A method, apparatus, and electronic device for migrating software data during factory production transition.

By automating the identification of production transition scenarios and controlling branch processes, the problem of software data migration relying on manual operation and data consistency during factory production transition was solved. This enabled closed-loop management of the entire process, ensuring the efficiency of data migration and business continuity, and avoiding production line downtime.

CN123115100APending Publication Date: 2026-11-17CHERY AUTOMOBILE CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202611635644.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-22
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In existing technologies, software data migration during factory production transitions relies on manual operation, making it difficult to guarantee data consistency, unable to flexibly cope with various production transition scenarios, and lacking full-process closed-loop management, leading to the risk of production line downtime.

Method used

The system introduces a scenario-based production transition and branch process control model, and adopts an automated data migration method, including full initialization of new factories and incremental migration processes of existing factories. It generates standardized JSON files through an automatic mapping mechanism from engineering view to factory view, controls the pause and resumption of data retrieval tasks, and ensures the correct allocation and synchronization of data version status.

Benefits of technology

It has implemented data processing strategies adapted to different production transition scenarios, ensuring improved data distribution efficiency, avoiding human error, guaranteeing business continuity, avoiding the risk of production line downtime, and achieving systematization, automation, and traceability of the entire process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123115100A_ABST
    Figure CN123115100A_ABST
Patent Text Reader

Abstract

This invention discloses a method, apparatus, and system for migrating software data during factory production conversion, relating to the fields of automotive manufacturing and industrial information technology. The method includes: receiving conversion application information, the application information including the vehicle model to be converted and a target factory identifier; determining whether the target factory is a new factory based on the target factory identifier; if it is a new factory, pausing the data retrieval task of the original factory, completely migrating the software data of the original factory to the target factory, and activating the data retrieval channel of the target factory after data consistency verification passes; if it is an existing factory, executing the corresponding data distribution logic according to the type of conversion material, mapping engineering view data to factory view data, and then distributing it to the target factory. This invention achieves automated, scenario-based, and closed-loop migration of software data during factory production conversion, improving conversion efficiency and data reliability.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of automobile manufacturing and industrial information technology, specifically to a method, apparatus and system for migrating software data during factory production transitions, applicable to software data migration and synchronization scenarios when vehicle manufacturers switch between multiple factories for different vehicle models. Background Technology

[0002] In the automotive manufacturing industry, with capacity adjustments and the introduction of new models, factory conversion (i.e., moving the production of a particular model from one factory to another) is a common strategic behavior for companies to respond to market changes and optimize capacity layout. Especially in the context of the deep evolution of software-defined vehicles, factory conversion is no longer just a physical relocation of production line equipment, but a cross-factory systemic reconstruction with the whole vehicle software data as the core carrier. Software data used for electrical testing and production line flashing, including controller assembly part numbers, software package numbers, hardware and software version binding relationships, configuration words, etc., have become key access conditions that determine whether a vehicle can be legally rolled off the production line and whether its functions are fully activated.

[0003] For example, the published document with authorization announcement number CN116976653B, entitled "A Multi-Factory Capacity Coordination Method Based on MES System," discloses a multi-factory capacity coordination scheme: the MES system of factory A sends a part transfer request to the MES system of factory B. After verification by factory B, factory A freezes the information of the parts to be processed and stores it in a dataset. Factory B then reads the dataset and arranges processing, while simultaneously calling the overhead crane system to complete the physical transport of the parts. This scheme realizes the allocation of production tasks between different factories, allowing parts to be transferred to another factory for processing when the capacity of one factory is insufficient.

[0004] However, the above solutions are geared towards coordinating the production capacity of physical parts, and their data interaction revolves around batch transfers of parts. They do not involve the systematic migration of software data (such as controller software packages, configuration words, version bindings, etc.) required for flashing on automotive production lines. In actual factory conversion processes, software data migration faces far more complex technical challenges:

[0005] On the one hand, the migration of software data during the conversion of existing factories is highly dependent on manual operation. Process personnel need to manually export data, organize mapping relationships offline, and then import it into the target factory's manufacturing execution system. The whole process lacks a unified standard process and automated interface, making it difficult to guarantee data integrity and accuracy. It is easy to cause production line stoppage due to missing configuration items or version relationships.

[0006] On the other hand, different production conversion scenarios (such as the first production of a brand-new factory, the addition of materials to an existing factory, the conversion of all materials, and the parallel production of new and old factories) require different data processing logics. Existing solutions cannot automatically select differentiated data migration strategies according to the production conversion scenario, and lack flexibility. At the same time, if the opening and closing of the data release channel is not properly handled during the production conversion, it is easy for the original factory to not receive new data or for the new factory to receive the wrong version of garbage data, making it difficult to guarantee business continuity.

