Industrial chain quality map generation method

Abstract

The invention relates to the technical field of atlas data processing, and particularly discloses an industrial chain quality atlas generation method, which is beneficial to understanding the mutual influence and dependency relationship among the production elements of the production links in an industrial chain by analyzing the internal association degree of the production elements of the production links to which the industrial chain belongs. Therefore, an industrial chain atlas with relatively high reliability is generated, the influence weight index of each quality problem index of each production element of each production link to which the industrial chain atlas belongs is evaluated, and each quality problem key index with greater influence on the industrial chain is accurately identified; and finally, carrying out classification processing on each quality problem key index of each production element of each production link to which the industrial chain atlas belongs, thereby generating the industrial chain quality atlas, and guiding the generation of the industrial chain quality atlas and accurately examining the quality problem of the industrial chain by finely analyzing multi-dimensional parameters and focusing on the core problem of the industrial chain. And powerful data support is provided for optimizing industrial chain configuration.

Description

Technical Field

[0001] The present invention relates to the technical field of graph data processing, and in particular to a method for generating an industrial chain quality graph. Background Art

[0002] Graphs play an important role in data analysis and intelligent decision support. However, traditional graph generation methods usually rely on simple statistical analysis and drawing, without deeply exploring the complex relationships and dependencies between the elements involved in the graph. At the same time, traditional graph generation methods fail to effectively integrate and display the derivative problems of the elements belonging to the graph, resulting in limitations in the identification of problems and the formulation of solution strategies during the use of graphs.

[0003] For example, the invention patent with announcement number CN111178940B is a method and system for automatically generating a sales speech map, wherein the method includes: obtaining conversation records between sales personnel and customers; determining a conversation event sequence based on the conversation records; determining the causal relationship between different types of events and the causal probability corresponding to the causal relationship based on the conversation event sequence; generating a sales speech map based on the causal relationship and causal probability; the above-mentioned application embodiment can automatically generate a sales speech map based on sales records, without the need for sales experts to manually summarize sales experience, thereby improving the accuracy and efficiency of sales evaluation, and sales personnel can improve the order conversion rate based on the sales speech map.

[0004] For example, the invention patent with announcement number CN116432965B involves a job capability analysis method and a tree diagram generation method based on a knowledge graph. The job capability analysis method includes crawling job information and company information; performing data cleaning on job information and company information to obtain recruitment information; processing the recruitment information according to preset data processing rules to obtain entity information and relationship information; converting entity information and relationship information into nodes and edges to construct a knowledge graph; identifying key job information based on the knowledge graph, and constructing a job capability model using the key job information and the knowledge graph; constructing a job capability model based on the knowledge graph to more accurately and systematically organize and summarize the key skills and capability requirements required for different positions, and provide guidance for the capabilities of relevant personnel.

[0005] However, in the process of implementing the embodiments of the present application, it was found that the above-mentioned technology has at least the following technical problems: the existing graph generation technology lacks clear graph construction guidelines in the process of constructing the graph, which will lead to deviations in information extraction and processing in the graph, affecting the accuracy and completeness of the graph, and further affecting the practicality and decision support effect of the generated graph. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention provides a method for generating an industrial chain quality map, which can effectively solve the problems involved in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for generating an industrial chain quality map, comprising: S1. acquiring the process parameters of each production factor of each production link belonging to the industrial chain, analyzing the internal correlation of each production factor of each production link belonging to the industrial chain, thereby generating an industrial chain map; S2. collecting each quality problem indicator of each production factor of each production link belonging to the industrial chain map, evaluating the influence weight index of each quality problem indicator of each production factor of each production link belonging to the industrial chain map, thereby extracting each key indicator of quality problems of each production factor of each production link belonging to the industrial chain map; S3. classifying and processing each key indicator of quality problems of each production factor of each production link belonging to the industrial chain map, thereby generating an industrial chain quality map.

[0008] As a further method, the key indicators of quality problems of each production factor in each production link belonging to the industrial chain map are extracted. The specific analysis process is: the influence weight index of each quality problem indicator of each production factor in each production link belonging to the industrial chain map is compared with the influence weight threshold stored in the industrial chain map database, and the quality problem indicators of each production factor in each production link belonging to the industrial chain map whose influence weight index is greater than the influence weight threshold are screened out, and marked as the key indicators of quality problems of each production factor in each production link belonging to the industrial chain map.

[0009] As a further method, the key indicators of quality problems of each production factor in each production link belonging to the industrial chain map are classified and processed. Specifically, the problem keyword adaptation index between the key indicators of quality problems of each production factor in each production link belonging to the industrial chain map and each problem classification label is analyzed, thereby classifying and processing the key indicators of quality problems of each production factor in each production link belonging to the industrial chain map.

[0010] As a further method, the industrial chain quality map is generated, and the specific analysis process is: after classifying and processing the key indicators of quality problems of each production factor in each production link belonging to the industrial chain map, they are automatically filled into the corresponding problem classification labels in the industrial chain map, thereby generating the industrial chain quality map.

