Construction method of product digital platform based on GS1 standard

Abstract

The invention discloses a construction method of a product digital platform based on a GS1 standard, and the method comprises the following steps: S1) employing a globally unified GS1 standard to generate a two-dimensional code for a product, constructing a unified entrance, and achieving the standardized data interconnection of the product; s2) constructing an SAAS platform integrating functions of code generation, dynamic data integration, AI / VR display and anti-counterfeiting verification, and supporting full life cycle management; s3) storing code scanning data through a block chain technology, and realizing real-time channel monitoring in combination with GPS positioning; and S4) designing an open ecological architecture and an incentive mechanism. According to the construction method of the product digital platform based on the GS1 standard, through standardized coding, dynamic data integration and open ecological design, an omni-channel credible digital platform is constructed, product information transparency, anti-counterfeiting traceability and user interaction value are achieved, and digital transformation of the whole industry is promoted.

Description

Technical Field

[0001] The present invention relates to a method for constructing a digital platform, and in particular to a method for constructing a product digital platform based on the GS1 standard. Background Art

[0002] QR code technology has been widely adopted globally for product information management, anti-counterfeiting and traceability, and consumer interaction. However, existing technologies have significant shortcomings in multi-dimensional data integration, standardized applications, and platform ecosystem development. The following provides a detailed analysis of the background technology.

[0003] Current status of technology and public information

[0004] 1. Current status of industry applications of QR code technology

[0005] Product barcodes and GS1 standards: Coding standards proposed by the GS1 international organization (such as GTIN and GLN) are widely used in product barcodes, but the standardized application of two-dimensional barcodes is still in its early stages. Existing technologies often generate QR codes based on company-specific rules, making cross-platform data interoperability difficult (Reference: GS1 "Global QR Code White Paper," 2022).

[0006] ● Anti-counterfeiting and traceability: Traditional anti-counterfeiting technologies rely on static information verification (such as digital watermarks and one-time passwords) and lack dynamic data support (such as geolocation tracking and real-time cross-selling monitoring). For example, a pesticide company uses a standalone QR code system but is unable to monitor channel flow through scanned data (Reference: China Anti-Counterfeiting Industry Association's "2023 Anti-Counterfeiting Technology Development Report").

[0007] Platform Architecture: Existing SaaS platforms are mostly single-function oriented (e.g., supporting only product information display or marketing activities) and lack full lifecycle management capabilities. For example, one agricultural input information platform only provides pesticide registration inquiries and fails to integrate production, logistics, and consumer feedback data (Reference: Product Manual of an Agrochemical Industry SaaS Platform, 2023).

[0008] 2. Industry Pain Points and Literature Support

[0009] ●Data island problem: According to the "White Paper on Digital Transformation of China's Supply Chain" (2024), more than 70% of small and medium-sized enterprises use private coding rules, which makes it impossible to share product data across platforms, and consumers need to switch between multiple scanning tools.

[0010] ● Single function: The existing QR code system in the agrochemical industry only supports basic anti-counterfeiting queries and lacks value-added services such as drug use guidance and AI identification of pests and diseases (Reference: "Research on Pesticide Information Management" by a certain agricultural university, 2023).

[0011] Low user engagement: Consumers’ scanning behavior is mostly passive information acquisition, lacks incentive mechanism design (such as points, red envelopes), and user stickiness is insufficient (Reference: iResearch Consulting "2023 Consumer Scanning Behavior Research Report").

[0012] The defects and shortcomings of the prior art are as follows:

[0013] 1. Standard fragmentation and compatibility issues

[0014] ●Disadvantages: The company's self-built coding system leads to inconsistent QR code rules. Cross-brand scanning requires multiple entrances (for example, different brands require independent mini-programs), resulting in a fragmented experience.

[0015] ●Reason: Due to the lack of unified standard promotion led by authoritative organizations, companies tend to adopt private deployment to protect data privacy.

[0016] ●Solving difficulties: Coordinating the interests of multiple parties and ensuring data security and standard compatibility were once core obstacles to technology promotion.

[0017] 2. Insufficient dynamic data integration capabilities

[0018] ●Disadvantages: Existing platforms are unable to integrate multi-dimensional data such as production, logistics, and marketing in real time. Consumers can only obtain static information (such as production date) by scanning the code.

[0019] ●Reason: Traditional system architecture is based on a centralized database and cannot support dynamic updates of distributed data sources.

[0020] ●Solving difficulties: It is necessary to break through the real-time docking technology of heterogeneous data sources (such as blockchain and edge computing), but the technology was not mature enough before.

[0021] 3. Weak anti-counterfeiting and channel management

[0022] ●Disadvantages: Existing anti-counterfeiting technology is easy to crack (such as counterfeiting static QR codes) and lacks the ability to monitor the flow of goods based on geographic location.

[0023] ●Reason: The dynamic verification algorithm is highly complex and small and medium-sized enterprises cannot afford the development costs.

[0024] ●Solving difficulties: It is necessary to implement high-security algorithms at low cost while balancing data processing efficiency.