[0007] Therefore, there is an urgent need for a systematic, automated, and traceable software data migration solution for factory transitions to solve the technical problems of existing technologies, such as reliance on manual operation, poor data consistency, weak scenario adaptability, and lack of closed-loop management. Summary of the Invention

[0008] The purpose of this invention is to provide a method, apparatus and system for migrating software data during factory production transition, so as to solve the technical problems in the prior art of relying on manual labor for software data migration during factory production transition, difficulty in ensuring data consistency, difficulty in coping with multiple production transition scenarios and lack of closed-loop management of the entire process.

[0009] To achieve the above objectives, the technical solution adopted by the present invention is a method for migrating software data during factory production transition, comprising:

[0010] Receive production conversion application information, which includes the vehicle model to be converted and the target factory identifier;

[0011] Determine whether the target factory is a brand new factory based on the target factory identifier;

[0012] If it is a brand new factory, the data retrieval task of the original factory is suspended, the software data of the original factory is completely migrated to the target factory, and the data retrieval channel of the target factory is enabled after the data consistency verification is passed.

[0013] If it is an existing factory, the corresponding data distribution logic is executed according to the type of transferred material, and the engineering view data is mapped to the factory view data and then sent to the target factory.

[0014] The suspension of the original factory data retrieval task includes:

[0015] Send a pull pause command to the manufacturing execution system of the original factory, the pull pause command carrying the conversion application number and vehicle model identifier;

[0016] The newly released software data received by the original factory during the suspension period is temporarily stored in the cache queue of the upstream platform. Once the data retrieval channel of the target factory is enabled, all data in the cache queue will be resent from the target factory.

[0017] The complete migration of software data from the original factory to the target factory includes:

[0018] By copying the underlying database tables, the software data tables in the original factory's manufacturing execution system are fully copied to the target factory's manufacturing execution system. The software data tables include a table showing the relationship between the engineering vehicle number and the material number, a table showing the relationship between the material number and the controller part number and the software package number, a software details table, and a software configuration word data table.

[0019] The step of executing the corresponding data distribution logic based on the type of transferred material includes:

[0020] If the target factory is an existing factory and the transferred material is a new material, then data distribution will be automatically triggered according to the release process number, and the engineering view data corresponding to the new material will be automatically added to the factory view of the target factory and distributed.

[0021] If the transferred material is part or all of the material, the specified software data is exported from the manufacturing execution system of the original plant, imported into the upstream platform, and then the upstream platform matches the engineering view data according to the imported list, generates the plant view data, and distributes it.

[0022] The process of mapping engineering view data to factory view data and then distributing it includes:

[0023] The upstream platform, based on the factory identifier of the target factory, filters software data corresponding to the model being converted for production from the engineering view, and generates six types of JSON data files according to a predefined interface protocol. These six types of JSON data files include:

[0024] Data files containing the relationship between the engineering vehicle number and the vehicle part number, the relationship between the engineering vehicle number and the part number, the relationship between the part number and the software package number and the factory, the software details data file, the relationship between the engineering vehicle number and the configuration word short code, and the complete configuration word data file;

[0025] The six types of JSON data files are pushed to the manufacturing execution system of the target factory via a scheduled task.

[0026] Factory conversion software data migration methods also include:

[0027] When there are multiple versions of the same part number or the same software package number, the data of each version shall be distributed in the order of release time.

[0028] The target factory's manufacturing execution system automatically assigns status identifiers to each version of data according to the order of receipt: the first received version is marked as release status, the second received version is marked as trial assembly status, and the third and subsequent received versions are marked as pending release status.

[0029] After enabling the data retrieval channel for the target factory, the following is also included:

[0030] Determine whether the original factory should continue producing the model to be converted;

[0031] If the original factory no longer produces the model to be converted, then the factory group status of the original factory will be set to invalid, or the overall status of the original factory will be set to shutdown.

[0032] A software data migration device for factory transition, comprising:

[0033] The application receiving module is used to receive production conversion application information, which includes the vehicle model to be converted and the target factory identifier;

[0034] The scene discrimination module is used to determine whether the target factory is a brand new factory based on the target factory identifier;

[0035] The data processing module is used to pause the data retrieval task of the original factory and completely migrate the software data of the original factory to the target factory when the target factory is a brand new factory; when the target factory is an existing factory, it executes the corresponding data distribution logic according to the type of transferred material, maps the engineering view data to the factory view data and then distributes it.

