A blockchain-based whole-process tracing method and system for producing preserved eggs

Abstract

The present application relates to product data management technical field, specifically to a kind of pratacookiess production whole-process traceability method and system based on block chain, comprising: raw material egg batch is registered, and in production, key parameters such as pit storage temperature and humidity, material liquid alkalinity are aggregated and captured, and tamper-proof process witness is generated;Witness is handled using the dual-track governance paradigm consisting of execution track and learning track, execution track is intervened in real time according to operation benchmark, and learning track is iteratively optimized to predictive model;Finally, process data is associated with subsequent supply chain turnover events, and is aggregated to form supply chain decision sand table that can be used for simulation deduction.The present application combines the fine data management and closed-loop control of production process, through the data association and sand table deduction throughout the whole chain, improves isolated production management to integrated supply chain collaborative decision-making level, and systematically improves the quality control precision and management decision level of non-standard product supply chain.

Description

Technical Field

[0001] This invention relates to the field of product data management technology, specifically to a blockchain-based method and system for tracing the entire production process of preserved eggs. Background Technology

[0002] In modern enterprise production and operation management, especially in the field of product data management, leveraging information technology to achieve data tracking throughout the entire product lifecycle has become crucial for improving management efficiency and data transparency. Among existing technologies, blockchain-based traceability systems provide a distributed data recording framework. By creating tamper-proof records for products and their circulation processes, they support cross-departmental and cross-enterprise supply chain data collaboration and monitoring. Simultaneously, combined with automated data collection equipment and RFID technologies, they enable standardized data collection and refined tracking management of materials and finished products, providing fundamental data support for quality control and logistics scheduling.

[0003] However, existing management systems reveal core technical problems when applied to the production of non-standard products with special processes: a lack of dynamic and reliable structured data management capabilities for key process parameters within the production process. Taking preserved egg production as an example, its core quality depends on special processes such as pickling and cooking. However, existing systems typically stop at batch management of raw materials and finished products, failing to delve into the internal process flow to effectively collect and correlate data on key control points that determine the final product quality, such as the ratio of raw materials to liquids, storage environment, and pickling time. This lack of data management leads to blind spots in quality control during production, making it difficult to accurately define production responsibilities and creating a data-driven dilemma for process optimization and management decisions.

[0004] Therefore, this invention proposes a blockchain-based method and system for tracing the entire production process of preserved eggs. Summary of the Invention

[0005] The purpose of this invention is to provide a blockchain-based method and system for tracing the entire production process of preserved eggs. By deeply contextualizing the production batches with the process environment and generating tamper-proof encrypted witnesses for key dynamic parameters in the production process, and by utilizing a dual-track governance paradigm to achieve real-time intervention and forward-looking optimization of the production process, this invention solves the problems mentioned in the background technology that traditional traceability systems cannot penetrate into the process flow, resulting in blind spots in quality control and insufficient data support for process optimization and management decisions.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A blockchain-based method for tracing the entire production process of preserved eggs includes: Register a unique production instruction for each batch of raw egg, generate an initial job record containing the initial material file, and record it on the distributed operations ledger; The initial work records are associated with the storage unit and process formula as production factors. By strategically anchoring the execution resources, work-in-process operation files associated with the production factors are generated. The process involves aggregating the work-in-process records with captured operational parameters and generating an immutable witness using an asymmetric encryption algorithm. The operational parameters include the alkalinity of the liquid material, the temperature and humidity of the storage cellar, and the maturation time. The immutable witness is submitted to a dual-track governance paradigm for processing. The dual-track governance paradigm includes an execution track and a learning track. Execution track: The digital agent generates and dispatches an immediate correction instruction to intervene in the current batch in real time based on the comparison result between the witness and the current operating benchmark. Learning track: The witness data is input into the predictive operation model to perform real-time inference to predict process deviations, and the model parameters are iteratively optimized based on the witness data and subsequent intervention and quality results. Based on real-time correction instructions and optimized model parameters, the flow events of subsequent supply chain nodes are correlated to build product data management archives, and finally aggregated to form a predictable supply chain decision-making sandbox.

[0007] Preferably, the initial material file specifically includes: creating a root data object as the starting point for product data management by recording initial transactions on a distributed operating ledger. The data structure of the root data object includes: raw egg purchase order number associated with the procurement process, enterprise digital voucher identifying the supplier, geographical location data of the raw material origin, purchase quantity and unit of measurement of the batch size, and hash value of electronic quarantine certificate issued by a third-party quarantine agency. The hash value is used for subsequent certificate file verification.

[0008] Preferably, the production factor association specifically includes: creating a relationship binding transaction on the distributed operations ledger, injecting production context information into the batch product data management record; the input field of the relationship binding transaction references the hash value of the transaction where the initial job record is located, and the data payload field records the physical asset number of the allocated storage unit, the version identifier of the adopted process formula, the employee digital identity certificate of the production operator on duty, and the planned storage start time; the output field of the transaction generates a unique identifier for the work-in-process operation file that is referenced in subsequent production stages.

[0009] Preferably, the immutable witness specifically includes: periodically performing data snapshot operations based on a preset time frequency to continuously add dynamic process data; the data snapshot collects the pH value of the liquid material, the temperature value of the storage environment, and the humidity value of the storage environment at the current time point through IoT sensors, and obtains the cumulative duration from the start time of storage through a timer; the collected four parameter values ​​are combined with the unique identifier of the work-in-process file to form a data block in JSON format; the hash value of the data block is calculated, and the hash value is digitally signed using the private key bound to the storage unit to form the immutable witness.