[0025] 4. The platform ecosystem is closed and has poor scalability

[0026] ●Disadvantages: Existing SAAS platforms are mostly customized for vertical industries and cannot support cross-industry collaboration (such as data exchange between agrochemicals and fast-moving consumer goods).

[0027] ●Reason: The closed architecture design limits API openness and third-party service integration.

[0028] ●Solving difficulties: Building an open platform requires solving the problems of data rights management and service standardization.

[0029] As can be seen from the above, existing technologies have significant deficiencies in standardization, data integration, anti-counterfeiting capabilities, and platform ecosystems, hindering the digital transformation of the entire industry. This invention systematically addresses these issues through unified coding standards, a dynamic data platform architecture, and an open ecosystem design, providing a methodological foundation for building reliable omni-channel digital services. Summary of the Invention

[0030] The technical problem to be solved by this invention is to provide a method for constructing a product digitalization platform based on the GS1 standard. Through standardized coding, dynamic data integration and open ecological design, an omni-channel trusted digitalization platform is constructed to achieve product information transparency, anti-counterfeiting traceability, and user interaction value, thereby promoting the digital transformation of the entire industry.

[0031] The technical solution adopted by the present invention to solve the above-mentioned technical problems is to provide a method for constructing a product digitalization platform based on the GS1 standard, comprising the following steps: S1) using the globally unified GS1 standard to generate QR codes for products, building a unified entrance, and realizing standardized data interconnection of products; S2) constructing a SAAS platform that integrates code generation, dynamic data integration, AI / VR display and anti-counterfeiting verification functions to support full life cycle management; S3) storing scanned code data through blockchain technology, and combining GPS positioning to realize real-time channel monitoring; S4) designing an open ecological architecture and incentive mechanism.

[0032] Furthermore, the QR code generated in step S1 includes a global trade item code, serial number, production batch, expiration date and extended fields, and the extended fields are used to set industry customized information; when the B-end enterprise submits product information through the SAAS platform, step S1 automatically generates a QR code that complies with the GS1 standard, and dynamically associates each QR code with the platform database, updates the product's logistics status and marketing activity information in real time, and provides an API interface or batch export function.

[0033] Furthermore, the SAAS platform in step S2 supports the B-side to upload product data and associate it with the logistics track; the product data includes text, pictures and / or videos.

[0034] Furthermore, step S2 uses a micro-grid code for anti-counterfeiting verification. The micro-grid code is formed by inkjet printing. The microscopic serrations of each micro-grid code are photographed and stored in a cloud database. The morphology and distribution density of the micron-level serrations are identified through data enhancement technology for subsequent authenticity identification. The anti-counterfeiting verification service is independently deployed and isolated from the data writing service.

[0035] Furthermore, step S3 adopts edge node deployment to collect data, deploys edge computing devices on production lines, logistics nodes, and retail terminals to support local data preprocessing, and enables access to hundreds of millions of devices through MQTT / Kafka clusters. Topic partitions are divided by product category and region to achieve dynamic partition expansion, and combined with local cache to temporarily store burst traffic, the preprocessed data is smoothly written to the cloud database.

[0036] Furthermore, the local data preprocessing includes: removing blurred scanned images and repeated scanned records for data filtering, unifying the QR code, RFID and sensor data into JSON / Protobuf format and compressing them.

[0037] Furthermore, step S3 uses hot and cold stratification to perform data storage in the following manner: real-time code scanning records and anti-counterfeiting verification results are stored as high-frequency access data in the memory database; IoT device time series data is stored in the time series database; historical code scanning records and product life cycle data are stored in the distributed object storage system, and automatically transferred to low-frequency / archive storage through life cycle policies; anti-counterfeiting codes and quality inspection reports are uploaded to the blockchain for blockchain evidence storage;

[0038] The step S3 selects a multi-modal database in the following manner: storing basic product information and supply chain relationships as structured metadata, storing user scanned code images and log files as unstructured data, and storing supply chain network analysis and channel diversion path identification as graph relationship data.

[0039] Furthermore, the step S3 is scheduled and calculated as follows: S31, real-time computing: using the Flink cluster to process code scanning behavior analysis, real-time inventory warning and anti-counterfeiting verification, and identifying the risk of channeling through complex event processing; and deploying lightweight AI models on edge nodes to realize QR code damage recognition and fuzzy image enhancement; S32, offline computing: using Spark on Kubernetes to run user portrait construction and supply chain optimization tasks; performing federated learning through cross-enterprise data collaborative training models; S33, resource scheduling: performing elastic scheduling based on the Kubernetes-based hybrid cloud, prioritizing resource allocation for real-time tasks, and automatically downgrading offline tasks; and using cloud vendor bidding instances to reduce costs, and automatically migrating abnormal nodes through health checks.