[0011] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:

[0012] (1) The present invention provides a method for generating an industrial chain quality map, analyzes the internal correlation of each production factor in each production link of the industrial chain, helps to understand the mutual influence and dependence between each production factor in each production link in the industrial chain, thereby generating an industrial chain map with strong reliability, and evaluates the influence weight index of each quality problem indicator of each production factor in each production link of the industrial chain map, accurately identifies each key indicator of quality problems that have a greater impact on the industrial chain, and finally classifies each key indicator of quality problems of each production factor in each production link of the industrial chain map, thereby generating an industrial chain quality map, through refined analysis of multi-dimensional parameters, focusing on the core issues of the industrial chain, guiding the generation of the industrial chain quality map, accurately examining the quality issues of the industrial chain, and providing strong data support for optimizing the configuration of the industrial chain.

[0013] (2) The present invention analyzes the number of improvement plans, adjustability and corresponding production factor position node connectivity of each quality problem indicator of each production factor in each production link of the industrial chain map within the production inspection cycle, thereby accurately reflecting the impact of each quality problem indicator on the overall operation of the industrial chain, thereby accurately identifying key quality problems and helping to guide the generation of the industrial chain quality map more reasonably and effectively.

[0014] (3) The present invention analyzes the keyword co-occurrence and average correlation level between the key indicators of each quality problem of each production factor in each production link of the industrial chain map and each problem classification label, thereby classifying the key indicators of each quality problem of each production factor in each production link of the industrial chain map, providing a clearer and more definite decision-making basis for improving the operating efficiency of the industrial chain. According to the classification results, more targeted improvement measures and optimization strategies can be formulated for the quality problems of the industrial chain to improve the effectiveness of decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0016] Figure 1 It is a schematic diagram of the method flow of the present invention.

[0017] Figure 2 It is a schematic diagram of the quality map (part) of the integrated circuit industry chain involved in the present invention. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0019] Reference Figure 1 As shown, the present invention provides a method for generating an industrial chain quality map, including: S1. acquiring the process parameters of each production factor of each production link of the industrial chain, analyzing the internal correlation of each production factor of each production link of the industrial chain, and thereby generating an industrial chain map; the above-mentioned process parameters of each production factor of each production link of the industrial chain can be extracted by recording the production logs of each production factor of each production link.

[0020] In a specific embodiment, the present invention analyzes the keyword co-occurrence and average correlation level between the key indicators of each quality problem of each production factor in each production link of the industrial chain map and each problem classification label, thereby classifying the key indicators of each quality problem of each production factor in each production link of the industrial chain map, providing a clearer and more definite decision-making basis for improving the operation efficiency of the industrial chain, and according to the classification results, more targeted improvement measures and optimization strategies can be formulated for the quality problems of the industrial chain to improve the effectiveness of decision-making.

[0021] The internal correlation of each production factor of each production link in the industrial chain is sorted in order from large to small, and the internal correlation order of each production factor of each production link in the industrial chain is obtained, thereby generating an industrial chain map; it should be explained that the greater the average internal correlation of each production link in the industrial chain, the greater the contribution of the production factor of the production link in the industrial chain to the industrial chain, which usually means that the production factor of the production link in the industrial chain is in the upstream position of the industrial chain or has a greater impact on the entire industrial chain. Therefore, the production factor of the production link in the industrial chain is placed in a priority position in the industrial chain map, and sorting the production factors of each production link in the industrial chain means sorting the contribution of each production factor of each production link in the industrial chain; the above-mentioned industrial chain map is generated, and the map builder can use graphic visualization software to generate and process the production factors of each production link in the industrial chain that have been sorted.

[0022] Furthermore, the internal correlation of each production factor in each production link of the industrial chain is analyzed, and the specific analysis process is: according to the process parameters of each production factor in each production link of the industrial chain, the number of occurrences of each production factor in each production link of the industrial chain is extracted, and the above number of occurrences is extracted from the production log recording each production factor in each production link.

[0023] The data coupling degrees corresponding to each occurrence frequency interval stored in the industrial chain map database are matched, thereby obtaining the data coupling degrees of each production factor in each production link of the industrial chain; the above-mentioned industrial chain map database refers to a database that stores data required to generate the industrial chain quality map; the above-mentioned data coupling degrees corresponding to each occurrence frequency interval stored in the industrial chain map database are matched, which means that each occurrence frequency interval corresponds to the data coupling degree, and each occurrence frequency interval that matches the occurrence frequency of each production factor in each production link of the industrial chain is found. The data coupling degree corresponding to the occurrence frequency interval is the data coupling degree of each production factor in each production link of the industrial chain; the above-mentioned data coupling degree represents the numerical value of the tightness of transmitted data, which is formulated by the map builder based on historical map data.

[0024] The initial time point corresponding to each occurrence of each production factor in each production link of the industrial chain is located, and the difference processing is performed with the obtained end time point corresponding to each occurrence of each production factor in each production link of the industrial chain to obtain the data support duration of each production factor in each production link of the industrial chain; the above-mentioned positioning of the initial time point corresponding to each occurrence of each production factor in each production link of the industrial chain refers to locating the timestamp of each occurrence of each production factor in each production link of the industrial chain from the production log recording each production factor in each production link, and obtaining the occurrence time point in the timestamp, which is to locate the initial time point corresponding to each occurrence of each production factor in each production link of the industrial chain; the method for obtaining the end time point corresponding to each occurrence of each production factor in each production link of the industrial chain is the same as the method for obtaining the initial time point corresponding to each occurrence of each production factor in each production link of the industrial chain.