[0036] The verification module is used to perform data consistency verification after the data migration is completed.

[0037] The channel control module is used to enable the data retrieval channel of the target factory after the data consistency verification is passed.

[0038] The version sorting module is used to sort the version data in chronological order of release time when there are multiple versions of the same part number or the same software package number.

[0039] The status allocation module is used to automatically assign release status, trial installation status, or pending release status to each version of data according to the order of receipt.

[0040] The view mapping module is used to filter software data corresponding to the production model from the engineering view based on the factory identifier of the target factory, and generate six types of JSON data files according to a predefined interface protocol.

[0041] The channel pause module is used to send a pull pause command to the manufacturing execution system of the original factory, and temporarily store newly released software data received during the pause in a cache queue;

[0042] The data resending module is used to resend all data in the cache queue from the target factory after the data retrieval channel of the target factory is enabled.

[0043] The status cleanup module is used to set the factory group status of the original factory to invalid or the overall status of the original factory to shutdown when the original factory no longer produces the model to be converted.

[0044] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the factory transition software data migration method.

[0045] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the factory transition software data migration method.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] First, the "conversion scenario identification + branch process control" mode is introduced. The system can automatically identify whether the target factory is a brand new factory and select the full data initialization process or the incremental data distribution process accordingly. This enables different conversion scenarios, such as the first production of a brand new factory and the replenishment of materials in an existing factory, to obtain appropriate data processing strategies, which solves the problem that existing technologies cannot flexibly cope with multiple conversion scenarios.

[0048] Second, a five-stage closed-loop process of "application-migration-verification-effectiveness-cleanup" has been established. From the initiation of the conversion application to the cleanup of the original factory status, the entire process is completed under systematic process control. The operators and operation time of each node can be traced, avoiding the arbitrariness and uncontrollability brought about by manual operation.

[0049] Third, through the automatic mapping mechanism from engineering view to factory view, the system automatically extracts software data from the engineering view at the source of R&D, generates standardized JSON files according to the interface requirements of the target factory and distributes them, replacing the work of manually sorting Excel and manually entering data, which greatly improves the efficiency of data distribution and reduces data omissions and version errors caused by human error.

[0050] Fourth, by adopting a method of distributing multiple versions of data in chronological order of release and automatically assigning status at the factory, the target factory is guaranteed to receive the correct data version status—the latest version is in the "released" status for mass production, the next newest version is in the "trial installation" status for trial production verification, and the earlier version is in the "pending release" status for future use. This can avoid production quality problems caused by confusion between new and old versions.

[0051] Fifth, by controlling the pause and resumption of data retrieval tasks, as well as the temporary storage and re-issuance of data during the production transition, a seamless switch was achieved where the original factory did not stop production and the new factory was put into operation on time during the production transition, ensuring business continuity and avoiding the risk of production line stoppage due to data channel interruption. Attached Figure Description

[0052] The following is a brief explanation of the content represented by each figure in this specification:

[0053] Figure 1 This is an overall flowchart of the factory conversion software data migration method of the present invention;

[0054] Figure 2 This is a schematic diagram of the new factory conversion branch process of the present invention;

[0055] Figure 3 This is a schematic diagram of the existing factory conversion process of the present invention;

[0056] Figure 4 This is a schematic diagram illustrating the structure of the six types of JSON interface data files of this invention;

[0057] Figure 5 This is a schematic diagram of the module structure of the factory conversion software data migration device of the present invention;

[0058] Figure 6 This is a schematic diagram of the architecture of the factory conversion software data migration system of the present invention. Detailed Implementation

[0059] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and through the description of the examples.

[0060] Although the invention has been shown and described herein with reference to specific embodiments, it is not intended to be limited to the details shown. Rather, various modifications in detail may be made within the equivalent scope and scope of the claims without departing from the invention. In the drawings, the same item numbers refer to the same elements.

[0061] Throughout this disclosure, various terms are used to describe the physical shape or arrangement of features. Many of these terms are used to describe features conforming to a cylindrical or generally cylindrical geometry with the feature as its radius and a central axis perpendicular to that radius. Unless otherwise specified, the terms are given the following meanings: The terms “longitudinal,” “longitudinal,” “axial,” and “axial” refer to a direction, dimension, or orientation parallel to the central axis. The terms “radial” and “radially” refer to a direction, dimension, or orientation perpendicular to the central axis. The terms “inward” and “inner” refer to a direction, dimension, or orientation extending radially toward the central axis. The terms “outward” and “outer” refer to a direction, dimension, or orientation extending radially away from the central axis.

