A product quality prediction method and system based on identification resolution

Abstract

The application discloses a product quality prediction method and system based on identification analysis, comprising the following steps: obtaining first basic data from an identification analysis secondary platform; obtaining one or more kinds of second basic data including production environment basic data and order basic data through an Internet of Things platform or an input mode; training a corresponding preset model based on the first basic data and / or the second basic data to obtain a product quality factor prediction model; and obtaining product quality prediction data by using the product quality factor prediction model and analyzing the product quality prediction data to predict the quality of a product. The method and system in the application selectively obtain data including upstream raw material total factor data, product historical experience data and the like from the identification analysis secondary platform, and collect total factor and total chain data of raw materials, environment and production through other data collection modes, and develop link modeling and multi-factor correlation comprehensive modeling, so that factors influencing product quality can be grasped, and product quality can be improved.

Description

Technical Field

[0001] This invention belongs to the field of product quality control technology, specifically relating to a product quality prediction method and system based on identifier resolution. Background Technology

[0002] Traditional quality inspection, whether sampling or comprehensive inspection, involves removing defective products from finished goods to ensure product quality before shipment. However, by this time, production is complete, and potential defects have already been produced. This post-production inspection method, which cannot prevent or control defects during the production process, can easily lead to serious waste and even irreparable losses. Later, statistical quality control emerged, but the data sources were primarily based on internal factory data. With the development of industrial production and technological advancements, quality requirements have become increasingly stringent. Furthermore, the application of intelligent manufacturing and the complexity of product manufacturing processes have made quality prediction a particularly prominent issue, rendering traditional statistical quality analysis inadequate for current production demands.

[0003] In recent years, with the rapid development of data acquisition and computer technologies, it has become possible to collect characteristics of quality parameters during the manufacturing process. These data can then be analyzed and predicted using quality prediction methods such as artificial neural networks and Bayesian methods. However, these methods cannot predict the current quality of products in real time and with high accuracy. Manufacturing, as a complex production process, involves a wide range of data, is influenced by various factors, and is characterized by numerous process parameters, significant nonlinearity, and dynamic changes, making it difficult to achieve accurate product quality predictions. Summary of the Invention

[0004] To address the technical problems encountered in the prior art mentioned above, this invention proposes a product quality prediction method and system based on identifier resolution. This method overcomes the problems in the prior art by selectively acquiring full-chain data including raw materials, environment, and production from a secondary identifier resolution platform. It establishes a multi-channel product information acquisition method and integrates the above data through system platform design to perform analysis and modeling, thereby achieving accurate prediction and analysis of product quality.

[0005] This invention discloses a product quality prediction method based on identifier resolution, comprising the following steps:

[0006] The system interacts with the platform to obtain first basic data based on identifier resolution; it obtains second basic data through an IoT platform or by input method; the first basic data on the platform is associated with the mapped identifier code;

[0007] Perform analysis based on at least one of the first basic data and / or the second basic data to obtain product quality prediction factors; perform multiple analyses based on the first basic data and the second basic data to obtain product quality prediction conclusions.

[0008] Furthermore, the second set of basic data includes raw material data:

[0009] The upstream raw material full-element data includes one or more of the following: raw material processing environment data, raw material equipment data, raw material production process data, raw material factory quality inspection data, raw material transportation data, raw material storage environment data, and raw material warehousing quality inspection data.

[0010] Furthermore, the first set of basic data includes historical product experience data and product quality data from the same industry:

[0011] The product historical experience data includes one or more of the following: historical product data of products produced within the factory, historical product quality inspection data, and historical customer complaint data.

[0012] The product quality data from the same industry includes one or more of the following: industry quality information and key quality parameter information.

[0013] Furthermore, the product quality predictor includes a model obtained based on the first basic data and / or the second basic data, which is:

[0014] The raw material model is obtained by training a pre-defined raw material model with comprehensive data on all upstream raw materials.

[0015] Train a pre-defined production environment model using basic production environment data to obtain a production model;

[0016] Train a pre-defined model for specific needs using basic order data to obtain an order model;

[0017] Train a pre-defined historical experience model using product historical experience data to obtain an experience model;

[0018] The industry product preset model is trained using data including product quality data from the same industry to obtain the industry model;

[0019] The product quality prediction conclusion is obtained by training a preset product quality prediction model based on output data from one or more of the raw material model, production model, order model, experience model, and industry model to obtain the product quality prediction conclusion.