[0025] The product output of the production equipment corresponding to each production factor in each production link of the industrial chain within the production inspection cycle is extracted from the process parameters of each production factor in each production link of the industrial chain, and the ratio is processed with the duration corresponding to the production inspection cycle to obtain the production speed of the production equipment corresponding to each production factor in each production link of the industrial chain; the above-mentioned production inspection cycle represents the cycle for inspecting the performance of the industrial chain, and the map builder determines the cycle duration according to the actual situation; the above-mentioned acquisition of the product output of the production equipment corresponding to each production factor in each production link of the industrial chain within the production inspection cycle refers to extracting from the production log of the production equipment.

[0026] The internal correlation of each production factor in each production link of the industrial chain is obtained by comprehensive analysis. In this embodiment, the internal correlation of each production factor in each production link of the industrial chain is obtained by comprehensive analysis of the data coupling degree, data support time and production speed of the corresponding production equipment of each production factor in each production link of the industrial chain. It is used to determine the value of the internal correlation degree of each production factor in each production link of the industrial chain. The specific expression is:

[0027]

[0028] Among them, σ D ab It is the internal correlation of the bth production factor of the ath production link in the industrial chain. The longer the data support time is, the higher the degree of dependence of the industrial chain on the data of the production factor of the production link, and at the same time, the data coupling degree of the production factor of the production link is higher. At the same time, if the production speed of the production equipment corresponding to a production factor of a production link is faster, it means that the production factor of the production link is the infrastructure in the industrial chain, which has a significant and strong impact on the overall operation and efficiency of the industrial chain, and jointly improves the internal correlation. Therefore, smooth analysis of the three parameters can deeply understand the operation mechanism and data dependency of the industrial chain, optimize the data circulation path of the industrial chain, and provide accurate data management strategies for the generation of industrial chain maps.

[0029] CS ab It is the data coupling degree of the bth production factor of the ath production link in the industrial chain. A high data coupling degree means that the data of this production factor is frequently exchanged and shared in other production links in the industrial chain, and other production links have a strong dependence on this data.

[0030] DSD ab It is the data support time of the bth production factor of the ath production link in the industrial chain. The longer the data support time is, the more deeply this production factor is involved in the formulation and execution of many important decisions in the industrial chain. Therefore, this production factor has an important position in the industrial chain.

[0031] IF ab It is the production speed of the production equipment corresponding to the bth production factor of the ath production link in the industrial chain. The faster the production speed, the simpler and more efficient the operation process of this production factor in the industrial chain is. This lays the foundation for the stable operation of the industrial chain and has a significant impact on the overall operation and efficiency of the industrial chain.

[0032] a is the number of each production link, a = {1, 2, 3, ..., t}, t is the total number of production links, b is the number of each production factor, b = {1, 2, 3, ..., p}, p is the total number of production factors.

[0033] z 1 It is the impact factor corresponding to the unit value of data coupling degree predefined in the industrial chain map database, which represents the degree of influence of the unit value of data coupling degree on the internal correlation. The fitting curve corresponding to the data coupling degree is obtained by fitting the relationship between the historical data importance index, the historical data flow frequency and the data coupling degree, and the fitting curve corresponding to the data coupling degree is introduced according to the real-time data importance index and the data flow frequency, thereby obtaining the impact factor corresponding to the unit value of the data coupling degree in this example. The value range of this value in this example is 0 to 1.

[0034] z 2 It is the impact factor corresponding to the unit value of data support time predefined in the industrial chain map database, which represents the degree of influence of the unit value of data support time on the internal correlation. The fitting curve corresponding to the data support time is obtained by fitting the relationship between the historical data exception processing time, the historical data response time and the data support time, and the fitting curve corresponding to the data support time is brought into the fitting curve according to the real-time data exception processing time and the data response time, thereby obtaining the impact factor corresponding to the unit value of the data support time in this example. The value range of this value in this example is 0 to 1.

[0035] z 3 It is the impact factor corresponding to the unit value of production speed predefined in the industrial chain map database, which represents the degree of influence of the unit value of production speed on the internal correlation. The fitting curve corresponding to the production speed is obtained by fitting the relationship between the historical market demand order volume, the historical raw material cleanliness and the production speed. The fitting curve corresponding to the production speed is introduced into the fitting curve according to the real-time market demand order volume and the raw material cleanliness, thereby obtaining the impact factor corresponding to the unit value of production speed in this example. The value range of this value in this example is 0 to 1.

[0036] S2. Collect the quality problem indicators of each production factor in each production link of the industrial chain map, evaluate the impact weight index of each quality problem indicator of each production factor in each production link of the industrial chain map, and extract the key indicators of each quality problem of each production factor in each production link of the industrial chain map; the above-mentioned collection of quality problem indicators of each production factor in each production link of the industrial chain map refers to the map builder collecting the quality problem indicators of each production factor in each production link of the industrial chain map through industry forums, official websites and other channels based on the historical problems of each production factor in each production link of the industrial chain map.

[0037] In a specific embodiment, the present invention analyzes the number of improvement plans, adjustability and corresponding production factor position node connectivity of each quality problem indicator of each production factor in each production link of the industrial chain map within the production inspection cycle, thereby accurately reflecting the degree of influence of each quality problem indicator on the overall operation of the industrial chain, thereby accurately identifying key quality problems and helping to guide the generation of the industrial chain quality map more reasonably and effectively.