[0010] Preferably, the digital agent specifically includes: receiving the tamper-proof witness and parsing the alkalinity of the liquid and the temperature and humidity parameters of the storage cellar; comparing the parsed parameter values ​​with multidimensional thresholds stored in the operation benchmark parameter library one by one, wherein the multidimensional thresholds define the normal fluctuation range of each parameter at different maturation stages; when any parameter value is detected to exceed the normal fluctuation range of its corresponding stage, a structured real-time correction instruction is immediately generated, wherein the data structure of the real-time correction instruction includes the network address of the target environmental control device, the parameter item to be adjusted, and the target value to be adjusted.

[0011] Preferably, the predictive operation model specifically includes: extracting full-cycle historical witness data of completed production batches from the distributed operation ledger to form a dataset containing multiple time series; using the finished product yield rate recorded by the quality inspection department for each batch as a label for supervised learning and associating it with the corresponding time series dataset; training a long short-term memory network containing an attention mechanism using the labeled dataset; deploying the trained model, receiving the witness data sequence generated for the current production batch, and outputting in real time the probability prediction value of process deviations occurring in the next 24 hours.

[0012] Preferably, the dual-track governance paradigm specifically includes: a data feedback loop connecting the execution track and the learning track; the loop is configured to: after the execution track issues an immediate correction instruction, continuously capture the sequence of changes in operational parameters over a period of time after the instruction is executed; package the issued instruction content, the sequence of parameter changes after execution, and the final quality inspection results of the batch into a structured experience dataset; and use the experience dataset as a new training sample to input into the predictive operation model of the learning track to perform incremental training of the model.

[0013] Preferably, the association of subsequent supply chain node transfer events specifically includes: after a batch of products completes the production process, an on-chain operation is performed at each handover point of responsibility in the subsequent processes of unloading, cleaning, grading, packaging, warehousing, unloading, and logistics transportation; the on-chain operation creates a new transfer transaction, in which the event type, the digital identity of the operator, the geographical location data of the location where the operation occurs, the quantity of products transferred, and the real-time temperature of the temperature-controlled logistics vehicle are recorded in the form of key-value pairs; the new transaction contains a hash pointer pointing to its preceding transaction, thus constructing a logistics and ownership change chain on the ledger.

[0014] Preferably, the supply chain decision-making sandbox specifically includes: performing a simulation involving the following steps: calling a data aggregation function to extract historical batch cost and time data in production, logistics, and warehousing from a distributed operations ledger, and constructing a parameterized supply chain digital twin model based on this; receiving externally input simulation scenarios through a command-line interface, wherein the simulation scenario is a set of adjustments to multiple parameters in the model; running Monte Carlo simulation calculations in the digital twin model, and quantitatively outputting the probability distribution and expected value of the simulation scenario for the total cost, delivery cycle, and expected profit margin of the batch products.

[0015] A blockchain-based traceability system for the entire production process of preserved eggs includes: Registration record module: Used to register a unique production instruction for each batch of raw egg, generate an initial operation record containing the initial material file, and record it on the distributed operations ledger; Production factor association module: used to associate the initial operation record with the storage unit, process formula and operator identity to generate work-in-process operation files associated with the production factors; Witness generation module: aggregates the work-in-process operation file with the captured material liquid alkalinity, cellar temperature and humidity, and maturation time; generates an immutable witness through an asymmetric encryption algorithm; Dual-track governance module: Execution track: The digital agent generates and dispatches immediate corrective instructions for real-time intervention in the current batch based on the comparison results between the witness and the current operational benchmark; Learning track: The witness data is input into the predictive operation model to perform real-time inference to predict process deviations, and the model parameters are iteratively optimized based on the witness data and subsequent intervention and quality results. Product Data Management Module: Based on real-time correction instructions and optimized model parameters, it associates the flow events of subsequent supply chain nodes and forms a predictable supply chain decision-making sandbox through data aggregation to jointly complete the batch product data management.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention constructs a continuous, high-fidelity dynamic process dataset for each production batch through the synergy of contextualized binding and tamper-proof witnessing. Based on this data, the execution track in the dual-track governance paradigm is not simply a threshold alarm, but constitutes a closed-loop control system for real-time feedback and intervention. When faced with batch-to-batch differences in the physical properties of raw materials or fluctuations in external environmental factors, it does not rely on fixed static process instructions, but accurately captures deviations in actual process parameters through tamper-proof witnessing, and generates targeted corrective instructions from the execution track. This enables the production system to proactively suppress the impact of various disturbances on the uniformity of final product quality, demonstrating higher process control robustness.

[0017] 2. This invention combines immutable witnessing with the learning track in a dual-track governance paradigm. Immutable witnessing provides high-fidelity process data sources for machine learning, while the learning track, based on this longitudinal historical data, establishes a quantitative mapping model from multi-dimensional process parameters to final product quality indicators. This model not only optimizes static process parameter benchmarks based on historical data, but more importantly, it uses the intervention measures of the execution track and their resulting process response data as key training samples. This enables the model to learn and solidify effective strategies for dealing with unexpected deviations, thereby dynamically adjusting process path recommendations. This adaptive optimization capability allows the process model to continuously evolve to adapt to long-term changes in raw materials, equipment, and the environment.