[0062] In this specification, relative terms such as “horizontal,” “vertical,” “upward,” “downward,” “top,” and “bottom,” and their derivatives (e.g., “horizontal,” “downward,” “upward,” etc.) should be interpreted as referring to the direction described or the direction shown in the accompanying drawings. These relative terms are for ease of description and are not generally intended to require a specific direction.

[0063] This invention is a software data migration method for factory production transition, applied to multi-factory production switchover scenarios in vehicle manufacturing enterprises. It is based on a three-tiered software data management system: "CGPDM R&D source release → CGVSP global vehicle software data platform → MES factory execution." The CGPDM system is the R&D-side product data management system, responsible for defining and releasing vehicle software configurations; the CGVSP system is the global vehicle software data platform, responsible for the conversion from engineering views to factory views, data distribution, and version control; the MES system is the factory manufacturing execution system, responsible for receiving, storing, and updating software data on the production line.

[0064] like Figure 1 As shown, the overall process of this method, from production conversion initiation to final shutdown, comprises 25 nodes, covering six major stages: project planning, scenario identification, branch execution, data verification, channel switching, and state cleanup. The process adopts a dual-branch structure: depending on whether the target factory is a completely new factory, it executes either the "New Factory Full Initialization Process" or the "Existing Factory Incremental Migration Process." The existing factory branch is further divided into two sub-scenarios: "All Materials Converted" and "New Materials Supplemented."

[0065] The following is combined Figure 1 The content details the specific implementation steps of the software data migration method for factory conversion, as follows:

[0066] The first stage is the startup and scene identification phase (steps 1 to 4).

[0067] Step S1: Process begins;

[0068] This step marks the beginning of the process. Process engineers trigger the production transition entry point in the CGVSP system. The system initializes the production transition transaction context and assigns a unique production transition transaction number for full-process traceability. The transaction context records metadata such as the production model, initiating employee ID, initiation time, and transaction status. A process log is also enabled to record the operator, time, content, and result of each subsequent operation.

[0069] Step S2: The project team plans the production transition model;

[0070] This step is completed manually by the project team to determine the production conversion plan and timeline, serving as input for subsequent automation processes and determining the objectives the equipment needs to perform. For example, the project team may organize a cross-departmental coordination meeting involving manufacturing, process, IT, and ABOM engineers to assess the feasibility of the production conversion and confirm the plan. The meeting will clarify the following key inputs:

[0071] The models to be converted to other production lines and the corresponding range of vehicle part numbers;

[0072] Source factory code and target factory code;

[0073] Target factory attributes (new factory / existing factory);

[0074] Types of production conversion (conversion of all materials / conversion of some materials / addition of new materials);

[0075] The plan includes switching timelines and parallel production cycles;

[0076] Time window and risk contingency plan for data cutover.

[0077] A formal conversion plan will be formulated to serve as the basis for subsequent steps.

[0078] Step S3: The process receives the production switch notification;

[0079] According to the production conversion plan, the process manager is notified of the production conversion requirements via formal email, specifying the vehicle model to be converted, the target factory, the timeline, and quality requirements. Based on this, the process manager initiates the software data migration for the production conversion, registering the conversion task in the CGVSP system and assigning a responsible person.

[0080] Step S4: Determine the target factory attributes;

[0081] Based on the input of the production conversion plan, process engineers confirm in the system whether the target factory is a brand new factory. The criteria for judgment are: whether the target factory already has factory master data in the CGVSP system, and whether any model is already in production.

[0082] If the target factory is a brand new factory (without factory master data or without any models in production), proceed to branch one: new factory data migration process, corresponding to subsequent steps S5 to S14.

[0083] If the target factory is an existing factory (the factory master data already exists and other models are in production), proceed to branch two: Existing factory data migration process, corresponding to subsequent steps S15 to S24.

[0084] Branch 1: Full Initialization Process of the New Factory (Steps 5 to 14)

[0085] This branch is suitable for scenarios where the target factory is a brand new factory that is going into production for the first time. Its features include full data migration, one-time initialization, and no historical data conflicts.

[0086] Step S5: The process submits a factory conversion application at CGVSP;

[0087] Process engineers create a factory conversion application form in the conversion management module of the CGVSP system, filling in the following key information:

[0088] Model codes and names for vehicles converted to other production models;

[0089] Target factory code, factory name, and factory location;

[0090] Source factory code (data source factory);

[0091] The effective date of the planned conversion;

[0092] Scope of production conversion (all materials / specific engineering vehicle number range).