[0038] Specifically, the quality problem indicators of each production factor in each production link belonging to the industrial chain map specifically include each occurrence time point of each quality problem indicator of each production factor in each production link belonging to the industrial chain map within the production inspection cycle; the above-mentioned occurrence time points are extracted from the production log that records each production factor in each production link, and each quality problem indicator may appear 0 times or several times within the production inspection cycle. For example, the problem indicator that the packaging materials required for high-end chips such as 5G rely on imports appears once in the production inspection cycle, and there is no corresponding time point for being resolved within the production inspection cycle.

[0039] Specifically, the extraction of key indicators of quality problems of each production factor in each production link of the industrial chain map, the specific analysis process is: compare the influence weight index of each quality problem indicator of each production factor in each production link of the industrial chain map with the influence weight threshold stored in the industrial chain map database, screen out the quality problem indicators of each production factor in each production link of the industrial chain map whose influence weight index is greater than the influence weight threshold, and mark them as key indicators of quality problems of each production factor in each production link of the industrial chain map; the above-mentioned influence weight threshold refers to the minimum value set in the process of evaluating the influence weight index, which is a value formulated by the map builder based on the historical map construction work; it needs to be explained that the above-mentioned screened out influence weight index The number of quality problem indicators of each production factor in each production link of the industrial chain map is greater than the impact weight threshold. If the influence weight index of a certain quality problem indicator of each production factor in each production link of the industrial chain map is greater than the impact weight threshold, it indicates that the quality problem indicator is difficult to solve and may be a historical problem that is difficult to find a solution in a short time. Therefore, it is necessary to screen out the quality problem indicator and put it into the industrial chain map to guide the improvement direction of the industrial chain. If the influence weight index of a certain quality problem indicator of each production factor in each production link of the industrial chain map is less than or equal to the impact weight threshold, it indicates that the quality problem indicator can be solved in a short time and the impact on the industrial chain is small. Therefore, it can be omitted.

[0040] Locate and count the number of post-production factor position nodes and the number of types of post-production factor position nodes corresponding to each quality problem indicator of each production factor in each production link of the industrial chain map, and the connectivity factor corresponding to the number of post-production factor position nodes and the connectivity factor corresponding to the number of types of post-production factor position nodes preset in the industrial chain map database, and perform product analysis respectively, and accumulate them to obtain the production factor position node connectivity corresponding to each quality problem indicator of each production factor in each production link of the industrial chain map. It needs to be explained that the production factor position node connectivity corresponding to each quality problem indicator of each production factor in each production link of the industrial chain map refers to the connectivity factor corresponding to the number of post-production factor position nodes and the number of post-production factor position nodes of each production factor position node in each production link of the industrial chain map. The result of the multiplication of the two sub-items plus the result of the multiplication of the number of types of post-production factor position nodes and the connectivity factor corresponding to the number of types of post-production factor position nodes. For example, the number of post-production factor position nodes of a certain production factor position node is 4, and the result of multiplying it by the connectivity factor 0.09 corresponding to the number of post-production factor position nodes is 0.36. The number of types of post-production factor position nodes of this production factor position node is 2, and the result of multiplying it by the connectivity factor 0.05 corresponding to the number of types of post-production factor position nodes is 0.1. Then the connectivity of this production factor position node is the sum of 0.36 and 0.1, which is 0.46. The connectivity factor corresponding to the number of post-production factor position nodes and the connectivity factor corresponding to the number of types of post-production factor position nodes are formulated by the map builder based on the historical map construction work experience, for example Figure 2 As shown in the integrated circuit industry chain quality map (part), the integrated circuit manufacturing (wafer processing) production link is displayed. The post-production factor position nodes of the photoresist coating node include exposure, etching, doping deposition, detection and others not listed. The number of these post-production factor position nodes is counted, which is the number of post-production factor position nodes of the photoresist coating position node of the integrated circuit manufacturing (wafer processing) production link of the industrial chain map. Exposure, etching, doping deposition and detection are all action plans. The number of types of these four post-production factor position nodes is 1. The types of post-production factor position nodes also include problem judgment types, resource supply types and quality control types.

[0041] The number of time points at which each quality problem indicator of each production factor in each production link of the industrial chain map appears within the production inspection cycle is counted, and the ratio is processed with the number of time points at which each quality problem indicator of each production factor in each production link of the obtained industrial chain map is resolved within the production inspection cycle, so as to obtain the adjustability of each quality problem indicator of each production factor in each production link of the industrial chain map within the production inspection cycle; the number of time points at which each quality problem indicator of each production factor in each production link of the above industrial chain map is resolved within the production inspection cycle is extracted from the production logs recording each production factor in each production link. If a quality problem indicator of each production factor in each production link of the industrial chain map has no number of resolved time points within the production inspection cycle, then the adjustability of the quality problem indicator within the production inspection cycle is 0.

[0042] Obtain the number of improvement plans for each quality problem indicator of each production factor in each production link belonging to the industrial chain map within the production inspection cycle, and obtain the impact weight index of each quality problem indicator of each production factor in each production link belonging to the industrial chain map through comprehensive evaluation; the number of improvement plans for each quality problem indicator of each production factor in each production link belonging to the above industrial chain map within the production inspection cycle can be extracted from the report recording the quality problem indicator improvement plan.

