A traceability management method and system for industrial products

Abstract

The invention discloses a traceability management method and system for industrial products, relates to the technical field of industrial product production management, establishes a comprehensive product traceability database, covers detailed records of product production, accessories production links and assembly processes; uses a semantic recognition model to deeply analyze historical product abnormal conditions, identifies abnormal state types and parameters, constructs a historical abnormal state parameter group, and establishes associations with abnormal accessories to form abnormal state-abnormal accessories association groups; classifies and aggregates these association groups and selects the best to construct an abnormal state-abnormal accessories corresponding library; uses the semantic recognition model again to identify and analyze the current product abnormal conditions, and quickly locks the abnormal accessories and their production links by matching the abnormal state-abnormal accessories corresponding library to achieve efficient traceability; the invention significantly improves the traceability efficiency and accuracy, reduces the traceability cost, and provides strong support for industrial product quality control.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial product production management, and in particular to a traceability management method and system for industrial products. Background Art

[0002] In the modern industrial manufacturing environment, the stability and reliability of product quality are the core elements to ensure the competitiveness of enterprises. However, with the complexity of product structure and the extension of production process, when abnormal conditions occur in products, how to efficiently and accurately locate the source of the problem has become a major challenge for enterprises. Traditional traceability methods usually rely on physical disassembly and inspection of products one by one, which is not only time-consuming and labor-intensive, but may also damage the product itself and increase unnecessary costs.

[0003] Especially when facing large-scale production or highly integrated industrial products, the description of abnormal conditions is often complex and vague, and it is difficult to directly point to specific accessories or production links. This makes it difficult for companies to quickly identify the problem without obvious instructions, resulting in inefficient tracing and even delaying the resolution of the problem. Summary of the invention

[0004] The purpose of the present invention is to provide a traceability management method and system that can quickly and accurately determine abnormal accessories or production links based only on the description of abnormal conditions without relying on product disassembly analysis.

[0005] The present invention discloses a traceability management method for industrial products, comprising:

[0006] Step S100, establishing a product traceability database, including the product production records, including the production link records of each accessory, and the accessory assembly process records;

[0007] Step S200, using a semantic recognition model to identify and analyze the historical product abnormal condition description, determine the historical abnormal state type and historical abnormal state parameters of the product, and record the combination of the historical abnormal state type and historical abnormal state parameters of the product as a historical abnormal state parameter group;

[0008] Step S300, determining the historical product abnormality inspection records corresponding to the historical abnormal state parameter group, analyzing the historical product abnormality inspection records, determining the abnormal accessories of each product, and establishing an association between the historical abnormal state parameter group and the abnormal accessories to obtain an abnormal state-abnormal accessories association group;

[0009] Step S400, classifying and grouping the abnormal state-abnormal accessory association groups to obtain abnormal state-abnormal accessory association sets, combining the abnormal state-abnormal accessory association sets to obtain abnormal state-abnormal accessory corresponding libraries, performing equivalent performance analysis of abnormal accessories between abnormal state-abnormal accessory association groups in the abnormal state-abnormal accessory association sets, and determining optimal selection parameters for the abnormal state-abnormal accessory association groups based on the analysis results;

[0010] Step S500, using the semantic recognition model to identify and analyze the current product abnormal condition description, determine the current abnormal state type and current abnormal state parameter of the product, and record the combination of the current abnormal state type and the current abnormal state parameter of the product as a current abnormal state parameter group;

[0011] Step S600, use the abnormal state-abnormal accessory correspondence library to match and analyze the current abnormal state parameter group, determine several matching abnormal state-abnormal accessory association groups, and select parameters for each abnormal state-abnormal accessory association group, prioritize the abnormal state-abnormal accessory association groups, and retrieve the corresponding abnormal accessory production link records and accessory assembly process records.

[0012] In some embodiments disclosed in the present invention, a method for identifying and analyzing a product abnormal condition description using a semantic recognition model includes:

[0013] Step S201, cleaning the product abnormal condition description, removing redundant words and punctuation marks, and performing word segmentation and part-of-speech tagging to obtain a processed abnormal condition text segment;

[0014] Step S202, based on a preset abnormal description keyword comparison set, scanning and analyzing the processed abnormal condition text segment, extracting the abnormal description keywords, and determining the abnormal state type to which the abnormal description keywords belong, locating the abnormal description keywords belonging to the same abnormal state type, and determining the abnormal description segment corresponding to the abnormal description keywords;

[0015] In step S203, an abnormal state parameter table is preset, which includes several abnormal description keywords, and each abnormal description keyword corresponds to several abnormal description segments, and each abnormal description segment corresponds to an abnormal state parameter. Using the abnormal state parameter table, each abnormal description segment is identified to determine the abnormal state parameter of the abnormal description segment.