[0093] After completing the form and submitting the application, the process will automatically be routed to the IT engineer.

[0094] Step S6: Maintain the master data of the factory switching production in the CGVSP system;

[0095] After receiving the process notification, the IT engineer creates new factory master data in the factory management module of the CGVSP system based on the application form information:

[0096] Enter the factory code, factory name, factory abbreviation, and business unit to which it belongs;

[0097] Set the "Factory Status" to "In Production";

[0098] Configure the factory's time zone, language, and data storage partitions;

[0099] Bind the MES system interface address, communication protocol and authentication key corresponding to the factory.

[0100] After the factory is built, click "Complete" in the process, and the process will jump to the next node. At the same time, the system will automatically record the operation log.

[0101] Step S7: Close the original factory MES data retrieval task;

[0102] After receiving the workflow notification, the MES system engineer logs into the source factory's MES system and closes the upstream software data retrieval task corresponding to that vehicle model. This step can pause the source factory's receipt of new release data from CGVSP during the production transition data cutover, preventing inconsistencies in the migration baseline caused by continued changes in the source data during the transition process.

[0103] During the pause in data retrieval, all newly released software data from the R&D side is retained in the cache queue of the CGVSP system and is not distributed to any factory to ensure that the migration baseline is frozen.

[0104] Step S8: Migrate all data from the original factory MES system to the new factory MES system;

[0105] MES system engineers migrated all software data for the vehicles being transferred from the source factory's MES system to the target factory's MES system using a backend database-level replication method. The migration employed a database-level tablespace replication scheme, rather than importing data row by row at the application layer, to ensure data integrity and migration efficiency.

[0106] The migration data set contains four core data table types:

[0107] Table showing the relationship between engineering vehicle serial number and vehicle part number;

[0108] Table showing the correspondence between material numbers and controller assembly part numbers and software package numbers;

[0109] Software details table (including software and hardware version, flash file name, checksum, and file size);

[0110] Software configuration word data table (including configuration word short code, complete configuration word, and configuration validity conditions).

[0111] The migration process employs a batch compression and transmission method based on table partitioning: each table's data is split into multiple data fragments according to the primary key range. Each fragment is LZ4 compressed at the source before transmission, and decompressed at the destination before being written to the database. The compression algorithm uses intra-frame compression with a fixed block size, and each frame is independently verified to ensure that the corruption of a single frame does not affect the overall data. Transmission supports breakpoint resumption: when network interruption causes transmission to stop, the system records the sequence number of the successfully transmitted fragment. After the network is restored, transmission resumes from the breakpoint without retransmission. Immediately after each fragment is transmitted, on-the-fly verification is performed, comparing the CRC32 checksums of the source and destination. If the verification fails, the fragment is automatically retransmitted.

[0112] Step S9: Data consistency check;

[0113] Process engineers verify the consistency of software data between the source factory's MES and the target factory's MES. The verification employs a two-tiered approach: full comparison plus sampling verification.

[0114] Full comparison: The system automatically compares the number of records, key field values, and data status indicators of each table in the two MES databases and outputs a difference report;

[0115] Sampling verification: Process personnel randomly select 5% of typical engineering vehicle numbers and manually verify the consistency of their complete configuration chain (vehicle material number → part number → software package number → configuration word).

[0116] The verification includes: the number of engineering vehicle numbers, the range of material numbers, the software package version, the configuration short code, the data status identifier, and the version release timestamp. After verification, the MES system engineer clicks "Complete" in the production conversion application form in the CGVSP system, and the process jumps to the next step.

[0117] Step S10: Add a new factory code and bind a factory group in the CGVSP system;

[0118] After receiving the process notification, the ABOM engineer adds a new factory group for the model being converted to production in the factory grouping module of the CGVSP system. Access the factory grouping detailed configuration page:

[0119] Select all engineering vehicle numbers under this model;

[0120] Set "Factory Grouping Status" to "Enabled";

[0121] Bind the target factory code to establish a mapping relationship between the engineering vehicle number and the target factory.

[0122] After completing the factory group configuration, click "Complete" in the application process.

[0123] Step S11: The engineering view is automatically generated into the factory view;

[0124] This step is automatically triggered by the system. Based on the already activated factory grouping configuration, the CGVSP system automatically generates the target factory's factory view data from the vehicle model's engineering view data.

[0125] The difference between the Engineering View and the Factory View is that the Engineering View is a full configuration view for the R&D side, containing all optional configurations and common data for all factories; the Factory View is a production-available view for a specific factory, containing only the valid data corresponding to the production configuration of that factory, and the data status matches the release progress.