[0018] 3. This invention continuously links the immutable witnessing of the production process with subsequent supply chain events, and constructs a supply chain decision-making sandbox at the top level, transforming underlying technical data into top-level insights. Decision-makers can dynamically stratify the value of products and differentiate pricing based on the refined process quality indicators recorded in the witnessing data to adapt to the needs of different markets. At the risk management level, because the decision-making sandbox embeds the efficiency and quality constraints of the real production process into the model, it can perform more precise supply chain risk stress tests and strategy simulations when facing external shocks such as demand fluctuations or logistical disruptions. It outputs cost and delivery cycle impact assessments that are more valuable than traditional systems, thereby supporting enterprises in developing more flexible response strategies. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the overall method of a blockchain-based traceability method for the entire production process of preserved eggs, as proposed in this invention application. Figure 2 This is a schematic diagram of the structure of a blockchain-based traceability system for the entire production process of preserved eggs, as proposed in this invention application. Figure 3 This is a flowchart illustrating the specific method of the supply chain decision-making sandbox in the embodiments of this invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] This invention provides a blockchain-based method for tracing the entire production process of preserved eggs, the technical solution of which is as follows: A blockchain-based method for tracing the entire production process of preserved eggs, the specific implementation steps of which include: S1. Register a unique production instruction for each batch of raw egg to be produced, including an initial material file that records static information such as the source of the raw materials, the supplier, and the batch number. Then, package the production instruction and the initial material file into an initial operation record and broadcast it to the distributed operation ledger for on-chain storage.

[0022] S2. Strategically anchor the initial work records that have been put on the chain to the physical storage units occupied by this production, the process formula version number adopted, and the digital identity of the operator; generate work-in-process operation files associated with production factors by creating binding transactions on the distributed operations ledger.

[0023] S3. During the aging process, key operational parameters such as the alkalinity of the liquid, the temperature and humidity of the aging environment, and the cumulative aging time are periodically captured by IoT sensors. The dynamic parameters are aggregated with the work-in-process operation files, and signed using an asymmetric encryption algorithm to generate an immutable process witness, which is then submitted to the distributed operation ledger.

[0024] S4. Submit the generated process witnesses to the dual-track governance paradigm for processing. On the execution track, the digital agent compares the real-time parameters in the witnesses with the preset operational benchmarks. If a deviation is found, it automatically generates and dispatches an immediate corrective instruction to intervene in the current production environment. On the learning track, the witness data sequence is input into the predictive operation model in real time for inference to predict process deviations, and the model parameters are iteratively optimized based on the witness data and subsequent intervention and quality results.

[0025] S5. Based on the real-time correction instructions recorded by the execution track and the model parameters optimized by the learning track, the system continuously associates and uploads data to the chain for subsequent supply chain nodes such as batch product release, grading, packaging, and logistics; it aggregates data from the entire process from production to sales to form a predictable supply chain decision-making sandbox for strategy simulation.

[0026] Example 1: This embodiment provides a specific application of a blockchain-based traceability method for the entire production process of preserved eggs. A company utilizes a distributed ledger environment based on the Hyperledger Fabric consortium blockchain and equips each storage unit in the storage workshop with industrial-grade pH and temperature / humidity sensors to perform fully digital and intelligent management of each batch of premium preserved eggs, from raw material acceptance to finished product delivery. For details, please refer to... Figure 1 .

[0027] Furthermore, a unique production instruction is registered for each batch of raw egg material to be produced, including an initial material file that records static information such as the source of the raw material, supplier, and batch number. Subsequently, the production instruction and the initial material file are packaged into an initial operation record and broadcast to the distributed operational ledger for on-chain storage and notarization. Corresponding to step S1 above, the specific process is as follows: The warehouse manager executes a production instruction creation process for a newly arrived batch of duck eggs. During this process, a root data object is generated as the initial material file. This object includes the purchase order number, the supplier's digital certificate ("Luyuan Ecological Farm"), the geographical location data of the raw material's origin, the purchase quantity, and a PDF quarantine certificate issued by the local quarantine department. The SHA-256 hash value of this certificate is then calculated and recorded. The initial transaction is submitted to the blockchain network, obtaining a unique transaction ID, which serves as the starting point for batch product data management.

[0028] By specifically defining the content of the initial material file, a complete and tamper-proof digital identity starting point is established for batch products, and the basic attributes of raw materials and vouchers are reliably solidified, providing a unique and reliable traceability root for data association in all subsequent stages.

[0029] Furthermore, the initial work records already on the blockchain are strategically anchored to the physical storage units occupied by this production, the process formula version number used, and the digital identity of the operator; by creating binding transactions on the distributed operations ledger, work-in-process operation files associated with production factors are generated. Corresponding to step S2 above, the specific process is as follows: The production workshop supervisor initiates a relationship binding transaction by binding batches to production resources. The input field references the ID of the initial transaction, and its data payload records the storage unit number "A-07", the process formula version number "Premium-V3.2", the on-duty operator's digital identity credential, and the planned storage start time. After the transaction is successfully recorded on the blockchain, a unique work-in-process record identifier is generated for that batch.

[0030] By binding transactions with relationships, abstract batch materials are strongly associated with concrete production resources and responsible entities, creating a production context with a clear process background. This allows subsequent process data to be accurately attributed to specific batches, equipment, and personnel, solving the problem of difficult-to-define production responsibilities in traditional management.

[0031] Furthermore, during the aging process, key operational parameters such as the alkalinity of the liquid, the temperature and humidity of the aging environment, and the cumulative aging time are periodically captured by IoT sensors. These dynamic parameters are aggregated with the work-in-process records, and signed using an asymmetric encryption algorithm to generate an immutable process witness, which is then submitted to the distributed operational ledger. Corresponding to step S3 above, the specific process is as follows: After production begins, the edge computing gateway deployed in cellar unit A-07 automatically collects data from pH and temperature / humidity sensors at preset time intervals. In this invention, the monitoring of "alkalinity of the feed solution" is characterized by the pH value collected by this pH sensor. At a certain point during the maturation process, "pH 8.9, temperature 21.5℃, humidity 78%" are collected, and the cumulative duration is obtained through a timer. The gateway combines these four parameter values ​​with the unique identifier of the work-in-process file into a JSON data block, calculates its hash value, and uses the private key bound to the hardware encryption module of the cellar unit to sign it, forming an immutable witness. Finally, it submits the data to the blockchain via an API interface.