[0016] In some embodiments disclosed in the present invention, a method for classifying and grouping abnormal state-abnormal accessory association groups includes:

[0017] Step S401, comparing the historical abnormal state parameter groups in the abnormal state-abnormal accessory association group, classifying the abnormal state-abnormal accessory association group for the first time based on the historical abnormal state type, and classifying the abnormal state-abnormal accessory association group after the first classification for the second time based on the historical abnormal state parameter;

[0018] Step S402, based on the equality of abnormal parts, the abnormal state-abnormal parts association groups after the second classification are classified for the third time to obtain a plurality of abnormal state-abnormal parts association sets;

[0019] Among them, the method for classifying the abnormal state-abnormal accessory association group based on the historical abnormal state parameters includes:

[0020] Several initial abnormal state parameter intervals are set for each historical state type, and the number of parameter mappings of the historical abnormal state parameters mapped into each initial abnormal state parameter interval is determined. Based on the number of parameter mappings, the corresponding initial abnormal state parameter interval is amplified to obtain the amplified abnormal state parameter interval, the historical abnormal state parameters belonging to the same amplified abnormal state parameter interval are determined, and the abnormal state-abnormal accessory association groups corresponding to the historical abnormal state parameters are classified.

[0021] In some embodiments disclosed in the present invention, a method for performing equivalent performance analysis of abnormal accessories between abnormal state-abnormal accessory association groups includes:

[0022] Step S403, analyzing each product production record in the product traceability database, constructing a product production process expression model, and associating each accessory expression component in the product production process expression model with the product traceability database, including associating each accessory expression component with a production link record and an accessory assembly process record;

[0023] Step S404, determining a product production process expression model corresponding to each abnormal state-abnormal accessory association group, and marking the accessory expression component in the product production process expression model based on the abnormal accessories in the abnormal state-abnormal accessory association group;

[0024] Step S405, compare the product production process expression models corresponding to the abnormal state-abnormal accessories association group with each other, classify the equivalent product production process expression models, and identify the number of product production process expression models in each category as the preferred selection parameter of the abnormal state-abnormal accessories association group corresponding to the product production process expression model.

[0025] In some embodiments disclosed in the present invention, the method for constructing a product production process expression model includes:

[0026] Step S4031, analyzing the production process of the product, determining a number of sub-production process sequences, and determining the process connection points between the sub-production process sequences;

[0027] Step S4032, determine the accessory expression components to be constructed based on the sub-production process sequence, and connect the accessory expression components in sequence to obtain the sub-production process expression model, and connect the sub-production process expression models to each other according to the process connection points between the sub-production process sequences to obtain the product production process expression model.

[0028] In some embodiments disclosed in the present invention, the method for determining that product production process expression models are equivalent to each other includes:

[0029] Step S4051, based on the historical product abnormality inspection records, a number of abnormal accessories corresponding to the abnormal condition of the same product are determined, and based on the determined number of abnormal accessories, the accessory expression components on the product production process expression model are divided into blocks to obtain a number of abnormal condition mapping blocks;

[0030] Step S4052, mark the abnormal situation mapping block to which the accessory expression component marked on the product production process expression model belongs, and based on the equality of the marked abnormal situation mapping block and the marked accessory expression component, determine the equivalent parameters between the product production process expression models, and based on the mutual equivalent parameters, classify the product process expression models equally.

[0031] In some embodiments disclosed in the present invention, a method for determining equivalent parameters between product production process expression models includes:

[0032] Step S40521, based on the equivalence between the abnormal condition mapping blocks between the product production process expression models, determine the sub-equivalence parameters, and based on the abnormal weight coefficient preset in the abnormal mapping block, correct the sub-equivalence parameters, and based on the equivalence of the accessory expression components marked in the abnormal mapping block, correct the sub-equivalence parameters again;

[0033] Among them, the expression for calculating equivalent parameters is:

[0034] ;

[0035] ;

[0036] in, are equivalent parameters, is the adjustment coefficient of the sub-equal parameter impact, is the abnormal weight coefficient of the i-th abnormal condition mapping block in the compared product production process expression model, is the block equivalence judgment function of the ith abnormal condition mapping block in the compared product production process expression model. If the ith abnormal condition mapping block in the compared product production process expression model has an equivalent abnormal condition mapping block, then Output 1, otherwise output 0. The number of abnormal condition mapping blocks in the production process expression model for the first compared product, The number of abnormal condition mapping blocks in the production process expression model for the second comparison product, The adjustment factor for the component equality effect of the accessory expression, is the equivalent impact calculation function of the accessory expression component of the i-th abnormal condition mapping block in the compared product production process expression model, is the component equivalence judgment function of the xth marked accessory expression component in the compared abnormal condition mapping block. If the xth marked accessory expression component in the compared abnormal condition mapping block has an equivalent marked accessory expression component, then Output 1, otherwise output 0. The number of marked accessory expression components in the first abnormal condition mapping block among the abnormal condition mapping blocks to be compared, The number of components expressed by the marked accessories in the second abnormal condition mapping block in the abnormal condition mapping block for comparison.

[0037] In some embodiments disclosed in the present invention, the method of matching and analyzing the current abnormal state parameter group using the abnormal state-abnormal accessory correspondence library includes:

[0038] Step S601, taking the current abnormal state type in the current abnormal state parameter group as the first search condition, performing the first search in the abnormal state-abnormal accessories corresponding library, taking the current abnormal state parameter in the current abnormal state parameter group as the second search condition, performing the second search in the abnormal state-abnormal accessories corresponding library, and determining a number of matching abnormal state-abnormal accessories association groups.