[0020] Furthermore, the first basic data and the second basic data are managed by different servers.

[0021] This invention discloses a product quality prediction system based on identifier resolution, the system comprising:

[0022] It includes a data acquisition layer, which is used to acquire first basic data and second basic data in multiple ways to generate a shared resource pool, which is managed at the cloud service layer.

[0023] The data acquisition layer includes an identifier resolution device, which maps the uploaded identifier code of the first basic data and performs identifier resolution on the first basic data sent to the application layer.

[0024] The application layer obtains resources from the cloud server through the network transport layer;

[0025] The application layer performs analysis based on at least one of the first basic data and / or the second basic data to obtain product quality prediction factors; and performs multiple analyses based on the first basic data and the second basic data to obtain product quality prediction conclusions.

[0026] Furthermore, the product quality predictor includes a model obtained based on the first basic data and / or the second basic data, which is:

[0027] The raw material model is obtained by training a pre-defined raw material model with comprehensive data on all upstream raw materials.

[0028] Train a pre-defined production environment model using basic production environment data to obtain a production model;

[0029] Train a pre-defined model for specific needs using basic order data to obtain an order model;

[0030] Train a pre-defined historical experience model using product historical experience data to obtain an experience model;

[0031] The industry product preset model is trained using data including product quality data from the same industry to obtain the industry model;

[0032] The product quality prediction conclusion is obtained by training a preset product quality prediction model based on output data from one or more of the raw material model, production model, order model, experience model, and industry model to obtain the product quality prediction conclusion.

[0033] Furthermore, the first basic data and the second basic data are managed by different servers.

[0034] Furthermore, the data acquisition layer includes an Internet of Things (IoT) platform for acquiring second basic data.

[0035] In summary, compared with the prior art, the above-described technical solutions conceived in this invention can achieve the following beneficial effects:

[0036] (1) Based on the Industrial Internet Identifier Resolution Level 2 Platform, this invention selectively acquires full-chain data including upstream raw materials, environment and production, performs single-factor or multi-factor quality impact data collection and analysis modeling, and conducts comprehensive quality management, thereby enabling the prediction of the current quality of products and improving production efficiency.

[0037] (2) This invention addresses the real-world challenges faced by manufacturing enterprises, such as complex production processes, long production lines, numerous design departments, cross-departmental and cross-workshop operations, massive and fragmented data, and insufficient data analysis and processing capabilities. Based on a secondary identifier resolution platform, it effectively collects data, including but not limited to ERP and MES data, and uses historical and real-time data to formulate product quality analysis requirements and build data models to predict future quality fluctuation trends. This method significantly reduces data collection and processing time for enterprises while identifying key factors affecting product quality through analysis. This allows for the early detection and correction of potential quality-affecting factors during production, substantially reducing defect rates and minimizing potential serious waste or irreparable losses. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 A flowchart illustrating a product quality prediction method based on identifier resolution provided in an embodiment of the present invention;

[0040] Figure 2 This is a schematic diagram of the composition architecture of a product quality prediction system based on identifier resolution, provided in an embodiment of the present invention.

[0041] Figure 3 This is a flowchart illustrating the specific server composition in an embodiment of the present invention for implementing a product quality prediction method based on identifier resolution. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0043] The terms "first," "second," or "third," etc., used in the specification, claims, or accompanying drawings of this application are used to distinguish different objects, not to describe a particular order. Furthermore, the terms "comprising" or "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such processes, methods, products, or apparatus.

[0044] like Figure 1 As shown, this invention first discloses a product quality prediction method based on identifier resolution, including the following steps:

[0045] Interact with the platform to obtain the first basic data based on identifier resolution; obtain the second basic data through the IoT platform or by input method; associate the first basic data on the platform with the mapped identifier code;

[0046] Perform analysis based on at least one of the first and / or second basic data to obtain product quality predictive factors; perform multiple analyses based on the first and second basic data to obtain product quality predictive conclusions.