[0032] By specifically defining the content composition and generation method of tamper-proof witnesses, the invisible production process is transformed into a series of credible, continuous, and high-frequency digital snapshots, providing an objective and irrefutable data foundation for quality control and process deviation analysis.

[0033] A production execution closed-loop adaptive control paradigm, by contextually binding production batches with physical units and driving digital agents to autonomously generate and dispatch real-time correction instructions based on periodically generated immutable witnesses.

[0034] An operational benchmark parameter library is constructed, containing the normal fluctuation range of key parameters at each process stage. A digital agent is configured to continuously monitor and parse newly generated immutable witnesses. When witness data deviates from the benchmark, the digital agent calculates the precise adjustment amount according to preset rules or algorithm models, and generates a structured instruction containing the target device address, parameter items, and target values, which is then directly sent to the corresponding environmental control actuator. The operational benchmark parameter library can be dynamically updated by a learning track to adapt to changes in external conditions.

[0035] This paradigm transforms the production process from passive monitoring to proactive, real-time adaptive adjustment, shortening the delay from problem discovery to resolution. It can effectively suppress the accumulation of small fluctuations in the production process into large quality defects, thereby significantly improving the stability of the non-standard product production process and the consistency of finished product quality.

[0036] Furthermore, the generated process witnesses are submitted to a dual-track governance paradigm for processing. In the execution track, the digital agent compares the real-time parameters in the witnesses with preset operational benchmarks. If a deviation is found, it automatically generates and dispatches immediate corrective instructions to intervene in the current production environment. In the learning track, the witness data sequence is input into the predictive operation model in real time for inference to predict process deviations. Based on the witness data and subsequent interventions and quality results, the model parameters are iteratively optimized. Corresponding to step S4 above, the specific process is as follows: Execution Track Operation: In a subsequent data snapshot, a witness report shows that the temperature value exceeds the normal range specified for the corresponding ripening stage in the operational baseline library. The execution track's digital agent is immediately triggered, automatically generating an immediate correction command containing the target device address "A-07-TempController", the parameter to be adjusted "target temperature", and the adjustment value "21℃", and dispatching it to the environmental control actuator.

[0037] By defining specific limits to the digital agent workflow, automated, precise, and real-time intervention in the production process can be achieved, transforming passive quality alarms into proactive closed-loop control and preventing the escalation of quality problems caused by delays or errors in human intervention.

[0038] Learning track operation: Upon receiving the witness data indicating excessive temperature, the predictive operation model of the learning track immediately appends it to the batch of data sequences and performs real-time forward inference calculations. At this stage, the model weights are not updated. Based on the current model weights, an updated prediction report is output in real time, quantifying the predicted probability of various key process deviations ("excessive increase in alkalinity", "persistently high temperature", "humidity falling below the threshold") occurring within the next 24 hours.

[0039] By specifying the limitations of predictive operating models, forward-looking predictions of future production risks are achieved, thereby enhancing the foresight of decision-making.

[0040] Dual-track collaborative operation: The correction instructions generated by the above execution track, the parameter sequence of temperature recovery to normal captured within a period of time after execution, and the quality inspection results of the batch that was finally rated as "superior" are packaged together into a new structured experience dataset, and fed back to the learning track as training samples for incremental training of the model.

[0041] By specifically defining the data feedback loop in the dual-track governance paradigm, a self-optimizing mechanism capable of learning from real intervention events is created. This mechanism can not only execute controls but also draw experience from the results of those controls, enabling the continuous accumulation and self-evolution of process knowledge.

[0042] An incremental evolution method for process models based on historical witness and intervention data is proposed, which performs continuous incremental training on predictive operation models by structuring the association of "process-intervention-outcome" data.

[0043] Extract the full-cycle witness data sequence of historical batches, the sequence of corrective instructions dispatched on the execution track, and the final quality inspection results from the distributed ledger; package these three types of data into a structured empirical dataset, in which intervention instructions are used as key feature variables; input this dataset as a new training sample into the predictive operations model, and update the model weights through retraining or online learning algorithms, so that it can not only predict deviations, but also evaluate the effectiveness of different interventions in specific contexts.

[0044] This method constructs a dynamic process knowledge base that can self-improve and continuously evolve, accelerates the iterative cycle of process optimization, and realizes the solidification and inheritance of effective intervention strategies.