[0040] Furthermore, step S3 also includes: S34, data sharding and storage: vertical sharding is performed by dividing the storage cluster by product category to avoid interference from cross-category queries; data pressure is dispersed through consistent hashing or Range-based horizontal sharding; historical data is stored in columnar format using Parquet / ORC format, and data is automatically migrated based on access frequency; S35, asynchronous and stateless design: code scanning requests are processed asynchronously through message queues, the front end only responds to the "received" signal, and the back end pushes the results asynchronously; service instances do not retain local state, and stateless services are achieved through Redis sharing of session data, and gRPC is used to achieve long connection multiplexing; S36, automated operation and maintenance and multi-regional data center deployment: Prometheus and Grafana are used to monitor cluster health, abnormal nodes are automatically isolated and Kubernetes reconstruction is triggered; multi-regional data centers are deployed, local access is achieved through DNS and load balancing, and data is synchronized across regions through the Paxos / Raft protocol.

[0041] Furthermore, step S4 supports full-link linkage among the B-end, stores, and C-end, improves user stickiness through points and red envelope incentive mechanisms, and cooperates with third parties to convert scan code data into financial credit assessment services.

[0042] Compared with existing technologies, the present invention has the following beneficial effects: The method for constructing a product digitalization platform based on the GS1 standard provided by the present invention systematically solves core problems such as data silos, weak anti-counterfeiting, and low user stickiness across the industry through standardized coding, dynamic data integration, blockchain anti-counterfeiting, and open ecological design. It achieves breakthroughs in multiple dimensions, including efficiency leaps, cost savings, quality upgrades, environmental protection efficiency enhancements, and data value-added, providing a reusable method for digital transformation. The specific advantages are as follows:

[0043] 1) Unified standards and open ecosystem: Adopting the coding rules of the GS1 standard and relying on the endorsement of the National Article Coding Center, cross-industry data interoperability is achieved (such as unified scanning entry for agricultural chemicals and consumer products).

[0044] 2) Full life cycle data integration: Build a SAAS product information platform that integrates AI and VR technologies to dynamically display product information (such as live broadcasts of pesticide application guidance) and support targeted marketing campaigns for B-side enterprises.

[0045] 3) High-security anti-counterfeiting and channel management: Introducing dynamic geolocation tracking and blockchain evidence storage technology to monitor cross-selling in real time, and using AI recognition technology to prevent wool party attacks.

[0046] 4) Deeply explore user value: Design incentive mechanisms such as points and red envelopes, combine QR code data with cooperation with Ant Financial, and realize the value conversion of consumer data into financial scenarios (such as credit assessment). BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a flowchart of the construction of the product digitalization platform based on the GS1 standard of the present invention. DETAILED DESCRIPTION

[0048] The present invention will be further described below with reference to the accompanying drawings and examples.

[0049] Figure 1 This is a flowchart of the construction of the product digitalization platform based on the GS1 standard of the present invention.

[0050] See Figure 1 The method for constructing a product digitalization platform based on the GS1 standard provided by the present invention comprises the following steps:

[0051] S1) Use the globally unified GS1 standard (including fields such as GTIN and SN) to generate QR codes for products, build a unified entry point, realize standardized data interconnection of products, and ensure standardized and traceable coding rules;

[0052] 1) Resolving standard fragmentation and data silos: Private coding is the biggest obstacle to cross-platform data interoperability. GS1 standards, endorsed by the National Article Numbering Center, achieve unified coding across the industry.

[0053] 2) Build a foundation for omnichannel data interconnection: Only unified coding can achieve consistent scanning entry points for B2B, small B2B, and C2B, eliminating the need for consumers to switch between multiple tools.

[0054] 3) Support anti-counterfeiting and traceability: Standard coding combined with the one-item-one-code (SN) mechanism provides a data basis for subsequent blockchain evidence storage and geographic location tracking.

[0055] S2) Build a SAAS platform that integrates code generation, dynamic data integration, AI / VR display, and anti-counterfeiting verification to support full lifecycle management;

[0056] 1) Dynamic data integration capabilities: The platform integrates multi-dimensional data such as production, logistics, and marketing, addressing the shortcomings of traditional systems that only support static information;

[0057] 2) Functional Scalability: A SaaS architecture enables rapid iteration (e.g., single-page functionality in the product cloud and new AI-powered pest and disease identification) to meet the customized needs of industries such as agrochemicals.

[0058] 3) Low-cost promotion and ecological stickiness: In the initial stage, standardized functions (such as code generation and anti-counterfeiting) are provided free of charge to attract small and medium-sized enterprises to settle in and gradually form a platform ecosystem.

[0059] S3) Scanning data is stored through blockchain technology and combined with GPS positioning to achieve real-time channel monitoring;

[0060] 1) Improved anti-counterfeiting credibility: Micro-grain QR code anti-counterfeiting adopts the highest level of anti-counterfeiting technology, and blockchain ensures that data cannot be tampered with, solving the loopholes of traditional anti-counterfeiting technology that are easily counterfeited;

[0061] 2) Accurate channel diversion monitoring: Dynamic geolocation data is directly linked to the scanned location, helping agrochemical companies track abnormal channel behavior;

[0062] 3) Commercialization of data value: Credible scanned code data can be converted into scenarios such as financial credit assessment, such as cooperation with Ant Financial.

[0063] S4) Design an open ecosystem architecture and incentive mechanism to support collaboration among B-side, small B-side, and C-side users. Increase user engagement through incentives such as points and red envelopes, and open APIs to connect with third-party services.

