim 5, characterized in that, The problem categories also include management problems; The process of obtaining the experience data to establish the experience base also includes: When the problem data corresponds to a management problem category, the experience data is obtained based on the analysis of a timeline chart of management problems.

7. A quality problem avoidance system, characterized in that, The system includes: The problem database module is used to store and output problem data for the quality problems corresponding to the parts. The problem database module is established based on the collected text information of the quality problems. The experience base module is used to store and output the experience data corresponding to the problem data, and the experience data includes at least the preventive measures and detection measures for quality problems. A specialized avoidance module is used to obtain a parts list for a target project, and based on the parts list, obtain problem data for the quality problems corresponding to the parts from a problem database module, and obtain experience data corresponding to the problem data from an experience database module; implement preventive measures for the target project based on the experience data, and determine whether the quality problems corresponding to the parts recur based on the detection measures; when the quality problems corresponding to the parts recur, the module feeds back the avoidance failure result to the problem database; when the quality problems corresponding to the parts do not recur, the module feeds back the avoidance success result to the experience database. When the problem database module collects new problem data, it performs deduplication on the new problem data based on the existing problem data in the problem database module, including: Based on the text information of the newly added question data, the importance score of each word in the newly added question data is obtained; Multiple keywords are determined based on the importance scores; Based on the frequency of occurrence of the keywords in the newly added question data, a first calculation vector is obtained; A second calculation vector is obtained based on the frequency of occurrence of the keywords in the original problem data; Based on the first and second calculation vectors, a cosine similarity is obtained, and based on the cosine similarity, it is determined whether the newly added question data is duplicated with the original question data. When the newly added problem data does not duplicate the existing problem data, the newly added problem data is stored in the problem database module.

8. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

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

Images