[0045] Furthermore, based on the real-time correction instructions recorded by the execution track and the model parameters optimized by the learning track, continuous data association and on-chaining are performed on the flow events of subsequent supply chain nodes such as batch product release, grading, packaging, and logistics; the entire process data from production to sales is aggregated to form a predictable supply chain decision-making sandbox for strategy simulation, corresponding to step S5 above, the specific process is as follows: Supply Chain Linked Operations: When a batch of products leaves the storage facility, all real-time corrective action records generated during the production process and the optimized model parameters (final quality prediction labels or process stability scores) are retrieved. Operators scan the storage unit identifier and work-in-process file identifier using their mobile terminals to create a transaction for leaving the storage facility. This transaction's data structure records the actual number of products leaving the facility, the operator's digital identity, and the geographical location of the operation. It also includes a field storing the hash value of the corrective action record and the optimized model parameters, thus binding production process data to the transaction event. Subsequently, the products are transported to the cleaning station. After cleaning, a cleaning transaction is created by scanning the code again, recording the cleaning batch number and the process water quality parameters used. In the grading stage, quality inspectors sort the products according to standards and create grading transactions. These transactions record the quantity of premium, superior, and qualified products corresponding to the batch in key-value pairs and are linked to the quality inspector's digital identity. During the packaging process, workers affix a QR code containing batch information to each box of products and perform a barcode scanning and warehousing operation. This operation creates a transaction containing "Operation Type: Packaging Warehousing," "Location: Finished Product Warehouse A," and "Quantity: 500 boxes." Subsequently, the products are shipped out of the warehouse and handed over to "Shunjie Logistics" for transportation. Another transaction record is created, recording the carrier's digital identity and the real-time temperature of the temperature-controlled logistics vehicle, and a hash pointer is used to point to the previous warehousing transaction.

[0046] By specifying continuous correlation operations, the boundaries of product data management are seamlessly extended from the production domain to the entire supply chain, constructing an uninterrupted, clearly defined end-to-end data chain that provides a unified and reliable data foundation for cross-enterprise logistics tracking, inventory management, and responsibility definition.

[0047] Decision-making simulation: Finally, a supply chain manager conducted a decision-making simulation exercise, inputting the scenario of "how to adjust the production plan if the demand forecast for South China next month increases by 30%". The simulation used data aggregation functions to extract historical data, including that batch, to build a digital twin model. Through Monte Carlo simulation calculations, it provided a quantitative recommendation: "It is recommended to start two storage units in advance and pre-order an additional 20% of packaging materials".

[0048] By defining specific workflows within the supply chain decision-making sandbox, aggregated end-to-end data is transformed into forward-looking decision-making capabilities, enabling enterprises to conduct scientific risk assessments and resource planning based on real and complete operational data, thereby improving the scientific rigor and accuracy of decision-making.

[0049] A supply chain risk simulation sandbox method integrating high-fidelity production data is proposed. By constructing a digital twin model based on real process data, the method can quantify and simulate the impact of external disturbances on the supply chain.

[0050] First, the system extracts and integrates data from the distributed ledger, covering the entire supply chain from raw materials to logistics, using data aggregation functions. Second, it constructs a parameterized digital twin model of the supply chain, where the model parameters for the production process (production cycle, probability distribution of yield rate) are directly derived from historical data through statistical analysis or fitting, rather than being statically set. Finally, it receives simulated scenarios from external input (rising raw material prices, transportation disruptions) and uses calculation methods such as Monte Carlo simulation to quantify the impact of these scenarios on key indicators such as total cost and delivery cycle.

[0051] By introducing real data from the bottom of the production process, the accuracy of supply chain simulation and decision-making is greatly improved, enabling managers to conduct stress tests on various complex risk scenarios, thereby developing more forward-looking and robust strategies and enhancing the resilience of the entire supply chain.

[0052] This embodiment transforms a traditional preserved egg production process into an intelligent production flow that is data-transparent, process-controllable in real time, and continuously optimized. The method of this invention not only solves the black-box problem of traditional traceability through penetrating process data collection, but also constructs a proactive quality control and process evolution closed loop through a dual-track governance paradigm, ultimately transforming technical data into decision-making capabilities. This addresses the core problems of blind spots in quality control and insufficient decision-making data support in the production of non-standard products in the background technology.

[0053] Example 2: This embodiment provides a blockchain-based traceability system for the entire production process of preserved eggs. The system consists of hardware devices and software modules deployed in the production and management environment. See details... Figure 2 .

[0054] The system's hardware includes: desktop terminals with barcode scanners configured in the raw material warehouse; mobile industrial tablets provided to workshop supervisors and operators in the production workshop; industrial-grade pH sensors and high-precision temperature and humidity sensors deployed in each storage unit, which are connected to an edge computing gateway responsible for data acquisition, local computing, and communication via a Modbus-RTU industrial bus; the gateway has a built-in hardware security module compliant with the TPM 2.0 standard for securely generating and storing the encrypted private key for the storage unit; and a central server for carrying core business logic.

[0055] The system's software modules are deployed on the aforementioned hardware, specifically including: Registration record module: It runs as a software application on the warehouse desktop terminal, providing a graphical user interface for warehouse administrators to input initial material file information and call the blockchain client SDK to encapsulate the initial operation record into a transaction and submit it to the blockchain.

[0056] Production factor association module: Deployed as a mobile application on the workshop director's industrial tablet. The application can use the camera to call the scanning function to read the identification of batches and storage units, and provide an interface to select process formulas and confirm operator identities, and finally generate and submit relationship binding transactions.

[0057] Witness generation module: This module is embedded in the edge computing gateway's software. It communicates with sensors via an industrial bus protocol, performs periodic data acquisition, JSON formatting, hash calculation, and digital signature operations, and sends the generated immutable witnesses to the central server's blockchain node via a secure network protocol.

[0058] The dual-track governance module is deployed on a central server. The execution track unit is a continuously running background service that monitors new witnesses on the blockchain and compares them with an SQL database (serving as an operational baseline parameter library). It generates correction instructions based on preset logic and sends them to the environmental control actuator via the MQTT IoT protocol. The learning track unit is an application integrating the machine learning framework TensorFlow. This application uses a Keras frontend to build a Long Short-Term Memory (LSTM) network model with an attention mechanism, responsible for pulling samples from historical blockchain data and performing offline training and online inference of the model. In this embodiment, the specific implementation architecture of the LSTM network model with an attention mechanism is as follows: Input layer: Receives a fixed-length batch of time series data (the past 24 time steps), with the data dimension being (number of batches, 24, 4), where the four features correspond to the pH value of the liquid, the storage temperature value, the storage humidity value, and the cumulative maturation time.