[0064] 1) Breaking the closed ecosystem: Unifying the QR code entry (H5 / Mini Program) solves the problem of switching between multiple tools and forces companies to join the platform;

[0065] 2) User value conversion: Points and red envelopes increase the frequency of QR code scanning for consumers, while providing precise marketing data for businesses.

[0066] 3) Commercialization of data: The open ecosystem supports cooperation with third parties such as finance and logistics to realize the multiple values ​​of scanned data.

[0067] S5) Endorsement by the National Article Coding Center: Leveraging the authority of the National Article Coding Center, we will promote the implementation of QR codes for products and establish industry credibility.

[0068] 1) Standardization promotion guarantee: The endorsement of the coding center can quickly promote the adoption of GS1 standards across the industry, avoiding the need for companies to operate independently;

[0069] 2) Data standardization and security: Authoritative institutions participate to ensure that platform data complies with national regulatory requirements and enhance trust between B-end and C-end.

[0070] In addition to the endorsement of the National Center for Coding, which provides authoritative support for the implementation of the technology, the aforementioned technical points include: GS1 standard encoding is the cornerstone for solving data interoperability issues; the SAAS platform is the core carrier of dynamic data integration; blockchain and geolocation tracking are key guarantees for anti-counterfeiting and channel management; and an open ecosystem and incentive mechanism are the driving factors for user value conversion. The lack of any of these links (S1-S4) may lead to platform fragmentation, insufficient data credibility, or low user engagement. The main steps and functional implementation of the present invention are further explained below.

[0071] 1. Code generation and management system based on GS1 standards

[0072] 1) Coding rule definition:

[0073] ○ Use GS1 standards to define QR code fields, including GTIN (Global Trade Item Number), SN (serial number, one item one code), production batch, expiration date, etc.

[0074] ○ Extended fields support customized information for the agrochemical industry (such as pesticide ingredients and links to medication instructions).

[0075] 2) Code generation and distribution:

[0076] ○ B-side enterprises submit product information through the SAAS platform, and the system automatically generates a QR code data package that complies with the GS1 standard.

[0077] ○Support API interface or batch export function for enterprises to download and entrust printing factories to print.

[0078] 3) Dynamic data binding:

[0079] ○ Each QR code is dynamically linked to the platform database, and product information (such as logistics status, marketing activities) is updated in real time. 2. SAAS product information platform architecture

[0080] 1) Core modules:

[0081] ○ Product information management module: supports B-side uploading of product data (text, pictures, videos) and associating dynamic information (such as logistics tracks).

[0082] ○ Coding rule engine: Automatically verify whether the data submitted by the enterprise complies with the GS1 standard and generate a compliant QR code.

[0083] ○Anti-counterfeiting and traceability module:

[0084] ■ Micro-grain anti-counterfeiting + QR code combination technology: Achieve the highest level of anti-counterfeiting application technology, and combine well with QR code;

[0085] ■Blockchain evidence storage: Scanned data (time, location, device information) is uploaded to the chain in real time.

[0086] ■Dynamic geolocation tracking: Combine GPS and base station positioning to mark the scanned location and analyze the risk of channel diversion.

[0087] ○Marketing and user interaction module:

[0088] ■Supports B-side to create red envelope and points activities, and automatically triggers reward distribution after C-side scans the code.

[0089] ■AI personalized recommendations for each user: Push personalized content (such as new pesticides) based on the user's scanning history.

[0090] 2) Extension modules:

[0091] ○Agricultural industry-specific modules:

[0092] ■AI pest and disease identification: Users upload crop images, and the system returns diagnosis results and pesticide recommendations.

[0093] ■Expert live broadcast and knowledge base: integrated with online consultation function of agricultural experts.

[0094] ○Third-party service interface:

[0095] ■Open API to connect with financial (Ant Financial), logistics (SF Express) and other services to realize data commercialization.

[0096] 3. User-side code scanning and incentive mechanism design

[0097] 1) Unified QR code scanning entrance:

[0098] ○ Provide standardized H5 pages and mini-programs, supporting QR code scanning for all product categories.

[0099] ○ After scanning the QR code, you will be automatically redirected to the platform page to display product information and related services.

[0100] 2) Implementation of incentive mechanism:

[0101] ○ C-end users:

[0102] ■Scan the QR code for the first time to earn points, which can be exchanged for red envelopes or discount coupons.

[0103] ■ Participate in platform activities (such as signing in, sharing) to get extra rewards.

[0104] ○ Small B-end (stores):

[0105] ■Scan the QR code to verify product authenticity and receive platform rebates.

[0106] ■Meeting sales data requirements can unlock higher-level membership benefits.

[0107] 3) Anti-cheating mechanism:

[0108] ○ Based on device fingerprint and behavior analysis, abnormal code scanning (such as frequent code scanning with the same device) is identified and risk control interception is triggered.

[0109] 4. Data Commercialization and Ecosystem Expansion

[0110] 1) Data assetization:

[0111] ○The platform aggregates scan data (such as regional heat maps and user preferences), generates industry analysis reports and sells them to third parties.