[0039] In some embodiments disclosed in the present invention, a traceability management system for industrial products is also disclosed, which is used to execute the traceability management method of any one of claims 1 to 3, including:

[0040] The first module is used to establish a product traceability database, including the product production records, the production link records of each accessory, and the accessory assembly process records;

[0041] The second module is used to identify and analyze the historical product abnormal condition descriptions using a semantic recognition model, determine the historical abnormal state type and historical abnormal state parameters of the product, and record the combination of the historical abnormal state type and historical abnormal state parameters of the product as a historical abnormal state parameter group;

[0042] The third module is used to determine the historical product abnormal inspection records corresponding to the historical abnormal state parameter group, analyze the historical product abnormal inspection records, determine the abnormal accessories of each product, and establish the association between the historical abnormal state parameter group and the abnormal accessories to obtain the abnormal state-abnormal accessories association group;

[0043] The fourth module is used to classify and group the abnormal state-abnormal accessory association groups to obtain abnormal state-abnormal accessory association sets, and combine the abnormal state-abnormal accessory association sets to obtain abnormal state-abnormal accessory corresponding libraries, perform equivalent performance analysis of abnormal accessories between abnormal state-abnormal accessory association groups in the abnormal state-abnormal accessory association sets, and determine the optimal selection parameters of the abnormal state-abnormal accessory association groups based on the analysis results;

[0044] The fifth module is used to identify and analyze the current abnormal state description of the product using the semantic recognition model, determine the current abnormal state type and the current abnormal state parameter of the product, and record the combination of the current abnormal state type and the current abnormal state parameter of the product as the current abnormal state parameter group;

[0045] The sixth module is used to use the abnormal state-abnormal accessory corresponding library to match and analyze the current abnormal state parameter group, determine several matching abnormal state-abnormal accessory association groups, and select parameters for each abnormal state-abnormal accessory association group, prioritize the abnormal state-abnormal accessory association groups, and retrieve the corresponding abnormal accessory production link records and accessory assembly process records.

[0046] The present invention discloses a traceability management method and system for industrial products, which relates to the technical field of industrial product production management, establishes a comprehensive product traceability database, covers detailed records of product production, parts production links and assembly processes; uses a semantic recognition model to deeply analyze historical product abnormal conditions, identifies abnormal state types and parameters, constructs a historical abnormal state parameter group, and establishes associations with abnormal parts to form abnormal state-abnormal parts association groups; classifies and aggregates these association groups and selects the best to construct an abnormal state-abnormal parts corresponding library; uses the semantic recognition model again to identify and analyze the current product abnormal conditions, and quickly locks the abnormal parts and their production links by matching the abnormal state-abnormal parts corresponding library to achieve efficient traceability; the present invention significantly improves the traceability efficiency and accuracy, reduces the traceability cost, and provides strong support for industrial product quality control.

[0047] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 The present invention is a method step diagram of a traceability management method for industrial products disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0050] The following will be combined with the accompanying drawings and specific embodiments to clearly and completely describe the technical solution of the present invention. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention and cannot be understood as limiting the scope of protection of the present invention. Those skilled in the art in this field can make some non-essential improvements and adjustments based on the content of the present invention described below. In the present invention, unless otherwise clearly specified and limited, the technical terms used in the present invention should be the common meanings understood by the technical personnel described in the present invention.

[0051] Example:

[0052] The purpose of the present invention is to provide a traceability management method and system that can quickly and accurately determine abnormal accessories or production links based only on the description of abnormal conditions without relying on product disassembly analysis.

[0053] The present invention discloses a traceability management method for industrial products. Figure 1 ,include:

[0054] Step S100, establishing a product traceability database, including the product production records, including the production link records of each accessory, and the accessory assembly process records.

[0055] This step is the basis of traceability management. By establishing a comprehensive product traceability database, enterprises can systematically record and manage all information related to product production. This includes the overall production records of the product, detailed records of the production links of each accessory, and the process records of accessories during assembly. Such a database provides solid data support for subsequent abnormal analysis and traceability, ensuring the accuracy and comprehensiveness of traceability.

[0056] Step S200, using a semantic recognition model to identify and analyze historical product abnormal condition descriptions, determine the product's historical abnormal state type and historical abnormal state parameters, and record the combination of the product's historical abnormal state type and historical abnormal state parameters as a historical abnormal state parameter group.

[0057] This step uses advanced semantic recognition technology to conduct an in-depth analysis of the description of historical product abnormal conditions. The semantic recognition model can accurately understand the key information in the text, such as the type of abnormality, performance characteristics, etc., and convert it into structured data. Through this process, we can extract valuable information from historical data to form a historical abnormal state parameter group, which provides strong support for the subsequent establishment of the association between abnormal states and abnormal accessories. The implementation of this step greatly improves the efficiency and accuracy of abnormality identification and lays a solid foundation for traceability management.

[0058] In some embodiments disclosed in the present invention, a method for identifying and analyzing a product abnormal condition description using a semantic recognition model includes:

[0059] Step S201 , cleaning the product abnormal condition description, removing redundant words and punctuation marks, and performing word segmentation and part-of-speech tagging to obtain a processed abnormal condition text segment.

[0060] This step is the starting point of semantic recognition analysis, and its main purpose is to clean and normalize the original description of product abnormal conditions. First, by removing redundant words, punctuation marks and other irrelevant information, noise interference is reduced and the accuracy of subsequent analysis is improved. Next, word segmentation and part-of-speech tagging are performed to divide the continuous text into independent vocabulary units and mark the part of speech of each word, such as noun, verb, etc. The output of this step is the processed abnormal condition text, which provides a clear and orderly text basis for subsequent keyword extraction and abnormal state type identification.