[0043] Furthermore, the influence weight index of each quality problem indicator of each production factor in each production link of the industrial chain map, in this embodiment, the influence weight index of each quality problem indicator of each production factor in each production link of the industrial chain map is obtained by comprehensively analyzing the number of improvement plans, adjustability and corresponding production factor position node connectivity of each quality problem indicator of each production factor in each production link of the industrial chain map within the production detection cycle, and is used to comprehensively determine the numerical value of the influence degree of each quality problem indicator of each production factor in each production link of the industrial chain map on the industrial chain. The specific expression is:

[0044]

[0045] Among them, μ abnIt is the influence weight index of the nth quality problem indicator of the bth production factor of the ath production link belonging to the industrial chain map. If the number of improvement plans for a problem indicator is small, the adjustability of the quality problem indicator is small, indicating that the quality problem indicator is difficult to adjust within the production inspection cycle, thereby increasing the influence weight index of the problem indicator. If the node connectivity of the production factor position corresponding to a quality problem indicator is large, it indicates that the production factor position to which the quality problem indicator belongs occupies a relatively core or key position in the industrial chain map, and the impact of the quality problem indicator on the production chain is also significant. Therefore, integrating and quantitatively analyzing the above parameters can accurately identify the key to the industrial chain problem, invest limited resources in key issues and important links, achieve efficient use of resources, and help optimize the industrial chain process and promote the progress of the industrial chain.

[0046] AJ abn It is the production factor position node connectivity corresponding to the nth quality problem indicator of the bth production factor in the ath production link of the industrial chain map. Production factors with high production factor position node connectivity can quickly spread the negative impact of quality problem indicators to multiple links of the industrial chain.

[0047] NP abn It is the adjustability of the nth quality problem indicator of the bth production factor of the ath production link in the industrial chain map within the production inspection cycle. A smaller adjustability indicates that it is difficult to respond to and solve the quality problem indicator quickly, increasing the risk of industrial chain interruption and loss.

[0048] AT abn It is the number of improvement plans for the nth quality problem indicator of the bth production factor of the ath production link in the industrial chain map within the production inspection cycle. Fewer improvement plans reflects that the root cause of the quality problem indicator is more complicated and it is difficult to find an effective solution directly, so the impact weight index is increased.

[0049] a is the number of each production link, a = {1,2,3,...,t}, t is the total number of production links, b is the number of each production factor, b = {1,2,3,...,p}, p is the total number of production factors, n is the number of each quality problem indicator, n = {1,2,3,...,m}, m is the total number of quality problem indicators.

[0050] mk 1It is the impact factor corresponding to the unit value of the production factor position node connectivity predefined in the industrial chain map database, which represents the degree of influence of the unit value of the production factor position node connectivity on the impact weight index. The fitting curve corresponding to the production factor position node connectivity is obtained by fitting the relationship between the number of historical post-production factor position nodes, the number of historical post-production factor position node types and the production factor position node connectivity. The fitting curve corresponding to the production factor position node connectivity is brought into the fitting curve according to the real-time number of post-production factor position nodes and the number of post-production factor position node types, thereby obtaining the impact factor corresponding to the unit value of the production factor position node connectivity in this example. The value range of this value in this example is 0 to 1.

[0051] mk 2 It is the impact factor corresponding to the adjustable degree unit value predefined in the industrial chain map database, which represents the degree of influence of the adjustable degree unit value on the impact weight index. The fitting curve corresponding to the adjustable degree is obtained by fitting the relationship between the resolution rate of historical quality problem indicators, the occurrence rate of historical quality problem indicators and the adjustability. The real-time resolution rate of quality problem indicators and the occurrence rate of quality problem indicators are brought into the fitting curve corresponding to the adjustability, thereby obtaining the impact factor corresponding to the adjustable degree unit value in this example. The value range of this value in this example is 0 to 1.

[0052] mk 3 It is the impact factor corresponding to the unit value of the number of improvement plans predefined in the industrial chain map database, which represents the degree of influence of the unit value of the number of improvement plans on the impact weight index. The fitting curve corresponding to the number of improvement plans is obtained by fitting the relationship between the historical improvement plan output speed, the historical improvement plan success rate and the number of improvement plans, and the fitting curve corresponding to the number of improvement plans is brought into the fitting curve according to the real-time improvement plan output speed and the improvement plan success rate, thereby obtaining the impact factor corresponding to the unit time length of the number of improvement plans in this example. The value range of this value in this example is 0 to 1.

[0053] In this exemplary embodiment, the influence weight index of each quality problem indicator of the bth production factor of the ath production link of the above industrial chain map and its corresponding parameter change table are shown in Table 1:

[0054] Table 1 The influence weight index of each quality problem indicator of the bth production factor in the ath production link of the industrial chain map and the change of its corresponding parameters

[0055]

[0056] In this example embodiment, the value of the impact factor corresponding to the unit value of the production factor location node connectivity is 0.7, the value of the impact factor corresponding to the unit value of the adjustability is 0.8, and the value of the impact factor corresponding to the unit value of the number of improvement plans is 0.9. Table 1 also reflects the positive correlation between the adjustability and the number of improvement plans, which means that when the adjustability is higher, there are usually more improvement plans to choose from. The impact weight index of the fourth quality problem indicator is 117%, and it is greater than the impact weight threshold, indicating that the quality problem indicator has a greater impact on the industrial chain and needs to be marked as a key quality problem indicator.

[0057] S3. Classify and process the key indicators of quality issues of each production factor in each production link of the industrial chain map, thereby generating an industrial chain quality map.