[0126] During the generation process, the system automatically performs the following steps:

[0127] Filter: Only retain the data corresponding to the engineering vehicle numbers selected in the factory group;

[0128] Mapping: Map and transform part numbers and package numbers in the engineering view according to the factory dimension;

[0129] Isolation: The generated factory view data is marked as "not to be distributed yet" and does not enter the distribution queue to avoid interfering with other factories;

[0130] Structured processing: Six types of standard JSON data files are generated according to the interface specifications and are ready for distribution.

[0131] After generation, the system determines the new original factory production mode:

[0132] If the original factory stops production and only the new factory is producing, proceed to step S12;

[0133] If the new original factory is producing on two parallel lines, proceed to step S14.

[0134] Step S12: Start the data retrieval task for the new factory;

[0135] After receiving the process notification, the MES system engineer initiates the software data retrieval task for the target factory's MES system. Once the retrieval task is initiated, the CGVSP system automatically resends all backlogged release software data since the pause in step S7 to the target factory's MES system in chronological order of release time.

[0136] The re-issuance process also employs a timed task-driven approach, distributing data in batches, with each batch's data size not exceeding a preset threshold to avoid network congestion. Once the re-issuance is complete, the new factory enters a normal data publishing and receiving state, and subsequent new data released by the R&D department is distributed at the normal pace.

[0137] After completing the operation, the MES system engineer finds the production conversion application process in the CGVSP system, clicks "Complete", and the process returns to the initiator, the process engineer.

[0138] Step S13: Clean up to factory condition;

[0139] After all production conversion work is completed and the new factory is in stable operation, the original factory status will be cleared:

[0140] If the original factory does not produce other models, the IT engineer will set the original factory's "factory status" to "stopped production" in the factory management module of the CGVSP system and close all its data retrieval channels.

[0141] If the original factory produces other models, the ABOM engineer will set the "Factory Group Status" of the original factory group corresponding to the model being converted to "Invalid" in the factory group module. This will only disable data distribution for that model and will not affect other models.

[0142] Finally, the process engineer closes the production conversion application form, and the entire process ends.

[0143] Step S14: New original factory dual-line parallel mode;

[0144] If the new and original factories operate in parallel after the production conversion, the MES system engineer will simultaneously initiate software data retrieval tasks for both the new and original factory MES systems. Once initiated, the CGVSP system will synchronously distribute all backlogged release data since the pause in step S7 to the MES systems of both factories.

[0145] During the parallel operation, CGVSP maintains separate plant views for each plant, with each plant independently counting version status to ensure data synchronization and non-interference. After the operation is completed, the process returns to the initiator, and the process engineer closes the production change request.

[0146] Branch 2: Incremental migration process for existing factories (steps 15 to 24).

[0147] This branch is suitable for scenarios where the target factory already exists and other models are in production. Its feature is incremental data migration without affecting existing models in production.

[0148] Step S15: Secondary judgment of the production conversion scenario;

[0149] Based on the production conversion plan, process engineers further determine the specific subtypes within the existing factory scenario:

[0150] If it falls under the "Adding Materials to an Existing Factory" scenario (the target factory already produces other vehicle models, and this time only new materials / engineering vehicle numbers are being added), proceed to step S23;

[0151] If it falls under the "all / partial material conversion" scenario (transferring all existing materials from the source factory to the target factory), continue to step S16.

[0152] Step S16: The process submits a factory conversion application at CGVSP;

[0153] Process engineers create a factory conversion request form in the CGVSP system. This differs from a completely new factory scenario:

[0154] If all materials are being converted to production, you only need to fill in the vehicle model; the system will automatically select all engineering vehicle numbers.

[0155] If it is a partial material conversion, you need to manually select the range of engineering vehicle numbers to be converted in the application form.

[0156] After completing the form, submit the application, and the process will be transferred to the ABOM engineer node.

[0157] Step S17: Maintain the plant grouping of transferred materials in the CGVSP system;

[0158] After receiving the application process, the ABOM engineer adds a target factory group for the model to be converted to production in the factory grouping module:

[0159] According to the scope of the application form, if all materials are to be converted to production, select all engineering vehicle numbers; if only some materials are to be converted to production, select the corresponding engineering vehicle numbers according to the list.

[0160] Set "Factory Grouping Status" to "Enabled";

[0161] The system automatically checks whether the selected vehicle number overlaps with existing groups in the factory. If there is an overlap, a warning is issued to avoid duplicate configuration.

[0162] After completing the configuration, click "Finish" to return the process to the process engineer.