[0061] Step S202, based on a preset abnormal description keyword comparison set, the processed abnormal condition text segment is scanned and analyzed, the abnormal description keywords are extracted, and the abnormal state type to which the abnormal description keywords belong is determined, the abnormal description keywords belonging to the same abnormal state type are located, and the abnormal description segment corresponding to the abnormal description keywords is determined.

[0062] The core of this step is to extract keywords directly related to product abnormalities and determine the abnormal state types to which these keywords belong. Through the preset abnormal description keyword comparison set, we can scan and analyze the processed abnormal condition text and quickly identify keywords related to the abnormality. These keywords are the bridge between abnormal conditions and abnormal state types. Subsequently, based on the semantic information and contextual relationship of the keywords, we can determine the abnormal state type to which each keyword belongs, such as temperature abnormality, pressure abnormality, etc. At the same time, the keywords belonging to the same abnormal state type are located, and the corresponding abnormal description segments are divided to provide accurate target areas for subsequent abnormal state parameter identification.

[0063] In step S203, an abnormal state parameter table is preset, which includes several abnormal description keywords, and each abnormal description keyword corresponds to several abnormal description segments, and each abnormal description segment corresponds to an abnormal state parameter. Using the abnormal state parameter table, each abnormal description segment is identified to determine the abnormal state parameter of the abnormal description segment.

[0064] This step is the ultimate goal of semantic recognition analysis, that is, to determine the abnormal state parameters in the abnormal description segment. In order to achieve this goal, we preset an abnormal state parameter table, which contains a large number of description keywords related to abnormal states and their corresponding abnormal description segments and abnormal state parameters. By using this parameter table, we can identify each abnormal description segment one by one and match the abnormal state parameters that best match the current segment description. These parameters specifically describe the state characteristics of the abnormality, such as the specific value of the temperature abnormality, the fluctuation range of the pressure abnormality, etc., which provide key data support for subsequent traceability analysis and problem solving. Through the implementation of this step, we can accurately extract valuable abnormal state information from the product abnormal condition description and provide strong data support for traceability management.

[0065] Step S300, determine the historical product abnormal inspection records corresponding to the historical abnormal state parameter group, analyze the historical product abnormal inspection records, determine the abnormal accessories of each product, and establish an association between the historical abnormal state parameter group and the abnormal accessories to obtain an abnormal state-abnormal accessory association group.

[0066] This step aims to identify the specific parts that caused the abnormality by comparing the historical abnormal state parameter group with the abnormal inspection record. By analyzing the various data in the abnormal inspection record in detail, such as the test results, abnormal performance, etc., we can determine which parts are directly related to the abnormal state. This process not only helps us establish a clear association between the abnormal state and the abnormal parts, but also provides a specific goal for subsequent tracing, making the tracing work more targeted.

[0067] Step S400, classify and group the abnormal state-abnormal accessory association groups to obtain abnormal state-abnormal accessory association sets, and combine the abnormal state-abnormal accessory association sets to obtain an abnormal state-abnormal accessory corresponding library, perform equivalent performance analysis of abnormal accessories between the abnormal state-abnormal accessory association groups in the abnormal state-abnormal accessory association sets, and determine the optimal selection parameters of the abnormal state-abnormal accessory association groups based on the analysis results.

[0068] This step aims to systematically organize and optimize the established abnormal state-abnormal accessory association groups. Through the classification set, we can arrange the association groups in order according to the dimensions such as abnormality type and accessory type, which is convenient for subsequent search and use. At the same time, the process of selecting the best parameters is to further screen and optimize the association groups to ensure that the selected association groups are more representative and accurate. The implementation of this step not only improves the efficiency of traceability, but also enhances the reliability of the traceability results.

[0069] In some embodiments disclosed in the present invention, a method for classifying and grouping abnormal state-abnormal accessory association groups includes:

[0070] Step S401, comparing the historical abnormal state parameter groups in the abnormal state-abnormal accessory association group, classifying the abnormal state-abnormal accessory association group for the first time based on the historical abnormal state type, and classifying the abnormal state-abnormal accessory association group after the first classification for the second time based on the historical abnormal state parameter;

[0071] Step S402, based on the equality of abnormal parts, the abnormal state-abnormal parts association groups after the second classification are classified for the third time to obtain a plurality of abnormal state-abnormal parts association sets;

[0072] Among them, the method for classifying the abnormal state-abnormal accessory association group based on the historical abnormal state parameters includes:

[0073] Several initial abnormal state parameter intervals are set for each historical state type, and the number of parameter mappings of the historical abnormal state parameters mapped into each initial abnormal state parameter interval is determined. Based on the number of parameter mappings, the corresponding initial abnormal state parameter interval is amplified to obtain the amplified abnormal state parameter interval, the historical abnormal state parameters belonging to the same amplified abnormal state parameter interval are determined, and the abnormal state-abnormal accessory association groups corresponding to the historical abnormal state parameters are classified.

[0074] In some embodiments disclosed in the present invention, a method for performing equivalent performance analysis of abnormal accessories between abnormal state-abnormal accessory association groups includes:

[0075] Step S403, analyzing each product production record in the product traceability database, constructing a product production process expression model, and associating each accessory expression component in the product production process expression model with the product traceability database, including associating each accessory expression component with a production link record and an accessory assembly process record;

[0076] Step S404, determining a product production process expression model corresponding to each abnormal state-abnormal accessory association group, and marking the accessory expression component in the product production process expression model based on the abnormal accessories in the abnormal state-abnormal accessory association group;

[0077] Step S405, compare the product production process expression models corresponding to the abnormal state-abnormal accessories association group with each other, classify the equivalent product production process expression models, and identify the number of product production process expression models in each category as the preferred selection parameter of the abnormal state-abnormal accessories association group corresponding to the product production process expression model.