[0058] Specifically, the key indicators of quality problems of each production factor in each production link belonging to the industrial chain map are classified and processed, specifically, the problem keyword adaptation index between each key indicator of quality problems of each production factor in each production link belonging to the industrial chain map and each problem classification label is analyzed, thereby classifying and processing the key indicators of quality problems of each production factor in each production link belonging to the industrial chain map; it needs to be explained that the specific classification processing process is: sorting the problem keyword adaptation index between a certain key indicator of quality problems of a certain production factor in a certain production link belonging to the industrial chain map and each problem classification label in order from small to large, extracting the problem classification label corresponding to the first-ranked key index of problems, then a certain key indicator of quality problems of a certain production factor in a certain production link belonging to the industrial chain map belongs to the problem classification label.

[0059] Specifically, the specific analysis process of generating the industrial chain quality map is as follows: after classifying and processing the key indicators of the quality problems of each production factor in each production link of the industrial chain map, the key indicators are automatically filled into the corresponding problem classification labels in the industrial chain map, thereby generating the industrial chain quality map; it should be explained that the above-mentioned automatic filling into the corresponding problem classification labels in the industrial chain map refers to filling the key indicators of the quality problems after the classification processing into the problem classification label area reserved in the industrial chain map through the automatic filling algorithm, such as Figure 2 Schematic diagram of the integrated circuit industry chain quality map (part), which is a part of the final generated integrated circuit industry chain quality map. If the problem classification label area is not filled, it will be a schematic diagram of the integrated circuit industry chain map (part).

[0060] Furthermore, the problem keyword adaptation index between the key indicators of each quality problem of each production factor in each production link of the industrial chain map and each problem classification label is specifically analyzed as follows: extracting each keyword of each problem classification label from the industrial chain map database, and each keyword of each problem classification label refers to the definition of each problem classification label collected from the document materials related to the industrial chain. Each problem classification label may have several definitions, and then the keywords of each problem classification label are extracted from the definitions of each problem classification label through the word frequency-inverse document frequency method; the name of each key indicator of quality problem refers to the text describing each key indicator of quality problem.

[0061] At the same time, locate each keyword of the key indicator name of each quality problem of each production factor in each production link of the industrial chain map, and compare them with the keywords of each problem classification label respectively. Count the number of keyword overlaps between each keyword of the key indicator name of each quality problem of each production factor in each production link of the industrial chain map and each keyword of each problem classification label, and perform ratio processing with the total number of keywords of each problem classification label to obtain the keyword co-occurrence degree between each key indicator of each quality problem of each production factor in each production link of the industrial chain map and each problem classification label. It should be explained that the above-mentioned overlaps with each problem classification label are The keyword overlap comparison of each keyword of the class label refers to using the set operation algorithm in the programming language to convert each keyword of each problem classification label into a set, and also converting each keyword of the key indicator name of each quality problem of each production factor in each production link of the industrial chain map into a set, and calculating the intersection of each keyword set of the key indicator name of each quality problem of each production factor in each production link of the industrial chain map with each keyword set of each problem classification label, so as to obtain the overlapped text of each keyword of each problem classification label and each key indicator name of each quality problem of each production factor in each production link of the industrial chain map.

[0062] The correlation levels of the keywords of the key indicators of each quality problem of each production factor in each production link of the industrial chain map and the keywords of each problem classification label are matched from the industrial chain map database, and the average correlation level between the key indicators of each quality problem of each production factor in each production link of the industrial chain map and each problem classification label is obtained by mean processing; the above correlation level refers to the map builder judging the semantic similarity of words based on the semantic dictionary-based method in the semantic similarity calculation, and then formulating the correlation level based on the actual situation. For example, if the semantic similarity is 80%, the correlation level is 8. The higher the correlation level, the closer the meanings expressed by the two words are.

[0063] The comprehensive analysis thus obtained is the problem keyword adaptation index of each key indicator of each quality problem of each production factor in each production link of the industrial chain map and each problem classification label. In this embodiment, the problem keyword adaptation index is obtained by comprehensive analysis of the keyword co-occurrence and average correlation level of each key indicator of each quality problem of each production factor in each production link of the industrial chain map and each problem classification label. It is used to evaluate the value of the degree of adaptation of each key indicator of each quality problem of each production factor in each production link of the industrial chain map and each problem classification label. The specific expression is:

[0064]

[0065] Among them, Φ abs_v It is the problem keyword adaptation index between the sth quality problem key indicator of the bth production factor of the ath production link and the vth problem classification label. The average correlation level is low, indicating that the semantics between the quality problem key indicator and the problem classification label are quite different, resulting in weak correlation between the two, which makes the keyword co-occurrence low, and together reduces the problem keyword adaptation index. By quantitatively analyzing the problem keyword adaptation index, keyword co-occurrence and average correlation level, the accuracy and reliability of the quality problem indicator classification are improved, which can reduce subjective assumptions and misjudgments in the classification process.

[0066] C_MK abs_v It is the keyword co-occurrence degree between the sth quality problem key indicator of the bth production factor of the ath production link and the vth problem classification label of the industrial chain map, which indicates the keyword overlap ratio of the quality problem indicator and the problem classification label. If the keyword co-occurrence degree is high, it means that the meanings expressed by the two are similar.

[0067] E_NS abs_v It is the average correlation level between the sth quality problem key indicator of the bth production factor of the ath production link belonging to the industrial chain map and the vth problem classification label. The average correlation level is a quantitative value of the similarity between keywords. If two keywords are very similar in semantics or often appear in the same context, their average correlation level may be very high.