[0047] Specifically, the platform is an identifier resolution platform, officially known as the Industrial Internet Identifier Resolution System. It serves as the nerve center supporting the interconnection of the Industrial Internet, similar in function to the domain name resolution system in the internet field. It comprises a root node, five national-level top-level nodes, multiple secondary nodes (platforms), enterprise nodes, and public recursive nodes. The secondary identifier resolution platform provides identifier registration and resolution services to industry nodes. By associating data with identifier codes, it connects heterogeneous identifier systems, supporting information positioning and resource sharing for the Industrial Internet platform. This enables information filing, monitoring, and authentication, and the data stored on the platform is entered into a cloud resource pool for further data utilization by subsequent application programs. Enterprise nodes provide identifier registration and resolution services based on the specific internal needs of each enterprise. The secondary nodes connect upwards to the national top-level nodes and downwards to assign identifier codes to enterprise nodes and provide identifier registration, identifier data, and identifier resolution services.

[0048] The core of the Industrial Internet Identifier Resolution System includes identifier coding, identifier resolution system, and identifier data services. Among them:

[0049] Identification and Resolution System: Utilizing identification codes, this system uniquely locates and queries information about machines and goods, serving as the prerequisite and foundation for precise integration of global supply chain systems and enterprise production systems, full lifecycle management of products, and intelligent services. Identification Data Services: Facilitating industrial identification data management and cross-enterprise, cross-industry, cross-regional, and cross-national data sharing through identification coding resources and the identification and resolution system.

[0050] Identifier resolution technology refers to the process of mapping object identifiers to the information required for actual information services, such as addresses, items, and spatial locations. In one implementation, by parsing the identifier code of an item, the server address storing its associated information can be obtained. Identifier resolution technology connects all production elements in industry, such as equipment, machines, and materials, by establishing a unified identifier system. It connects fragmented data and applications through the resolution system, enabling the understanding of data sources, flow processes, and uses. Vertically, it can connect products, machines, workshops, and factories, achieving large-scale collection of underlying identifier data, data sharing between information systems, and analysis and application of identifier data. Horizontally, it can connect upstream and downstream enterprises, allowing them to query data on demand using identifier resolution. Small and medium-sized enterprises can be horizontally connected to form a platform, sharing data on demand using identifier resolution. End-to-end, it can connect the entire lifecycle of design, manufacturing, logistics, and use, achieving true lifecycle management.

[0051] Figure 2 The schematic diagram of the architecture of a product quality prediction system based on identifier resolution provided in one embodiment of the present invention includes a data acquisition layer for acquiring first basic data and second basic data in multiple ways to generate a shared resource pool, wherein the shared resource pool is managed at the cloud service layer.

[0052] The data acquisition layer includes an identifier resolution device, which maps the uploaded identifier code to the first basic data and performs identifier resolution on the first basic data sent to the application layer.

[0053] The application layer obtains resources from the cloud server through the network transport layer;

[0054] The application layer performs analysis based on at least one of the first and / or second basic data to obtain product quality prediction factors; and performs multiple analyses based on the first and second basic data to obtain product quality prediction conclusions.

[0055] The system comprises a data acquisition layer, a network transmission layer, and cloud infrastructure services and facilities (servers, storage, networks, etc.). It provides the hardware platform for cloud operations and a shared resource pool for collaborative development, data sharing, and business collaboration. The SaaS application layer monitors data within the shared resource pool, enabling standard chaining, target chaining, quality analysis, quality prediction, solution improvement and optimization, cloud production, and cloud procurement. The data acquisition layer includes identifier resolution gateways, identifier resolution all-in-one machines, and industrial intelligent devices, handling various data types such as environmental data, supplier data, raw material data, production data, flow data, business data (ERP / MES / DCS, etc.), network data, image / video data, and actively identified data.

[0056] In one embodiment of the present invention, the first basic data includes:

[0057] The entire supply chain of environmental data, including raw material warehouses, production workshops, finished product warehouses, and transportation information, is collected and uploaded to a secondary identifier resolution platform. This means that the factory's IoT platform collects environmental data from the entire supply chain, including production workshops and finished product warehouses, and uploads it to the secondary identifier resolution platform for use by other production units.

[0058] Supplier information, raw material information, and semi-finished product data are synchronized to the secondary identification and resolution platform in real time. Data is primarily entered into the factory's internal system or obtained from the IoT platform. Inspection and storage information for raw materials upon arrival is also internal factory data, obtained through system entry or the IoT platform. If the raw materials have identification tags, detailed information about the raw materials before arrival can be retrieved from the secondary identification and resolution platform based on these tags. After obtaining the information, it is encoded according to rules and registered with the secondary identification and resolution platform's external interface.