[0059] LSTM layer: Contains two stacked LSTM units, used to extract long-term dependency features from time series.

[0060] Attention layer: A temporal attention mechanism is applied on top of the output of the LSTM layer. This mechanism calculates the weight score of the hidden state at each time step, enabling the model to dynamically focus on historical moments that have a greater influence on future bias predictions.

[0061] Output layer: The weighted average feature vector is input into the fully connected layer and finally output as a multivariate prediction vector through the Sigmoid activation function. This vector contains the probability predictions of "alkalinity deviation", "temperature deviation" and "humidity deviation" occurring in the next 24 hours (in steps of 4 hours).

[0062] Product Data Management Module: Deployed on a central server, this module provides managers with a web-based data analysis and visualization platform. The platform parses and displays end-to-end data from the blockchain and includes a built-in simulation engine for a supply chain decision-making sandbox. This engine utilizes Python's NumPy and SimPy libraries to perform Monte Carlo simulations, allowing managers to input simulation scenarios and obtain quantitative analysis results through the web interface.

[0063] This system, through the collaborative work of its various modules, and based on high-fidelity process witnessing and a dual-track governance paradigm, constructs a robust production system capable of proactively suppressing production disturbances and achieving closed-loop adaptive control. Furthermore, through the continuous evolutionary capability of the learning track, it establishes a dynamic process knowledge base capable of self-optimization. Ultimately, the system seamlessly links the underlying refined production data with upper-level supply chain events, transforming technical data into precise insights through a decision-making sandbox, thereby systematically improving the quality control accuracy, risk resilience, and management decision-making level of the non-standard product supply chain.

[0064] Example 3: This embodiment provides a further detailed explanation of the dual-track governance paradigm. When an immutable witness from cellar A-07 is submitted to the distributed ledger, both cells of the dual-track governance module are triggered simultaneously.

[0065] Within the execution track, the workflow begins with parsing the witnessed data to extract real-time parameters (pH 8.9, temperature 21.5℃) and the corresponding work-in-process record identifier. Next, this identifier is used to query the operational baseline parameter library to obtain the parameter thresholds for the current production stage. Then, a comparison operation is performed; if a parameter deviation is detected, a structured, immediate corrective instruction is generated according to preset rules. This instruction clearly identifies the target equipment, the parameter to be adjusted, and the target value. Finally, this instruction is sent via the MQTT IoT protocol to the corresponding environmental control actuator (temperature relay or ventilation system controller) to complete the closed-loop intervention.

[0066] The learning track's workflow is divided into two independent stages: 1. Real-time inference: When a witness data point is received, it is appended to the corresponding time-series dataset and immediately fed into the deployed LSTM model for inference. The model's output layer is designed as a multivariate prediction vector, capable of simultaneously outputting the probability predictions of three "critical process deviations"—"alkalinity deviation," "temperature deviation," and "humidity deviation"—occurring within the next 24 hours (in 4-hour increments). 2. Incremental training: The data feedback loop is activated after batch production. During production, the learning track receives witness data in real time, along with correction instructions dispatched by the execution track and parameter sequences indicating process recovery after instruction execution; and after the batch production is completed, the final quality inspection results (excellent / good rate label) are obtained. At this point, all this data is automatically packaged into a structured empirical dataset, used as new training samples, and input into the predictive operation model at once for incremental training and iterative optimization, thereby achieving continuous self-optimization of the model.

[0067] Example 4: This embodiment provides a more detailed description of the process for generating tamper-proof witnesses. This process is executed autonomously on the edge computing gateway of each storage unit. The process begins with a timer task triggered, for example, every 30 minutes, where the gateway reads raw data from sensors via an industrial bus protocol. Next, the gateway uses an internal timer to obtain the cumulative duration since the start of storage and encapsulates the four parameters (pH, temperature, humidity, and cumulative duration) along with a unique identifier from the work-in-process record into a structured JSON object. The specific structure of this JSON object is: {"batch_id": "ZZP-A07-001", "timestamp": 1678886400, "ph": 8.9, "temp": 21.5, "humidity": 78, "duration_hours": 120.5}. The gateway performs a SHA-256 hash on this JSON object to generate a unique digital fingerprint. It then invokes its built-in hardware security module to digitally sign the hash digest using the device's private key stored within it, ensuring the data's source is trustworthy and its integrity is protected. Finally, the original JSON data block and the generated digital signature are combined to form the "data payload" of a transaction and broadcast to the blockchain network. Furthermore, this edge computing gateway is configured with a network outage response mechanism. When the gateway loses connection to the blockchain node network of the central server, it initiates a local caching mode, continuing to generate and sign witnesses at a preset frequency, and encrypting and storing them in a local non-volatile storage queue. Once the network connection is restored, the gateway immediately pushes all witness data packets in the cache queue to the distributed operational ledger in chronological order, ensuring that production process data remains complete, continuous, and chronologically correct even under network fluctuations.