[0163] Step S18: Import the production conversion data into the CGVSP;

[0164] The process engineer accesses the export module of the source factory's MES system, filters available data by "data status" (including three statuses: released, trial assembly, and pending release), and exports a software data list in a specific format to an Excel file. The exported list includes key fields such as: engineering vehicle number, part number, software package number, version number, and configuration short code.

[0165] Then, enter the factory view module of the CGVSP system, click the "Import" button, select the Excel file exported from MES, and perform the import operation. Before importing, the system performs format and duplicate checks. If there are format errors or duplicate data, an error message will be displayed and the import will be terminated.

[0166] Step S19: CGVSP automatically processes data and distributes it to MES;

[0167] The CGVSP system parses the imported Excel list data, matches the corresponding data from the engineering view, adds it to the target factory's factory view, and generates a standard JSON format distribution file, which is then distributed to the target factory's MES system via a scheduled task queue.

[0168] The core matching logic for this step is as follows: generate JSON files for each of the six data categories:

[0169] Engineering vehicle number and vehicle part number relationship data: Based on the imported engineering vehicle number list, generate the corresponding vehicle part number mapping relationship JSON.

[0170] Engineering vehicle number and part number relationship data: Based on the engineering vehicle number and part number in the imported list, match the BOM structure in the engineering view. If the match is successful, generate the relationship data JSON; if the match is not successful, display an error message and stop the import.

[0171] Part number / package number and factory, abbreviation relationship data: Based on the imported part number and package number, match the factory dimension attribute in the engineering view to generate factory mapping JSON.

[0172] Software details data: Includes complete attributes in JSON format, such as software and hardware version, file name, file size, SHA256 checksum, and release date.

[0173] Engineering vehicle number and configuration word short code relationship data: Match the latest configuration word short code corresponding to each engineering vehicle number for production conversion, and generate a mapping JSON.

[0174] Complete configuration word data: Includes the full JSON of the configuration word content, value range, and dependency conditions.

[0175] After all matches are successful, the generated six types of JSON files are pushed to the target factory's MES via a scheduled task queue. If multiple versions of the same part number or software package number exist, they are sent out in the order of release time. When the MES receives the files, it automatically assigns a status according to the order of arrival: the first received file is in the "Released" status, the second received file is in the "Trial Assembly" status, and the third and subsequent files are in the "Pending Release" status.

[0176] If any errors occur during the import process, the process personnel should confirm and update the data based on the error message and then re-execute the import operation.

[0177] Step S20: Data consistency verification and status processing;

[0178] After the target factory's MES system receives the data, the process engineer exports the data list from the new original factory's MES system and performs a consistency comparison to confirm the completeness of the transition data. Once completeness is confirmed, the data status indicators are further compared; if inconsistencies are found, they are manually adjusted in the new MES system.

[0179] The consistency verification in this step uses a hash tree verification method: a Merkle root hash value is calculated for each type of data file, and the source end and the target end are compared with the root hash to quickly verify the overall integrity; if the root hash is inconsistent, the specific data fragment with the difference is located layer by layer to improve the verification efficiency.

[0180] After data confirmation and processing, the process engineer closes the process of the production change application form in the CGVSP system.

[0181] Step S21: Determine the original factory's production status;

[0182] After the new factory officially starts production, ABOM engineers will determine whether the models being converted to be produced will continue to be produced at the original factory based on the process inputs.

[0183] If the original factory continues production, the process will naturally end, and the two factories will produce in parallel.

[0184] If the original factory stops production of the model to be switched to, proceed to step S22.

[0185] Step S22: Clean up to factory condition;

[0186] Perform factory cleanup, following the same procedure as step S13:

[0187] If the original factory has no other models, then the entire production line will be marked as "discontinued".

[0188] If the original factory has other models, simply set the corresponding group to "invalid".

[0189] Step S23: Configure factory grouping for new materials in existing factories.

[0190] For the sub-scenario of "adding materials to existing factories", ABOM engineers can directly find the existing group of the corresponding vehicle model in the factory grouping module of the CGVS system, click "edit" to enter the settings page, check the newly added engineering vehicle number, and complete the factory grouping configuration for the new material.

[0191] In this scenario, there is no need to create a separate conversion application form; it can be added directly to the existing group, making the process more streamlined.

[0192] Step S24: CGVSP automatically distributes new material data;

[0193] CGVSP, based on ABOM grouping changes, automatically adds the software data of new materials from the engineering view to the plant view according to the initial full-version release logic, generates a standard JSON file, and distributes it to the target plant's MES system via a scheduled task. After distribution, the new material can be put into production at the target plant.