[0068] a is the number of each production link, a={1,2,3,...,t}, t is the total number of production links, b is the number of each production factor, b={1,2,3,...,p}, p is the total number of production factors, s is the number of each key indicator of quality problems, s={1,2,3,...,g}, g is the total number of key indicators of quality problems, v is the number of each problem classification label, v={1,2,3,...,d}, d is the total number of problem classification labels.

[0069] zl 1 It is the impact factor corresponding to the unit value of keyword co-occurrence predefined in the industrial chain map database, which represents the degree of influence of the unit value of keyword co-occurrence on the problem keyword adaptation index. The fitting curve corresponding to the keyword co-occurrence is obtained by fitting the relationship between the historical keyword co-occurrence frequency, the historical keyword co-occurrence number and the keyword co-occurrence, and the real-time keyword co-occurrence frequency and the keyword co-occurrence number are brought into the fitting curve corresponding to the keyword co-occurrence, thereby obtaining the impact factor corresponding to the unit value of the keyword co-occurrence in this example. The value range of this value in this example is 0 to 1.

[0070] zl 2 It is the impact factor corresponding to the average correlation level unit value predefined in the industrial chain map database, which represents the degree of influence of the average correlation level unit value on the problem keyword adaptation index. The fitting curve corresponding to the average correlation level is obtained by fitting the relationship between the number of historical keyword direct associations, the number of historical keyword indirect associations and the average correlation level. The real-time number of keyword direct associations and the number of keyword indirect associations are brought into the fitting curve corresponding to the average correlation level, thereby obtaining the impact factor corresponding to the average correlation level unit value in this example. The value range of this value in this example is 0 to 1.

[0071] In this example embodiment, the change table of the problem keyword adaptation index and its corresponding parameters between the sth quality problem key indicator of the bth production factor of the ath production link belonging to the above industrial chain map and each problem classification label is shown in Table 2:

[0072] Table 2 Changes in the problem keyword adaptation index and its corresponding parameters between the key indicators of the sth quality problem of the bth production factor in the ath production link of the industrial chain map and each problem classification label

[0073]

[0074]

[0075] In this example embodiment, the value of the impact factor corresponding to the unit value of the keyword co-occurrence is 0.5, the value of the impact factor corresponding to the unit value of the average correlation level is 0.4, the keyword co-occurrence is positively correlated with the average correlation level, the sth quality problem key indicator of the bth production factor of the ath production link belonging to the industrial chain map has the highest problem keyword adaptation index with the 4th problem classification label, then the sth quality problem key indicator of the bth production factor of the ath production link belonging to the industrial chain map belongs to the 4th problem classification label.

[0076] In a specific embodiment, the present invention provides a method for generating an industrial chain quality map, analyzes the internal correlation of each production factor in each production link of the industrial chain, helps to understand the mutual influence and dependence between each production factor in each production link in the industrial chain, thereby generating an industrial chain map with strong reliability, and evaluates the influence weight index of each quality problem indicator of each production factor in each production link of the industrial chain map, accurately identifies each key indicator of quality problems that have a greater impact on the industrial chain, and finally classifies and processes each key indicator of quality problems of each production factor in each production link of the industrial chain map, thereby generating an industrial chain quality map, focusing on the core issues of the industrial chain through refined analysis of multi-dimensional parameters, guiding the generation of the industrial chain quality map, accurately examining the quality problems of the industrial chain, and providing strong data support for optimizing the configuration of the industrial chain.

[0077] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A method for generating an industrial chain quality map, characterized in that: include: S1. Obtain the process parameters of each production factor in each production link of the industrial chain, analyze the internal correlation of each production factor in each production link of the industrial chain, and generate an industrial chain map; S2. Collect the quality problem indicators of each production factor in each production link of the industrial chain map, evaluate the impact weight index of each quality problem indicator of each production factor in each production link of the industrial chain map, and extract the key indicators of each quality problem of each production factor in each production link of the industrial chain map; S3. Classify and process the key indicators of quality issues of each production factor in each production link of the industrial chain map, thereby generating an industrial chain quality map.

2. The method for generating an industrial chain quality map according to claim 1, characterized in that: The analysis of the internal correlation of each production factor in each production link of the industrial chain is as follows: According to the process parameters of each production factor in each production link of the industrial chain, the occurrence times of each production factor in each production link of the industrial chain are extracted, and matched with the data coupling degrees corresponding to each occurrence frequency interval stored in the industrial chain map database, thereby obtaining the data coupling degrees of each production factor in each production link of the industrial chain; Locate the initial time point corresponding to each occurrence of each production factor in each production link of the industrial chain, and perform difference processing with the end time point corresponding to each occurrence of each production factor in each production link of the industrial chain to obtain the data support duration of each production factor in each production link of the industrial chain; The product output of the production equipment corresponding to each production factor in each production link of the industrial chain within the production inspection cycle is extracted from the process parameters of each production factor in each production link of the industrial chain, and the ratio is processed with the duration corresponding to the production inspection cycle to obtain the production speed of the production equipment corresponding to each production factor in each production link of the industrial chain. Based on this comprehensive analysis, the internal correlation of each production factor in each production link of the industrial chain is obtained.

3. The method for generating an industrial chain quality map according to claim 2, characterized in that: The specific analysis process of generating the industrial chain map is as follows: The internal correlation of each production factor in each production link of the industrial chain is sorted in order from large to small, and the internal correlation order of each production factor in each production link of the industrial chain is obtained, thereby generating an industrial chain map.