[0078] In some embodiments disclosed in the present invention, the method for constructing a product production process expression model includes:

[0079] Step S4031, analyze the production process of the product, determine a number of sub-production process sequences, and determine the process connection points between the sub-production process sequences.

[0080] At the beginning of building a product production process expression model, the first task is to conduct a comprehensive and in-depth analysis of the product's production process. This step involves a detailed combing of the entire production process from raw material procurement to final product output. Through analysis, we can clearly identify the key sub-production process sequences included in the product production process, which are the basic units that constitute the complete production process. At the same time, we also need to determine the process connection points between these sub-production process sequences, that is, how they are interconnected and interdependent. These connection points are the key to ensuring the continuity and smoothness of the production process, and are also an important reference for the subsequent construction of the production process expression model.

[0081] Step S4032, determine the accessory expression components to be constructed based on the sub-production process sequence, and connect the accessory expression components in sequence to obtain the sub-production process expression model, and connect the sub-production process expression models to each other according to the process connection points between the sub-production process sequences to obtain the product production process expression model.

[0082] After completing the analysis of the product production process, the next step is to determine the accessory expression components to be constructed based on the sub-production process sequence. These components are the basic elements that constitute the product production process expression model, and they represent the various accessories or semi-finished products involved in the production process. By connecting these accessory expression components in the order of the sub-production process sequence, we can obtain the sub-production process expression model. These sub-models describe different stages or links in the product production process. Finally, according to the process connection points between the sub-production process sequences, we connect these sub-production process expression models to each other to construct a complete product production process expression model. This model presents the entire production process of the product in a structured way, including key information such as each production process, the accessories used, the assembly relationship, and the process connection points, providing strong support for the production management and optimization of the enterprise.

[0083] In some embodiments disclosed in the present invention, the method for determining that product production process expression models are equivalent to each other includes:

[0084] Step S4051, based on the historical product abnormality inspection records, determine a number of abnormal accessories corresponding to the abnormal condition of the same product, and based on the determined abnormal accessories, block the accessory expression components on the product production process expression model to obtain a number of abnormal condition mapping blocks.

[0085] In the process of determining whether the product production process expression models are equivalent to each other, it is first necessary to conduct an in-depth analysis of the historical product abnormality inspection records. The purpose of this step is to determine the abnormal conditions of the products recorded in the inspection records and the abnormal accessories corresponding to the same abnormal condition. These abnormal accessories are the key link between product abnormalities and the production process. Subsequently, based on these determined abnormal accessories, the corresponding accessory expression components are found on the product production process expression model, and the blocks are demarcated to form several abnormal condition mapping blocks. These blocks are specific areas on the model that reflect the abnormal conditions of the product, and they will serve as an important basis for subsequent comparison and determination of model equivalence.

[0086] Step S4052, mark the abnormal situation mapping block to which the accessory expression component marked on the product production process expression model belongs, and based on the equality of the marked abnormal situation mapping block and the marked accessory expression component, determine the equivalent parameters between the product production process expression models, and based on the mutual equivalent parameters, classify the product process expression models equally.

[0087] After completing the demarcation of the abnormal condition mapping blocks, the next step is to clearly mark the abnormal condition mapping blocks to which the accessory expression components marked on the product production process expression model belong. The purpose of this step is to clearly identify which areas on the model are associated with product abnormal conditions. Subsequently, based on the equivalence of these marked abnormal condition mapping blocks and the accessory expression components they contain, we can begin to determine the equivalent parameters between the product production process expression models. These equivalent parameters may be certain specific structures of the model, the arrangement of accessories, the order of production processes, etc., which can reflect the similarity and consistency between models. Finally, based on these equivalent parameters, we can classify the product process expression models equally, and classify those models with similar or identical characteristics into one category, so as to facilitate subsequent management and optimization. This method can help enterprises better understand and compare different production process expression models, and provide strong support for the improvement and optimization of production processes.

[0088] In some embodiments disclosed in the present invention, a method for determining equivalent parameters between product production process expression models includes:

[0089] Step S40521, based on the equivalence between the abnormal condition mapping blocks between the product production process expression models, determine the sub-equivalent parameters, and based on the abnormal weight coefficient preset in the abnormal mapping block, correct the sub-equivalent parameters, and based on the equivalence of the accessory expression components marked in the abnormal mapping block, correct the sub-equivalent parameters again.