[0059] Processing data, such as equipment parameters, energy consumption data, and process parameters, are synchronized to the secondary identifier resolution platform in real time. This data is obtained from the IoT platform, encoded according to rules, and then used to call the external interfaces of the secondary identifier resolution platform and register with it.

[0060] Production data and business data [including ERP (Enterprise Resource Planning) data, MES (Manufacturing Execution System) data, etc.] are synchronized in real time to the identifier resolution secondary platform along with the production process. This involves registering business data such as customer orders from the ERP system, raw material supplier information, and production plans from the MES system to the identifier resolution secondary platform after being coded according to rules.

[0061] Based on the secondary identifier resolution platform, we obtain quality data information of identical products registered by similar companies in the same industry. Our product quality prediction system retrieves quality data information of identical products registered by similar companies in the same industry based on the company number on the secondary identifier resolution platform. The identifier registration rule codes on the secondary identifier resolution platform include country code, industry code, company code, category code, basic classification code, and sequence code, etc.; based on the company code following the industry code, we parse the corresponding company's product quality data information on the secondary identifier resolution platform.

[0062] In one embodiment of the present invention, the second basic data includes:

[0063] Equipped with an IoT data acquisition platform, it gathers data such as equipment operating parameters, production environment, and video feeds. Equipment operating parameter data (e.g., voltage, equipment pressure) and production environment data (e.g., workshop humidity, temperature) serve as inputs to the quality prediction model. When product quality is correlated with equipment operating parameter data or production environment data, these data can be used as indicators for quality judgment. Data transmission is achieved using WiFi, 4G / 5G, GPRS networks, or the Internet.

[0064] It gathers basic order data such as current orders, personalized customer needs, and customer complaints (i.e., basic data about customer orders, including but not limited to special quality requirements of products in the order, process-specific requirements, order parameter data, and sample quality inspection parameter data).

[0065] like Figure 3 As shown in the preferred embodiment of the present invention, the cloud server performs different information management through multiple different servers. In the first aspect, it can perform analysis on the data in a single server to obtain some factors in the quality correlation factors, or it can perform correlation analysis on the data in multiple different servers to obtain an overall analysis model of quality conclusions.

[0066] The first server 1 is used to obtain different information about materials through different channels and methods. Among them, raw material storage information, raw material warehousing quality inspection information, and material quality label information are obtained through data entry. The information obtained from the secondary platform includes material information from the secondary platform, including information on the entire process of raw material production and transportation. Real-time physical property parameters are obtained through data entry / Internet platform protocol.

[0067] Based on the data obtained above, the predicted values ​​of material quality, the key material parameter values ​​that affect product quality, and the raw material model are obtained through the server.

[0068] The raw material model will be trained using data such as raw material factory inspection information, transportation information, pre-warehouse inspection information, and storage environment information obtained from the secondary platform for identifier resolution (obtained by querying the external interface of the secondary platform for identifier resolution based on the raw material identifier).

[0069] The input to this raw material preset model is a complete set of upstream raw material data, including factory inspection information, transportation information, pre-warehouse inspection information, and storage environment information. The output is a raw material model that reflects the correspondence between the complete set of upstream raw material data and product quality. It can obtain predicted material quality values ​​and key material parameters that affect product quality.

[0070] The second server 2 will use basic production environment data, such as machine status data (e.g., voltage, current, temperature, and time), process production data (e.g., operation data, elasticity, speed, tension, and weight), and production environment data (e.g., factory humidity, temperature, light, and dust), obtained from the IoT acquisition platform through device collection or smart device interfaces, to train the pre-set production environment model and obtain the production model.

[0071] The input to this pre-set production environment model consists of basic production environment data such as machine and equipment status data, process production data, and production environment data. The output is a production model that reflects the correspondence between basic production environment data and product quality, and it also outputs key environmental parameter indicators that affect product quality.

[0072] The third server (3) will use order parameter information imported from the ERP system, customer customized demand data, specific key process data, key indicator parameters, and sampling quality inspection parameter data to train a specific demand preset model, thus obtaining an order model. The input to this specific demand preset model consists of basic order data such as order parameter information, customer customized demand data, specific key process data, key indicator parameters, and sampling quality inspection parameter data. The output is an order model that reflects the correspondence between basic order data and product quality.