[0068] Example 5: This embodiment provides a more detailed explanation of the simulation process of the supply chain decision-making simulation. For specific details, please refer to [link / reference]. Figure 3 This sand table offers simulation capabilities for a variety of scenarios: Scenario 1: Cost Fluctuation Impact Assessment Taking a supply chain manager's desire to assess the impact of a 15% increase in the supply price of a major raw material supplier on the company's annual profit as an example, the deduction process is as follows: The deduction begins with the manager initiating a new simulation on the decision-making sandbox interface. The sandbox's data aggregation function is then triggered, automatically retrieving full-process data for all batches over the past year from the distributed ledger. Specifically, the function links the "purchase order number" in the initial work record to the raw material procurement cost in the financial system, links the "tiered" transaction to the yield rate data, uses "immutable witness" to calculate the actual maturation time (capital occupation time), and uses logistics "flow transactions" to calculate transportation time and cost. Based on this real historical data, the system constructs a supply chain digital twin model containing multiple random parameters. The parameters of its production process ("maturation cycle" is fitted with a normal distribution with a mean of 480 hours and a standard deviation of 20 hours, and "yield rate" is fitted with a Beta distribution) are directly derived from real process witnesses, rather than static settings. Subsequently, the manager inputs scenario parameters through the command-line interface to set the raw material cost increase coefficient (+15%). The simulation engine then starts, executing tens of thousands of Monte Carlo simulations. In each run, the engine samples from various probability distributions of the digital twin model and substitutes them into cost and profit calculation formulas that apply scenario parameters. After all simulations are completed, the simulation platform performs statistical analysis on tens of thousands of profit results and ultimately presents the expected impact and risk range of the scenario on annual profits to managers in the form of probability distribution charts and key indicators, providing data support for subsequent decisions (whether to raise prices or change suppliers).

[0069] Scenario 2: Logistics Disruption Stress Test. A supply chain manager wants to assess the impact of a 48-hour transportation disruption caused by weather on on-time delivery rate and inventory costs. 1. Scenario Input: The manager selects the "Logistics Disruption" simulation scenario in the simulation interface and inputs the parameters "Disruption Node: South China Logistics Center" and "Disruption Duration: 48 hours." 2. Data Aggregation and Model Call: The simulation engine uses aggregation functions to retrieve historical data, focusing on analyzing the timestamps of "outbound" and "transportation" transactions to build a baseline model of logistics turnaround time. 3. Simulation Calculation: The simulation engine forcibly adds a 48-hour delay variable to the logistics model. Subsequently, the engine simulates the order flow for the next quarter and extrapolates the "domino effect" of this delay on the delivery cycles of all subsequent batches, while calculating the increase in finished goods inventory in transit (work-in-process cost) and warehouse backlog caused by delayed delivery. 4. Result Output: The simulation ultimately outputs a stress test report: quantitatively predicting that the "on-time delivery rate" will decrease from 98% to 75%, and the "average inventory cost" will increase by 22%. Based on these results, the report automatically recommends response strategies, such as "immediately activate backup logistics provider B" or "temporarily increase regional safety stock by 30%." This simulation enables managers to develop contingency plans based on data, enhancing the resilience of the supply chain.

[0070] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for tracing the whole process of production of preserved eggs based on a blockchain, characterized in that, include: Register a unique production instruction for each batch of raw egg, generate an initial job record containing the initial material file, and record it on the distributed operations ledger; The initial work records are associated with the storage unit and process formula as production factors. By strategically anchoring the execution resources, work-in-process operation files associated with the production factors are generated. The work-in-process records and captured operational parameters are aggregated through an edge computing gateway; an immutable witness is generated using the device's private key and an asymmetric encryption algorithm. The operating parameters include the alkalinity of the liquid, the temperature and humidity of the storage cellar, and the maturation time. The immutable witness specifically includes: periodically performing data snapshot operations based on a preset time frequency to continuously add dynamic process data; the data snapshots are collected by IoT sensors at the current time point, including the pH value of the liquid material, the temperature value of the storage environment, and the humidity value of the storage environment, and the cumulative duration from the start of storage is obtained by a timer; the four parameter values ​​collected are combined with the unique identifier of the work-in-process file to form a data block in JSON format; the hash value of the data block is calculated, and the hash value is digitally signed using the private key bound to the storage unit to form the immutable witness; The immutable witness is submitted to a dual-track governance paradigm for processing. This paradigm includes an execution track and a learning track. The execution track generates and issues an immediate correction instruction for real-time intervention in the current batch based on the comparison between the witness and the current operational baseline. The operational baseline defines the normal fluctuation range of each operational parameter at different maturation stages and is dynamically updated by the learning track. The learning track inputs the witness data into a predictive operation model, performs real-time inference to predict process deviations, and iteratively optimizes the model parameters based on the witness data and subsequent intervention and quality results. Specifically, the digital agent receives the immutable witness and parses the alkalinity of the liquid and the temperature and humidity of the storage cellar. It then compares the parsed parameter values ​​with multidimensional thresholds stored in the operational baseline parameter library, where the multidimensional thresholds define the normal fluctuation range of each parameter at different maturation stages. When any parameter value is detected to exceed the normal fluctuation range of its corresponding stage, a structured immediate correction instruction is immediately generated. The data structure of the immediate correction instruction includes the network address of the target environmental control device, the parameter to be adjusted, and the target value to be adjusted. Based on real-time correction instructions and optimized model parameters, the flow events of subsequent supply chain nodes are correlated to construct product data management archives, and finally aggregated to form a predictable supply chain decision sandbox. The production process model parameters of the supply chain decision sandbox are obtained by statistical fitting of historical batch witness data. 2.The method according to claim 1, wherein, The initial material file specifically includes: By recording initial transactions on a distributed operational ledger, a root data object is created as the starting point for product data management. The data structure of the root data object includes: raw egg purchase order number associated with the procurement process, enterprise digital certificate identifying the supplier, geographical location data of the raw material origin, purchase quantity and unit of measurement of the batch size, and hash value of electronic quarantine certificate issued by a third-party quarantine agency. The hash value is used for subsequent certificate file verification. 3.The method according to claim 1, characterized in that, The aforementioned production factor linkages specifically include: A relationship binding transaction is created on the distributed operations ledger to inject production context information into the batch's product data management record. The input field of the relationship binding transaction references the hash value of the transaction in which the initial job record is located, while the data payload field records the physical asset number of the assigned storage unit, the version identifier of the process formula used, the employee digital identity credential of the production operator on duty, and the planned storage start time. The output field of the transaction generates a unique identifier for the work-in-process operation file that is referenced in subsequent production stages.