[0090] Among them, the expression for calculating equivalent parameters is:

[0091] ;

[0092] ;

[0093] in, are equivalent parameters, is the adjustment coefficient of the sub-equal parameter impact, is the abnormal weight coefficient of the i-th abnormal condition mapping block in the compared product production process expression model, is the block equivalence judgment function of the ith abnormal condition mapping block in the compared product production process expression model. If the ith abnormal condition mapping block in the compared product production process expression model has an equivalent abnormal condition mapping block, then Output 1, otherwise output 0. The number of abnormal condition mapping blocks in the production process expression model for the first compared product, The number of abnormal condition mapping blocks in the production process expression model for the second comparison product, The adjustment factor for the component equality effect of the accessory expression, is the equivalent impact calculation function of the accessory expression component of the i-th abnormal condition mapping block in the compared product production process expression model, is the component equivalence judgment function of the xth marked accessory expression component in the compared abnormal condition mapping block. If the xth marked accessory expression component in the compared abnormal condition mapping block has an equivalent marked accessory expression component, then Output 1, otherwise output 0. The number of marked accessory expression components in the first abnormal condition mapping block among the abnormal condition mapping blocks to be compared, The number of components expressed by the marked accessories in the second abnormal condition mapping block in the abnormal condition mapping block for comparison.

[0094] Step S500, using the semantic recognition model to identify and analyze the current product abnormal condition description, determine the product's current abnormal state type and current abnormal state parameters, and record the combination of the product's current abnormal state type and current abnormal state parameters as a current abnormal state parameter group.

[0095] This step is a key part of traceability management. When a new product anomaly occurs, we use the semantic recognition model again to identify and analyze the abnormal condition. Through the model's rapid response and accurate analysis, we can quickly determine the abnormal state type and related parameters of the current product, and provide basic data for subsequent matching analysis with the abnormal state-abnormal accessories corresponding library. The implementation of this step ensures the timeliness and accuracy of traceability work, and provides strong support for rapid problem solving.

[0096] Step S600, use the abnormal state-abnormal accessory correspondence library to match and analyze the current abnormal state parameter group, determine several matching abnormal state-abnormal accessory association groups, and select parameters for each abnormal state-abnormal accessory association group, prioritize the abnormal state-abnormal accessory association groups, and retrieve the corresponding abnormal accessory production link records and accessory assembly process records.

[0097] This step is the final link of traceability management. By matching and analyzing the current abnormal state parameter group with the abnormal state-abnormal accessories corresponding library, we can quickly lock the abnormal accessories and their production links that match the abnormal state. Subsequently, by retrieving relevant information such as the production records and quality inspection reports of these accessories, we can fully understand the production process and quality status of the accessories, and provide detailed traceability basis for problem location and resolution. The implementation of this step not only improves the efficiency and accuracy of traceability, but also provides a strong guarantee for enterprises to strengthen quality control and improve product quality.

[0098] In some embodiments disclosed in the present invention, the method of matching and analyzing the current abnormal state parameter group using the abnormal state-abnormal accessory correspondence library includes:

[0099] Step S601, taking the current abnormal state type in the current abnormal state parameter group as the first search condition, performing the first search in the abnormal state-abnormal accessories corresponding library, taking the current abnormal state parameter in the current abnormal state parameter group as the second search condition, performing the second search in the abnormal state-abnormal accessories corresponding library, and determining a number of matching abnormal state-abnormal accessories association groups.

[0100] When using the abnormal state-abnormal accessories corresponding library to match and analyze the current abnormal state parameter group, we first use the abnormal state type in the current abnormal state parameter group as the first search condition. The purpose of this step is to quickly locate all records related to the current abnormal state type in the abnormal state-abnormal accessories corresponding library, thereby narrowing the search scope and improving the matching efficiency. Next, we use the specific abnormal state parameters in the current abnormal state parameter group as the second search condition and further screen the result set of the first search. The purpose of this step is to accurately match those records that not only match the abnormal state type, but also match the abnormal state parameters. Through these two searches, we can determine several abnormal state-abnormal accessories association groups that fully match the current abnormal state parameter group. These association groups provide us with valuable information about the accessories, failure modes, and possible solutions that may be involved in the current abnormal state, which helps us quickly locate problems, analyze causes, and take effective countermeasures.

[0101] In some embodiments disclosed in the present invention, a traceability management system for industrial products is also disclosed, which is used to execute the traceability management method of any one of claims 1 to 3, including:

[0102] The first module is used to establish a product traceability database, including the product production records, the production link records of each accessory, and the accessory assembly process records.

[0103] The second module is used to use the semantic recognition model to identify and analyze the historical product abnormal condition descriptions, determine the product's historical abnormal state type and historical abnormal state parameters, and record the combination of the product's historical abnormal state type and historical abnormal state parameters as a historical abnormal state parameter group.

[0104] The third module is used to determine the historical product abnormal inspection records corresponding to the historical abnormal state parameter group, analyze the historical product abnormal inspection records, determine the abnormal accessories of each product, and establish the association between the historical abnormal state parameter group and the abnormal accessories to obtain the abnormal state-abnormal accessories association group.

[0105] The fourth module is used to classify and group the abnormal state-abnormal accessory association groups to obtain abnormal state-abnormal accessory association sets, and combine the abnormal state-abnormal accessory association sets to obtain an abnormal state-abnormal accessory corresponding library, perform equivalent performance analysis of abnormal accessories between the abnormal state-abnormal accessory association groups in the abnormal state-abnormal accessory association sets, and determine the optimal selection parameters of the abnormal state-abnormal accessory association groups based on the analysis results.

[0106] The fifth module is used to use the semantic recognition model to identify and analyze the current abnormal condition description of the product, determine the current abnormal state type and current abnormal state parameters of the product, and record the combination of the current abnormal state type and current abnormal state parameters of the product as the current abnormal state parameter group.