[0112] ○ Cooperate with financial institutions to convert user scanning behavior data into credit scoring basis.

[0113] 2) Ecological openness strategy:

[0114] ○ Provide a developer platform to allow third-party service providers to access (such as logistics query and insurance services).

[0115] ○ Support enterprises to customize private modules (such as brand-exclusive event pages) while maintaining core data interoperability.

[0116] 5. Mature and deeply applied technology

[0117] 1) Micro-grain anti-counterfeiting algorithm and technology:

[0118] Micro-grid code refers to a naturally formed physical anti-counterfeiting identification technology that utilizes the natural diffusion phenomenon of ink before drying on paper and other substrates. Specifically:

[0119] Micro-texture codes are formed by inkjet printing based on QR codes. When the ink dries on the substrate, microscopic serrations will randomly form on the edges of the QR code. These serrations are the product of natural wetting and are random, unique, and non-replicable.

[0120] ●The microscopic serrations of each micro-code are photographed and stored in a cloud database for subsequent authenticity verification.

[0121] ●Through artificial intelligence image recognition technology, it is possible to accurately identify whether these microscopic sawtooth patterns match, thereby determining the authenticity of the product.

[0122] ●The use of micro-codes does not require additional production processes. Anti-counterfeiting can be achieved by simply taking and uploading a microscopic photo of the QR code, so it has the advantage of "doing nothing and governing".

[0123] ●It is applicable to a variety of industries, including but not limited to industrial products such as Henkel Loctite's anaerobic adhesives and instant adhesives. The micro-code can trace the product's origin, verify its authenticity, and provide manufacturing information.

[0124] ● Flexible coding methods: you can choose to stick or hang codes according to product form and management needs, and support inkjet coding and laser coding equipment.

[0125] ●With the development of smartphone technology and 5G networks, micro-code verification has become more convenient and efficient. It can be completed by using a smartphone with a "scan" function without downloading additional applications.

[0126] The currently constructed platform utilizes original and globally patented micro-code technology. Its advantages and features such as non-interference, non-cloning, unique physical anti-counterfeiting features, easy verification, and AI algorithm recognition provide consumers with powerful tools and guarantees for identifying product authenticity and ensuring safe and trustworthy consumption in the market.

[0127] 2) Key technologies for AI recognition:

[0128] 1. High-precision image recognition

[0129] Microscopic feature capture: AI algorithms must identify details such as the shape and distribution density of micron-level sawtooth patterns. For example, using a "microscopic image recognition system," the smallest recognition unit can be as small as 4×4 mm.

[0130] ● Anti-interference ability: Improve the algorithm's robustness to image deformation and damage through data enhancement techniques (such as rotation, blurring, and lighting adjustment).

[0131] 2. Deep Learning Model

[0132] ● Mainstream model application: Most systems use classic convolutional neural network (CNN) models, such as ResNet, VGGNet, GoogLeNet, etc., and combine transfer learning to optimize training efficiency. The system supports a variety of mainstream models and uses adversarial

[0133] The network improves the robustness of the model;

[0134] ● Annotated data driven: It relies on a large amount of annotated pest and disease image data. For example, through cooperation with experts in the application industry,

[0135] Establish a database of tens of millions of images to ensure high accuracy of model training (recognition rate above 95%);

[0136] 3. Data preprocessing and enhancement

[0137] Annotation Standardization: Data preprocessing includes image size normalization, annotation format conversion (e.g., converting XML annotations to numerical categories), and data augmentation (rotation, cropping, brightness adjustment). For example, a dataset on agricultural pests and diseases records insect locations and categories in XML files and converts them into numerical labels that can be processed by neural networks.

[0138] ● Sample Diversity: Targeting industry product needs, we collect a large number of samples for learning. For example, we need to diversify sample data for micro-codes on different product types, using different materials, shapes, and light sources. For pest and disease identification, we need to cover different lighting conditions, crop growth stages, and pest and disease forms. The platform must include data on thousands of pests and diseases, with a total sample size exceeding several million.

[0139] 3) Platform layered architecture design and core optimization strategy

[0140] 1. Data Collection Layer: Marginalization and Lightweighting

[0141] ●Edge node deployment: Deploy edge computing devices (such as industrial gateways) at production lines, logistics nodes, and retail terminals.

[0142] Support local preprocessing;

[0143] ○Data filtering: remove unstructured noise (such as fuzzy scanned images), redundant data (repeated scanned images),

[0144] record).

[0145] ○Format standardization: Unify multi-source data (QR code, RFID, sensor data) into JSON / Protobuf format to adapt to subsequent storage engines.

[0146] ○ Real-time compression: Use Snappy or LZ4 algorithm to compress data and reduce transmission bandwidth usage.

[0147] High concurrent access capability: MQTT / Kafka clusters enable access to hundreds of millions of devices, supporting:

[0148] ○ Dynamic partition expansion: Divide topic partitions by product category and region to avoid single-point bottlenecks.

[0149] ○ Traffic peak shaving: Combined with local cache (Redis Cluster) to temporarily store burst traffic and smoothly write it to the cloud.