[0073] Furthermore, key demand parameters that affect product quality can be obtained from the above information.

[0074] The fourth server will use historical product quality prediction data (data predicted by the model), historical product quality inspection data (actual product quality data from quality monitoring), and historical customer complaint data obtained from the secondary platform based on identifier resolution to train the historical experience preset model. The input of this historical experience preset model is the aforementioned historical product experience data, and the output is an experience model reflecting the correspondence between historical product experience data and product quality.

[0075] The fifth server (5) will use industry quality information and key quality parameter information obtained from the secondary platform based on identifier resolution, along with other product quality data from the same industry, to train a pre-set industry product model, thus obtaining an industry model. The input to this pre-set industry product model is the product quality data from the same industry, and the output is an industry model reflecting the correspondence between the product quality data and product quality.

[0076] The sixth server 6 uses the output data (i.e., the correspondence between the model's input data and product quality) of one or more of the five services (which correspond to the aforementioned five models and are used to carry the quality prediction model) that are dynamically and freely configured via the network based on the first server 1 to the fifth server 5 (e.g., if the cable depends on the quality of the raw materials, and the quality of the raw materials is not affected or is minimally affected by the workshop environment, then a raw material model can be configured instead of a workshop environment model). This data is then used as the input to the base server 6 to train the preset product quality prediction model, thus obtaining the product quality prediction model.

[0077] It should be noted that the specific training and modeling techniques mentioned above for training a preset model with corresponding data to obtain the final model can utilize existing artificial intelligence and machine learning technologies, and various deep learning model algorithms to obtain trained data and conclusions. No specific limitations are made here regarding raw material models, order models, production models, quality prediction models, etc.

[0078] In one preferred embodiment of the present invention, quality prediction includes the following steps:

[0079] Prediction Step 1: Based on the statistical analysis of the correspondence between upstream raw material full-element data and product quality, production environment basic data and product quality, and order basic data and product quality obtained from base servers 1, 2, and 3, the corresponding key parameters affecting product quality (including key raw material parameters, key environmental parameters, and key demand parameters) are calculated. The predicted key parameters affecting product quality are compared with the corresponding sample inspection parameters from the quality inspection department. The corresponding parameters are dynamically learned and adjusted, and the corresponding models are trained (i.e., the corresponding parameters are dynamically learned and adjusted, and the corresponding raw material model, production model, or order model is trained to obtain a better raw material model, production model, or order model).

[0080] Prediction Step Two: Based on Prediction Step One, the output information of the experience model of the basic server 4 (i.e., the correspondence between historical product experience data and product quality) is added as input to the product quality prediction model. Statistical analysis is performed simultaneously with the data from Prediction Step One. The calculated quality prediction index information is compared and analyzed with the sample inspection parameters of the quality inspection department. The corresponding parameters are dynamically learned and adjusted, and the corresponding model is trained, learned, and optimized.

[0081] Prediction Step 3: Based on Prediction Step 2, the output information of the industry model of the base server 5 (i.e., the correspondence between product quality data and product quality in the same industry) is added as input to the product quality prediction model. The base server 6 then performs calculation and comparative analysis on the prediction and calculates the product quality prediction data in real time based on the quality parameters of the sampled inspection. The product quality prediction data is analyzed to predict the quality of the product in real time.

[0082] Prediction Step 4: Analyze and compare the product quality and inspection quality parameters predicted by the product quality prediction model, input the deviation information into the product quality prediction model, and the product quality prediction model will dynamically correct and adjust and output the predicted quality parameter information in real time.

[0083] Step 5: Query and analyze historical product experience data and industry product quality data obtained from the secondary identifier resolution platform. Through data calculation and modeling analysis, obtain industry product quality prediction data. Based on the analysis of this prediction data, determine the overall product quality situation in the industry and predict future product quality. If peer companies or industry leaders have connected to the secondary identifier resolution platform, they can obtain industry product quality information and predict future industry product quality from that platform.

[0084] As a further preferred option, when the quality index information output by the product quality prediction model does not match the detection quality, the deviation information is entered into the product quality prediction model. The product quality prediction model reads the deviation information and dynamically learns and adjusts the relevant model parameters.