[0107] The sixth module is used to use the abnormal state-abnormal accessory corresponding library to match and analyze the current abnormal state parameter group, determine several matching abnormal state-abnormal accessory association groups, and select parameters for each abnormal state-abnormal accessory association group, prioritize the abnormal state-abnormal accessory association groups, and retrieve the corresponding abnormal accessory production link records and accessory assembly process records.

[0108] The present invention discloses a traceability management method and system for industrial products, which relates to the technical field of industrial product production management, establishes a comprehensive product traceability database, covers detailed records of product production, parts production links and assembly processes; uses a semantic recognition model to deeply analyze historical product abnormal conditions, identifies abnormal state types and parameters, constructs a historical abnormal state parameter group, and establishes associations with abnormal parts to form abnormal state-abnormal parts association groups; classifies and aggregates these association groups and selects the best to construct an abnormal state-abnormal parts corresponding library; uses the semantic recognition model again to identify and analyze the current product abnormal conditions, and quickly locks the abnormal parts and their production links by matching the abnormal state-abnormal parts corresponding library to achieve efficient traceability; the present invention significantly improves the traceability efficiency and accuracy, reduces the traceability cost, and provides strong support for industrial product quality control.

[0109] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present invention can be implemented by hardware, or by software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present invention.

[0110] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A traceability management method for industrial products, characterized in that: include: Establish a product traceability database, including the product production records, the production link records of each accessory, and the accessory assembly process records; The semantic recognition model is used to identify and analyze the historical product abnormal condition descriptions, determine the historical abnormal state type and historical abnormal state parameters of the product, and record the combination of the historical abnormal state type and historical abnormal state parameters of the product as a historical abnormal state parameter group; Determine the historical product abnormality inspection records corresponding to the historical abnormal state parameter group, analyze the historical product abnormality inspection records, determine the abnormal accessories of each product, and establish an association between the historical abnormal state parameter group and the abnormal accessories to obtain an abnormal state-abnormal accessories association group; Classify and aggregate the abnormal state-abnormal accessory association groups to obtain the abnormal state-abnormal accessory association set, and combine the abnormal state-abnormal accessory association sets to obtain the abnormal state-abnormal accessory corresponding library, perform equivalent performance analysis of abnormal accessories between the abnormal state-abnormal accessory association groups in the abnormal state-abnormal accessory association set, and determine the optimal selection parameters of the abnormal state-abnormal accessory association group based on the analysis results; The semantic recognition model is used to identify and analyze the current abnormal condition description of the product, determine the current abnormal state type and the current abnormal state parameter of the product, and record the combination of the current abnormal state type and the current abnormal state parameter of the product as the current abnormal state parameter group; Use the abnormal state-abnormal parts corresponding library to match and analyze the current abnormal state parameter group, determine several matching abnormal state-abnormal parts association groups, and prioritize the abnormal state-abnormal parts association groups according to the optimal selection parameters of each abnormal state-abnormal parts association group, and retrieve the corresponding abnormal parts production link records and parts assembly process records; Methods for identifying and analyzing product abnormal condition descriptions using semantic recognition models include: Clean the description of product abnormal conditions, remove redundant words and punctuation marks, and perform word segmentation and part-of-speech tagging to obtain the processed abnormal condition text segment; Based on the preset abnormal description keyword comparison set, the processed abnormal condition text segment is scanned and analyzed to extract the abnormal description keywords, and the abnormal state type to which the abnormal description keywords belong is determined, the abnormal description keywords belonging to the same abnormal state type are located, and the abnormal description segment corresponding to the abnormal description keywords is determined; An abnormal state parameter table is preset, the abnormal state parameter table includes a number of abnormal description keywords, and each abnormal description keyword corresponds to a number of abnormal description sections, and each abnormal description section corresponds to an abnormal state parameter, and each abnormal state parameter table is used to identify each abnormal description section to determine the abnormal state parameter of the abnormal description section; The method of classifying and grouping abnormal state-abnormal accessory association groups includes: Compare the historical abnormal state parameter groups in the abnormal state-abnormal accessory association group, classify the abnormal state-abnormal accessory association group for the first time based on the historical abnormal state type, and classify the abnormal state-abnormal accessory association group after the first classification for the second time based on the historical abnormal state parameter; Based on the equality of abnormal accessories, the abnormal state-abnormal accessories association groups after the second classification are classified for the third time to obtain several abnormal state-abnormal accessories association sets; Among them, the method for classifying the abnormal state-abnormal accessory association group based on the historical abnormal state parameters includes: A number of initial abnormal state parameter intervals are set for each historical state type, the number of parameter mappings of the historical abnormal state parameters mapped into each initial abnormal state parameter interval is determined, and based on the number of parameter mappings, the corresponding initial abnormal state parameter interval is amplified to obtain an amplified abnormal state parameter interval, the historical abnormal state parameters belonging to the same amplified abnormal state parameter interval are determined, and the abnormal state-abnormal accessory association groups corresponding to the historical abnormal state parameters are classified; The method for performing equivalent performance analysis of abnormal accessories between abnormal state-abnormal accessory association groups includes: Analyze each product production record in the product traceability database, build a product production process expression model, and associate each accessory expression component in the product production process expression model with the product traceability database, including associating each accessory expression component with the production link record and the accessory assembly process record; Determine the product production process expression model corresponding to each abnormal state-abnormal accessory association group, and mark the accessory expression component in the product production process expression model based on the abnormal accessories in the abnormal state-abnormal accessory association group; The product production process expression models corresponding to the abnormal state-abnormal parts association group are compared with each other, the identical product production process expression models are classified, and the number of product production process expression models in each category is identified as the optimal selection parameter of the abnormal state-abnormal parts association group corresponding to the product production process expression model; Methods for constructing a product production process expression model include: Analyze the production process of the product, determine several sub-production process sequences, and determine the process connection points between the sub-production process sequences; Determine the accessory expression components to be constructed based on the sub-production process sequence, and connect the accessory expression components in sequence to obtain the sub-production process expression model, and connect the sub-production process expression models to each other according to the process connection points between the sub-production process sequences to obtain the product production process expression model; Methods for identifying product production process expression models as equivalent include: Based on the historical product abnormality inspection records, several abnormal accessories corresponding to the abnormal conditions of the same product are determined, and based on the determined abnormal accessories, the accessory expression components on the product production process expression model are divided into blocks to obtain several abnormal condition mapping blocks; The abnormal condition mapping blocks to which the accessory expression components marked on the product production process expression model belong are marked, and based on the equality of the marked abnormal condition mapping blocks and the marked accessory expression components, the equivalent parameters between the product production process expression models are determined, and based on the mutual equivalent parameters, the product process expression models are equivalently classified.