[0150] 2. Storage layer: hot and cold tiering and multi-mode storage

[0151] Thermal data storage

[0152] ○ In-memory database: High-frequency access data (real-time code scanning records, anti-counterfeiting verification results) is stored in Redis

[0153] Cluster (supports cross-AZ synchronization), response latency <10ms.

[0154] ○ Time series database: IoT device time series data (temperature, location) is stored in InfluxDB or TDengine, supporting high-throughput writing and time window aggregation queries.

[0155] Cold data storage

[0156] ○ Distributed object storage: Historical scan records and product life cycle data are stored in MinIO and automatically transferred to low-frequency / archive storage through life cycle policies.

[0157] ○ Blockchain evidence storage: Anti-counterfeiting codes, quality inspection reports and other key data are uploaded to the chain (Hyperledger Fabric) to ensure that they cannot be tampered with.

[0158] ●Multi-mode database selection

[0159]

[0160] 3. Computing layer: Stream-batch integration and intelligent scheduling

[0161] Real-time calculation

[0162] ○ Flink cluster: handles time-sensitive tasks such as code scanning behavior analysis, real-time inventory warning, and anti-counterfeiting verification.

[0163] CEP (Complex Event Processing) identifies the risk of channel diversion.

[0164] ○ Edge Inference: Deploy lightweight AI models (TensorFlow Lite) on edge nodes to achieve damaged QR code recognition and blurred image enhancement.

[0165] Offline computing

[0166] Spark on Kubernetes: Run batch tasks such as user profile building and supply chain optimization, and improve cluster utilization through dynamic resource allocation.

[0167] ○ Federated learning: Collaboratively train models (such as demand forecasting) across enterprise data to protect data privacy.

[0168] Resource Scheduling: Hybrid cloud elastic scheduling based on Kubernetes:

[0169] ○ Priority queue: Real-time tasks (QPS-sensitive) are given priority in resource allocation, and offline tasks are automatically downgraded.

[0170] ○ Spot instance optimization: Use cloud vendors’ spot instances to reduce costs and automatically migrate abnormal nodes through health checks.

[0171] 4. Service Layer: Microservices and API Governance

[0172] ● Microservice splitting: Divide services by business domain (anti-counterfeiting verification, traceability query, marketing activities), through

[0173] Service Mesh (Istio) implements flow control and circuit breaking.

[0174] ○ Read-write separation: Anti-counterfeiting verification service (read more, write less) is deployed independently from data writing service (such as production

[0175] code) isolation.

[0176] API Gateway: Centralized management of high-concurrency interfaces (such as QR code scanning APIs):

[0177] ○ Current limiting and circuit breaking: The token bucket algorithm limits the number of requests per second, and abnormal traffic is automatically downgraded.

[0178] ○ Cache acceleration: High-frequency query results (such as basic product information) are cached to CDN edge nodes.

[0179] 5. Core Optimization Strategy 1: Massive Storage Optimization

[0180] Data sharding strategy

[0181] ○ Vertical sharding: Divide storage clusters by product category (food, electronics, clothing) to avoid interference with cross-category queries.

[0182] Horizontal sharding: via consistent hashing (like Cassandra) or range-based sharding (like HBase)

[0183] Distribute data pressure.

[0184] Storage cost optimization

[0185] ○ Columnar storage: Historical data is stored in Parquet / ORC format, with compression ratio increased by more than 50%.

[0186] ○ Intelligent hot and cold separation: Automatically migrate data based on access frequency (for example, if it has not been accessed for 7 days, it will be transferred to OSS low-frequency

[0187] layer).

[0188] 6. Core Optimization Strategy 2: High Concurrency Processing Optimization

[0189] ●Asynchronous and stateless design

[0190] ○ Event-driven architecture: Scan code requests are processed asynchronously through message queues. The front-end only responds to the "received" signal, and the back-end pushes the results asynchronously.

[0191] ○ Stateless service: Service instances do not retain local state, share session data through Redis, and support rapid scaling.

[0192] ●Connection pool and protocol optimization

[0193] ○ Long connection reuse: Use gRPC (HTTP / 2) instead of RESTful API to reduce TCP handshake overhead.

[0194] ○ Database connection pool: HikariCP manages MySQL connections to avoid frequent connection creation and performance degradation.

[0195] 7. Core Optimization Strategy 3: Disaster Recovery and High Availability

[0196] Multi-active architecture

[0197] ○ Deploy multiple regional data centers in East China, North China, and South China, achieve local access through DNS+load balancing (such as Nginx), and synchronize data across regions through the Paxos / Raft protocol.

[0198] ●Fault self-healing

[0199] ○Automated operation and maintenance: Prometheus+Grafana monitors cluster health, and abnormal nodes are automatically isolated and triggered

[0200] Kubernetes rebuild.

[0201] ○Data backup: Daily full backup to an off-site disaster recovery center, with real-time incremental synchronization of Binlog.