4. The method for generating an industrial chain quality map according to claim 1, characterized in that: The quality problem indicators of each production factor in each production link of the industrial chain map specifically include each occurrence time point of each quality problem indicator of each production factor in each production link of the industrial chain map within the production inspection cycle.

5. The method for generating an industrial chain quality map according to claim 4, characterized in that: The impact weight index of each quality problem indicator of each production factor in each production link of the industrial chain map is evaluated, and the specific evaluation process is as follows: Locate and count the number of post-production factor position nodes and the number of post-production factor position node types corresponding to each quality problem indicator of each production factor in each production link of the industrial chain map, and the connectivity factor corresponding to the number of post-production factor position nodes and the connectivity factor corresponding to the number of post-production factor position node types preset in the industrial chain map database, perform product analysis respectively, and accumulate and process to obtain the production factor position node connectivity corresponding to each quality problem indicator of each production factor in each production link of the industrial chain map; Count the number of time points at which each quality problem indicator of each production factor in each production link of the industrial chain map appears within the production inspection cycle, and perform ratio processing with the number of time points at which each quality problem indicator of each production factor in each production link of the industrial chain map is resolved within the production inspection cycle, so as to obtain the adjustability of each quality problem indicator of each production factor in each production link of the industrial chain map within the production inspection cycle; The number of improvement plans for each quality problem indicator of each production factor in each production link of the industrial chain map within the production inspection cycle is obtained, and the impact weight index of each quality problem indicator of each production factor in each production link of the industrial chain map is obtained by comprehensive evaluation.

6. The method for generating an industrial chain quality map according to claim 5, characterized in that: The influence weight index of each quality problem indicator of each production factor in each production link of the industrial chain map is specifically expressed as follows: Among them, μ abn AJ is the impact weight index of the nth quality problem indicator of the bth production factor in the ath production link of the industrial chain map, abn is the node connectivity of the production factor position corresponding to the nth quality problem indicator of the bth production factor in the ath production link of the industrial chain graph, NP abn AT is the adjustability of the nth quality problem indicator of the bth production factor in the ath production link of the industrial chain map within the production inspection cycle, abn is the number of improvement plans for the nth quality problem indicator of the bth production factor of the ath production link in the industrial chain map within the production inspection cycle, mk1 is the impact factor corresponding to the unit value of the production factor position node connectivity predefined in the industrial chain map database, mk2 is the impact factor corresponding to the unit value of the adjustable degree predefined in the industrial chain map database, mk3 is the impact factor corresponding to the unit value of the number of improvement plans predefined in the industrial chain map database, a is the number of each production link, a={1,2,3,...,t}, t is the total number of production links, b is the number of each production factor, b={1,2,3,...,p}, p is the total number of production factors, n is the number of each quality problem indicator, n={1,2,3,...,m}, m is the total number of quality problem indicators.

7. The method for generating an industrial chain quality map according to claim 6, characterized in that: The specific analysis process of extracting the key indicators of each quality problem of each production factor in each production link of the industrial chain map is as follows: The influence weight index of each quality problem indicator of each production factor in each production link of the industrial chain map is compared with the influence weight threshold stored in the industrial chain map database, and the quality problem indicators of each production factor in each production link of the industrial chain map whose influence weight index is greater than the influence weight threshold are screened out and marked as the key indicators of each quality problem of each production factor in each production link of the industrial chain map.

8. The method for generating an industrial chain quality map according to claim 1, characterized in that: The described classification processing of the key indicators of quality problems of each production factor in each production link belonging to the industrial chain map is specifically to analyze the problem keyword adaptation index between the key indicators of quality problems of each production factor in each production link belonging to the industrial chain map and each problem classification label, thereby classifying and processing the key indicators of quality problems of each production factor in each production link belonging to the industrial chain map.

9. The method for generating an industrial chain quality map according to claim 8, characterized in that: The problem keyword adaptation index between the key indicators of each quality problem of each production factor in each production link of the industrial chain map and each problem classification label is analyzed in detail as follows: Extract each keyword of each problem classification label from the industrial chain map database, and locate each keyword of each key indicator name of each quality problem of each production factor in each production link of the industrial chain map at the same time, and compare with each keyword of each problem classification label for keyword overlap, count the number of keyword overlaps between each keyword of each key indicator name of each quality problem of each production factor in each production link of the industrial chain map and each keyword of each problem classification label, and perform ratio processing with the total number of keywords of each problem classification label to obtain the keyword co-occurrence degree between each key indicator of each quality problem of each production factor in each production link of the industrial chain map and each problem classification label; The correlation levels between the keywords of the key indicators of each quality problem of each production factor in each production link of the industrial chain map and the keywords of each problem classification label are matched from the industrial chain map database, and the average correlation level between the key indicators of each quality problem of each production factor in each production link of the industrial chain map and each problem classification label is obtained by mean processing. Based on this, the adaptation index of the key indicators of each quality problem of each production factor in each production link of the industrial chain map and the problem keywords of each problem classification label is obtained through comprehensive analysis.

10. The method for generating an industrial chain quality map according to claim 9, characterized in that: The specific analysis process of generating the quality map of the industrial chain is as follows: After classifying the key indicators of quality problems of each production factor in each production link of the industrial chain map, they are automatically filled into the corresponding problem classification labels in the industrial chain map, thereby generating an industrial chain quality map.

Images