2. A traceability management method for industrial products according to claim 1, characterized in that: Methods for determining equivalent parameters between product production process expression models include: Based on the equivalence between the abnormal condition mapping blocks between the product production process expression models, the sub-equivalence parameters are determined, and based on the abnormal weight coefficient preset in the abnormal mapping block, the sub-equivalence parameters are corrected, and based on the equivalence of the accessory expression components marked in the abnormal mapping block, the sub-equivalence parameters are corrected again; Among them, the expression for calculating equivalent parameters is: ； ； in, are equivalent parameters, is the adjustment coefficient of the sub-equal parameter impact, is the abnormal weight coefficient of the i-th abnormal condition mapping block in the compared product production process expression model, is the block equivalence judgment function of the ith abnormal condition mapping block in the compared product production process expression model. If the ith abnormal condition mapping block in the compared product production process expression model has an equivalent abnormal condition mapping block, then Output 1, otherwise output 0. The number of abnormal condition mapping blocks in the production process expression model for the first compared product, The number of abnormal condition mapping blocks in the production process expression model for the second comparison product, The adjustment factor for the component equality effect of the accessory expression, is the equivalent impact calculation function of the accessory expression component of the i-th abnormal condition mapping block in the compared product production process expression model, is the component equivalence judgment function of the xth marked accessory expression component in the compared abnormal condition mapping block. If the xth marked accessory expression component in the compared abnormal condition mapping block has an equivalent marked accessory expression component, then Output 1, otherwise output 0. The number of marked accessory expression components in the first abnormal condition mapping block among the abnormal condition mapping blocks to be compared, The number of components expressed by the marked accessories in the second abnormal condition mapping block in the abnormal condition mapping block for comparison.

3. A traceability management method for industrial products according to claim 1, characterized in that: The method of using the abnormal state-abnormal accessory corresponding library to match and analyze the current abnormal state parameter group includes: The current abnormal state type in the current abnormal state parameter group is used as the first search condition, and the first search is performed in the abnormal state-abnormal accessories corresponding library. The current abnormal state parameter in the current abnormal state parameter group is used as the second search condition, and the second search is performed in the abnormal state-abnormal accessories corresponding library to determine several matching abnormal state-abnormal accessories association groups.

4. A traceability management system for industrial products, characterized in that: A traceability management method for executing any one of claims 1 to 3, comprising: The first module is used to establish a product traceability database, including the product production records, the production link records of each accessory, and the accessory assembly process records; The second module is used to identify and analyze the historical product abnormal condition descriptions using a semantic recognition model, determine the historical abnormal state type and historical abnormal state parameters of the product, and record the combination of the historical abnormal state type and historical abnormal state parameters of the product as a historical abnormal state parameter group; The third module is used to determine the historical product abnormal inspection records corresponding to the historical abnormal state parameter group, analyze the historical product abnormal inspection records, determine the abnormal accessories of each product, and establish the association between the historical abnormal state parameter group and the abnormal accessories to obtain the abnormal state-abnormal accessories association group; The fourth module is used to classify and group the abnormal state-abnormal accessory association groups to obtain abnormal state-abnormal accessory association sets, and combine the abnormal state-abnormal accessory association sets to obtain abnormal state-abnormal accessory corresponding libraries, perform equivalent performance analysis of abnormal accessories between abnormal state-abnormal accessory association groups in the abnormal state-abnormal accessory association sets, and determine the optimal selection parameters of the abnormal state-abnormal accessory association groups based on the analysis results; The fifth module is used to identify and analyze the current abnormal state description of the product by using the semantic recognition model, determine the current abnormal state type and the current abnormal state parameter of the product, and record the combination of the current abnormal state type and the current abnormal state parameter of the product as the current abnormal state parameter group; The sixth module is used to use the abnormal state-abnormal accessory corresponding library to match and analyze the current abnormal state parameter group, determine several matching abnormal state-abnormal accessory association groups, and select parameters for each abnormal state-abnormal accessory association group, prioritize the abnormal state-abnormal accessory association groups, and retrieve the corresponding abnormal accessory production link records and accessory assembly process records.

Images