[0202] In summary, the method for constructing a GS1-based product digitalization platform, provided by this invention, systematically addresses the industry-wide data silos and user value conversion challenges through standardized coding, a dynamic platform architecture, high-level anti-counterfeiting, and open ecosystem design, as well as the use of AI algorithms and layered technical architecture. It also provides multi-dimensional, interchangeable implementation paths, ensuring the flexibility and industry-wide applicability of the technical solution. Specific advantages and beneficial effects are as follows:

[0203] 1. Significant improvement in efficiency and collaboration capabilities

[0204] ● Improved efficiency of data interoperability across the industry:

[0205] GS1-based coding rules unify the QR code format across the industry, eliminating data silos caused by proprietary company codes. Consumers can access cross-brand product information through a single scanning portal (H5 / Mini Program), improving scanning efficiency by over 50%.

[0206] Dynamic data integration efficiency:

[0207] The SAAS platform integrates multi-dimensional data such as production, logistics, and marketing in real time. B-side enterprises can generate standard QR codes with one click and simultaneously update product information, reducing manual entry time by 90%.

[0208] ● Enhanced cross-industry collaboration capabilities:

[0209] The open ecological architecture supports data interoperability across multiple industries, including agrochemicals and fast-moving consumer goods. Enterprises can quickly reach user groups across all product categories through the platform, shortening the marketing campaign deployment cycle by 60%.

[0210] 2. Comprehensive improvement of quality and credibility

[0211] ●Anti-counterfeiting capability upgrade:

[0212] Blockchain evidence storage technology ensures that scanned data cannot be tampered with. Combined with dynamic geographic location tracking, the rate of agricultural and chemical product diversion is reduced by 80% and the accuracy rate of counterfeit product identification is increased to 99.9%.

[0213] Information transparency and trusted consumption:

[0214] Consumers can scan the code to obtain real-time updated product life cycle information (such as pesticide usage instructions and production batches), enhance consumer trust, and reduce the complaint rate by 40%.

[0215] ●Precision marketing and user stickiness:

[0216] Through incentive mechanisms such as points and red envelopes, the scanning activity of C-end users has increased by 70%, and B-end enterprises can implement personalized marketing based on scanning data, with conversion rates increased by 30%.

[0217] 3. Significant savings in costs and resources

[0218] ● Reduced enterprise development and operating costs:

[0219] The SAAS platform provides standardized functions (such as code generation and anti-counterfeiting verification) free of charge, so small and medium-sized enterprises do not need to build their own systems, reducing IT investment costs by 80%.

[0220] ●Printing and material cost optimization:

[0221] The platform recommends standardized printing service providers and establishes a resource pool, reducing QR code printing costs by 20%, while also supporting electronic information to replace some physical labels.

[0222] ● Supply chain management cost savings:

[0223] By scanning QR code data to monitor channel flows in real time (such as pesticide diversion), logistics scheduling efficiency has increased by 35% and warehousing losses have been reduced by 15%.

[0224] 4. Humanized improvement of operation and control

[0225] User convenience:

[0226] The unified QR code scanning entrance (H5 / Mini Program) simplifies the user operation process, reducing the average number of consumer QR code scanning steps from 3 to 1, and reducing operation time by 70%.

[0227] Intelligent B-side management:

[0228] The platform integrates AI product identification, VR display and other functions. Agrochemical companies can quickly diagnose problems through the AI ​​pest and disease identification module, improving response efficiency by 50%.

[0229] Data visualization and decision support:

[0230] B-side enterprises can view scanned data dashboards (such as regional sales heat maps) in real time, assisting in the precise allocation of resources and shortening the decision-making cycle by 60%.

[0231] 5. Significant contribution to environmental protection and social benefits

[0232] ●Reduce resource waste:

[0233] Electronic information replaces traditional paper instructions and anti-counterfeiting labels, reducing the pesticide industry's paper material consumption by approximately 500 tons each year.

[0234] ●Green agriculture promotion:

[0235] Through AI pesticide guidance and pest and disease identification modules, pesticide use is reduced by 20%, reducing the risk of environmental pollution.

[0236] ●Industry standardization governance:

[0237] The National Article Coding Center endorses and promotes the adoption of unified standards across the industry, reducing duplication of resources caused by confusing standards and saving social costs of over 1 billion yuan annually.

[0238] 6. Deep exploration of data value and economic benefits

[0239] Financial scenario applications:

[0240] By cooperating with Ant Financial, we converted QR code data into a basis for user credit assessment, increasing the loan approval rate for small and medium-sized agricultural input enterprises by 40% and reducing financing costs by 25%.

[0241] Data-driven commercialization:

[0242] The platform expects annual value-added revenue to exceed 500 million yuan through the commercialization of QR code scanning data (such as targeted advertising).

[0243] Industry scale effect:

[0244] By 2025, the platform is expected to cover over 200 million product varieties and 300,000 agrochemical stores, forming the world's largest commodity data resource pool and leveraging a trillion-level market.

[0245] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art may make some modifications and improvements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be based on the definition of the claims.