4. The blockchain-based method for tracing the entire production process of preserved eggs according to claim 1, characterized in that, The predictive operation model specifically includes: The distributed operations ledger extracts full-cycle historical witness data of completed production batches to form a dataset containing multiple time series. The finished product yield rate recorded by the quality inspection department for each batch is used as a label for supervised learning and associated with the corresponding time series dataset. The labeled dataset is used to train a long short-term memory network with an attention mechanism. The trained model is deployed to receive the witness data sequence generated for the current production batch and output the probability prediction value of process deviations in the next 24 hours in real time.

5. The blockchain-based method for tracing the entire production process of preserved eggs according to claim 1, characterized in that, The dual-track governance paradigm specifically includes: A data feedback loop connecting the execution track and the learning track is configured to: continuously capture the sequence of operational parameter changes over a period of time after the execution track issues an immediate correction instruction; package the issued instruction content, the sequence of parameter changes after execution, and the final quality inspection results of the batch into a structured experience dataset; and input the experience dataset as a new training sample into the predictive operation model of the learning track to perform incremental training of the model.

6. The blockchain-based method for tracing the entire production process of preserved eggs according to claim 1, characterized in that, The association of subsequent supply chain node transfer events specifically includes: After a batch of products completes its production process, an on-chain operation is performed at each point of responsibility handover during subsequent processes such as unloading from the cellar, cleaning, grading, packaging, warehousing, unloading, and logistics transportation. The on-chain operation creates a new transaction, which records the event type, the digital identity of the operator, the geographical location of the operation, the quantity of products handed over, and the real-time temperature of the temperature-controlled logistics vehicle in the form of key-value pairs. The new transaction contains a hash pointer to its preceding transaction, thus constructing a logistics and ownership change chain on the ledger.

7. The blockchain-based method for tracing the entire production process of preserved eggs according to claim 1, characterized in that, The supply chain decision-making sandbox specifically includes: The simulation process includes the following steps: calling a data aggregation function to extract historical batch cost and time data from the distributed operations ledger for production, logistics, and warehousing, and constructing a parameterized supply chain digital twin model based on this data; receiving externally input simulation scenarios through a command-line interface, where each simulation scenario is a set of adjustments to multiple parameters in the model; and running Monte Carlo simulation calculations in the digital twin model to quantitatively output the probability distribution and expected value of the simulation scenario for the total cost, delivery cycle, and expected profit margin of the batch products.

8. A blockchain-based traceability system for the entire production process of preserved eggs, characterized in that, include: Registration record module: Used to register a unique production instruction for each batch of raw egg, generate an initial operation record containing the initial material file, and record it on the distributed operations ledger; Production factor association module: used to associate the initial operation record with the storage unit, process formula and operator identity as production factors, and generate work-in-process operation files associated with production factors through strategic anchoring of execution resources; Witness generation module: Aggregates the work-in-process operation file with the captured material liquid alkalinity, cellar temperature and humidity, and maturation time through an edge computing gateway; An immutable witness is generated using the device's private key and an asymmetric encryption algorithm; The immutable witness specifically includes: periodically performing data snapshot operations based on a preset time frequency to continuously add dynamic process data; the data snapshots are collected by IoT sensors at the current time point, including the pH value of the liquid material, the temperature value of the storage environment, and the humidity value of the storage environment, and the cumulative duration from the start of storage is obtained by a timer; the four parameter values ​​collected are combined with the unique identifier of the work-in-process file to form a data block in JSON format; the hash value of the data block is calculated, and the hash value is digitally signed using the private key bound to the storage unit to form the immutable witness; Dual-track governance module: Execution track: The digital agent generates and dispatches an immediate correction instruction for real-time intervention in the current batch based on the comparison results between the witnessed data and the current operating baseline; the operating baseline defines the normal fluctuation range of each operating parameter at different maturation stages and is dynamically updated by the learning track; Learning track: Inputs the witnessed data into the predictive operation model, performs real-time inference to predict process deviations, and iteratively optimizes the model parameters based on the witnessed data and subsequent intervention and quality results; The digital agent specifically includes: receiving the tamper-proof witnessed data and parsing the alkalinity of the liquid and the temperature and humidity of the storage cellar; comparing the parsed parameter values ​​with the multidimensional thresholds stored in the operating baseline parameter library one by one, the multidimensional thresholds defining the normal fluctuation range of each parameter at different maturation stages; when any parameter value is detected to exceed the normal fluctuation range of its corresponding stage, a structured immediate correction instruction is immediately generated, the data structure of which includes the network address of the target environmental control device, the parameter item to be adjusted, and the target value to be adjusted; Product data management module: Based on real-time correction instructions and optimized model parameters, it associates the flow events of subsequent supply chain nodes, constructs product data management archives, and forms a predictable supply chain decision sandbox through data aggregation. The production process model parameters of the supply chain decision sandbox are derived from the statistical fitting of historical batch witness data, and together they complete the product data management of the batch.

Images