[0085] After quality inspectors submit quality inspection information to the secondary platform for identifier resolution, the system reads the quality inspection information from the secondary platform for identifier resolution and dynamically learns and adjusts the parameters of the relevant models.

[0086] For key information or key algorithm data for specific quality requirements, manual input is performed, and the system model dynamically learns and adjusts the parameters of the relevant model according to the input information.

[0087] This invention leverages a secondary identifier resolution platform to continuously improve product quality across the industry and supply chains. It selectively acquires and analyzes data from the platform, encompassing all elements and the entire supply chain, including upstream raw materials, the environment, and production processes. This allows for comprehensive control over the quality of finished and semi-finished products from the source of raw materials, thereby enhancing market competitiveness. Furthermore, this invention enables early detection of anomalies in the product industry and supply chains. Before production, it identifies key factors affecting product quality (including critical raw material parameters, environmental parameters, and demand parameters), allowing for proactive identification and correction of potential quality-affecting factors during production. This significantly reduces defect rates, minimizes potential serious waste or irreparable losses, and enables real-time prediction and assessment of product quality, thereby reducing customer complaints and other safety issues.

[0088] It should be noted that the flowcharts or block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in the flowchart or block diagram may represent a module, segment, or portion of code, which includes one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. Furthermore, it should be noted that each block in the block diagram or flowchart, and combinations of blocks in the block diagram or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0089] The description in this specification is merely illustrative of the invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the content of this specification or exceed the scope defined in the claims, they should all fall within the protection scope of this invention.

Claims

1. A product quality prediction method based on identifier resolution, characterized in that, Including the following steps: Interact with the platform to obtain the first basic data based on identifier resolution. The first basic data includes product historical experience data and product quality data of the same industry. The second basic data is acquired through an IoT platform or by data entry. This second basic data includes upstream raw material full-element data, production environment basic data, order basic data, product historical experience data, and industry peer product quality data. The upstream raw material full-element data includes one or more of the following: raw material processing environment data, raw material equipment data, raw material production process data, raw material factory inspection data, raw material transportation data, raw material storage environment data, and raw material warehousing inspection data. The first basic data on the platform is associated with a mapped identifier code. A raw material preset model is trained using the upstream raw material full-element data to obtain a raw material model, which is used to reflect the correspondence between the upstream raw material full-element data and product quality; The production environment is trained using the basic production environment data to obtain a production model, which reflects the correspondence between the basic production environment data and product quality. A specific demand preset model is trained using the basic order data to obtain an order model, which reflects the correspondence between the basic order data and product quality. The historical experience data of the product is used to train a preset historical experience model to obtain an experience model, which is used to reflect the correspondence between the historical experience data of the product and the product quality. The industry product preset model is trained using the product quality data of the same industry to obtain the industry model, which is used to reflect the correspondence between the product quality data of the same industry and product quality. Based on the output data of one or more of the raw material model, the production model, the order model, the experience model, and the industry model, a product quality prediction preset model is trained to obtain a product quality prediction model, which is used to predict the quality of the product.

2. The product quality prediction method based on identifier resolution as described in claim 1, characterized in that, The product historical experience data includes one or more of the following: historical product data of products produced within the factory, historical product quality inspection data, and historical customer complaint data. The product quality data from the same industry includes one or more of the following: industry quality information and key quality parameter information.

3. The product quality prediction method based on identifier resolution as described in claim 1, characterized in that, The first basic data and the second basic data are managed by different servers.

4. A product quality prediction system based on identifier resolution, characterized in that, The system includes: The data acquisition layer is used to acquire first basic data and second basic data in multiple ways to generate a shared resource pool, which is managed at the cloud service layer. The data acquisition layer includes an identifier resolution device, which performs identifier code mapping on the uploaded first basic data and performs identifier resolution on the first basic data sent to the application layer. The application layer obtains resources from the cloud service layer through the network transport layer; The application layer executes the method as described in any one of claims 1-3 to predict the quality of the product.

5. The product quality prediction system based on identifier resolution according to claim 4, characterized in that, The first basic data and the second basic data are managed by different servers.

6. The product quality prediction system based on identifier resolution according to claim 4, characterized in that, The data acquisition layer includes an Internet of Things (IoT) platform for acquiring second-level basic data.

Images