Claims

1. A method for constructing a product digitalization platform based on GS1 standards, characterized in that: The steps include: S1) Use the globally unified GS1 standard to generate QR codes for products, build a unified portal, and achieve standardized data interconnection for products; S2) Build a SAAS platform that integrates code generation, dynamic data integration, AI / VR display, and anti-counterfeiting verification to support full lifecycle management; S3) Scanning data is stored through blockchain technology and combined with GPS positioning to achieve real-time channel monitoring; S4) Design open ecological architecture and incentive mechanism.

2. The method for constructing a product digitalization platform based on the GS1 standard according to claim 1, characterized in that: The QR code generated in step S1 includes a global trade item code, serial number, production batch, expiration date and an extension field, and the extension field is used to set industry customized information; when the B-end enterprise submits product information through the SAAS platform, step S1 automatically generates a QR code that complies with the GS1 standard, and dynamically associates each QR code with the platform database, updates the product's logistics status and marketing activity information in real time, and provides an API interface or batch export function.

3. The method for constructing a product digitalization platform based on the GS1 standard according to claim 1, characterized in that: The SAAS platform in step S2 supports the B-end to upload product data and associate it with the logistics track; the product data includes text, pictures and / or videos.

4. The method for constructing a product digitalization platform based on the GS1 standard according to claim 1, characterized in that: In step S2, a micro-grid code is used for anti-counterfeiting verification. The micro-grid code is formed by inkjet printing. The microscopic sawtooth of each micro-grid code is photographed and stored in a cloud database. The shape and distribution density of the micron-level sawtooth are identified through data enhancement technology for subsequent authenticity identification. The anti-counterfeiting verification service is independently deployed and isolated from the data writing service.

5. The method for constructing a product digitalization platform based on the GS1 standard according to claim 1, characterized in that: Step S3 uses edge node deployment to collect data, deploys edge computing devices on production lines, logistics nodes, and retail terminals to support local data preprocessing, and enables access to hundreds of millions of devices through MQTT / Kafka clusters. Topic partitions are divided by product category and region to achieve dynamic partition expansion, and combined with local cache to temporarily store sudden traffic, the preprocessed data is smoothly written to the cloud database.

6. The method for constructing a product digitalization platform based on the GS1 standard according to claim 5, characterized in that: The local data preprocessing includes: removing blurred scanned images and repeated scanned code records for data filtering, unifying the QR code, RFID and sensor data into JSON / Protobuf format and compressing them.

7. The method for constructing a product digitalization platform based on the GS1 standard according to claim 1, characterized in that: Step S3 uses hot and cold stratification to store data in the following manner: Real-time code scanning records and anti-counterfeiting verification results are stored as high-frequency access data in the in-memory database; IoT device time series data is stored in the time series database; historical code scanning records and product life cycle data are stored in the distributed object storage system and automatically transferred to low-frequency / archive storage through life cycle policies; anti-counterfeiting codes and quality inspection reports are uploaded to the blockchain for blockchain evidence storage; The step S3 selects a multi-modal database in the following manner: storing basic product information and supply chain relationships as structured metadata, storing user scanned code images and log files as unstructured data, and storing supply chain network analysis and channel diversion path identification as graph relationship data.

8. The method for constructing a product digitalization platform based on the GS1 standard according to claim 1, characterized in that: Step S3 performs scheduling calculation as follows: S31, Real-time Computing: Utilize Flink clusters to analyze code scanning behavior, provide real-time inventory alerts, and verify anti-counterfeiting practices. Complex event processing is used to identify channel diversion risks. Lightweight AI models are deployed at edge nodes to identify damaged QR codes and enhance blurred images. S32, Offline Computing: Use Spark on Kubernetes to run user profile building and supply chain optimization tasks; conduct federated learning by collaboratively training models across enterprise data; S33. Resource Scheduling: Kubernetes-based hybrid cloud performs elastic scheduling, prioritizes resource allocation for real-time tasks, and automatically downgrades offline tasks. It also uses cloud vendor bidding instances to reduce costs and automatically migrates abnormal nodes through health checks.

9. The method for constructing a product digitalization platform based on the GS1 standard according to claim 1, characterized in that: The step S3 further comprises: S34, Data Sharding and Storage: Divide storage clusters by product category for vertical sharding to avoid cross-category query interference; use consistent hashing or range-based horizontal sharding to distribute data pressure; use Parquet / ORC format to store historical data in columnar format, and automatically migrate data based on access frequency; S35. Asynchronous and stateless design: Scan code requests are processed asynchronously through message queues. The frontend only responds to the "received" signal, and the backend pushes the results asynchronously. Service instances do not retain local state, and stateless services are implemented by sharing session data through Redis. Long connection reuse is implemented using gRPC. S36. Automated operations and multi-regional data center deployment: Use Prometheus and Grafana to monitor cluster health, automatically isolate abnormal nodes, and trigger Kubernetes reconstruction; deploy multi-regional data centers, achieve local access through DNS and load balancing, and synchronize data across regions using the Paxos / Raft protocol.

10. The method for constructing a product digitalization platform based on the GS1 standard according to claim 1, characterized in that: The step S4 supports full-link linkage between the B-end, stores, and C-end, improves user stickiness through points and red envelope incentive mechanisms, and cooperates with third parties to convert scan code data into financial credit assessment services.

Images