[0194] Step S25: Process ends;

[0195] Once all production conversion operations are completed, the production conversion application form is closed, the process is archived, and all operation logs are stored in the production conversion file for future auditing and traceability.

[0196] like Figure 5 As shown, the software data migration device for this factory's production conversion includes: an application receiving module, a scene discrimination module, a data processing module, a verification module, and a channel control module;

[0197] The application receiving module receives production conversion application information, including the vehicle model to be converted and the target factory identifier. This module provides a user interface for process engineers to fill out the production conversion application form, supporting operations such as vehicle model selection, factory code input, and material number selection. After submission, the application automatically enters the approval process, with real-time progress display.

[0198] The scenario discrimination module is used to determine whether the target factory is a brand new factory based on the target factory identifier. This module queries the factory information database, comparing the factory code in the application form with the list of existing factories. If a match is found, the factory is considered an existing factory; otherwise, it is considered a brand new factory. The discrimination result serves as the basis for subsequent data processing modules to select different processing logics.

[0199] The data processing module, the core module of this device, is used to pause the data retrieval task from the original factory and completely migrate the software data of the original factory to the target factory when the target factory is a brand new factory; when the target factory is an existing factory, it executes the corresponding data distribution logic according to the type of transferred materials, maps the engineering view data to the factory view data, and then distributes it. The data processing module further includes: a new factory data migration submodule, an incremental data distribution submodule, a view mapping submodule, and a version sorting submodule.

[0200] The verification module is used to perform data consistency verification after data migration is complete. This verification includes comparing key fields of the source and target data, supporting both full hash comparison and sampling comparison modes. Upon completion of verification, a verification report is generated, indicating the number of consistent items, the number of differing items, and details of the differences.

[0201] The channel control module is used to enable the data retrieval channel of the target factory after the data consistency verification is passed. At the same time, during the production start-up, the channel control module is also responsible for sending a pull pause command to the original factory MES and automatically re-sending all published data temporarily stored during the pause after the channel is restored.

[0202] like Figure 6 As shown, the factory conversion software data migration system includes: the upstream software data platform, the original factory manufacturing execution system, and the target factory manufacturing execution system.

[0203] The upstream software data platform (CGVSP system) is the core hub of the entire system. It receives production conversion application information, determines the target factory type based on the target factory identifier, and executes the corresponding data processing logic based on the determination result. It then maps the engineering view data to factory view data and distributes it. The upstream software data platform communicates with the MES systems of each factory through a standard JSON interface protocol.

[0204] The original equipment manufacturer's (OEM) Manufacturing Execution System (MES) communicates with the upstream software data platform to receive pull pause commands and suspend data pull tasks during production transitions. The OEM MES stores all software data required for production at the factory, including engineering vehicle numbers, material numbers, software package numbers, configuration words, etc.

[0205] The target factory manufacturing execution system (target factory MES) communicates with the upstream software data platform to receive and store factory view data sent from the upstream software data platform. The target factory MES automatically assigns data status according to the version order of the received data.

[0206] The three systems work together to form a complete software data migration chain for factory conversion: the upstream platform is responsible for centralized management and intelligent distribution of data, the original factory MES is responsible for providing and pausing source data, and the target factory MES is responsible for receiving and storing the migrated software data.

[0207] When the target factory is an existing factory and only new materials need to be added, this invention provides a more simplified automated process. In this scenario, process engineers only need to specify the "new material" type and the corresponding engineering vehicle number in the production conversion application form. After the ABOM engineer selects the newly added engineering vehicle number in the factory group, all subsequent data processing is automatically completed by the system: the CGVS system automatically identifies the engineering view data of the new material, automatically generates the factory view, and automatically sends it to the target factory's MES through a scheduled task, without the need for manual export and import operations. This scenario has the highest degree of automation and the least human intervention, making it particularly suitable for scenarios where existing factories continuously introduce new models or configurations.

[0208] To further enhance the security and reliability of the data migration process, the present invention may also include the following enhanced embodiments:

[0209] Resumable Download Example: When the transmission of a JSON data file is interrupted due to network fluctuations, the system automatically records the position of the successfully transmitted data. Once the network is restored, the transmission resumes from the point of interruption, without needing to retransmit from the beginning. This implementation method is particularly suitable for full data migration scenarios with large data volumes, effectively avoiding migration failures caused by network instability.

[0210] Data is transmitted in encrypted form. JSON data transmission between CGVS and MES uses an HTTPS encrypted channel. At the same time, the content of the data file is signed and verified. After receiving the data, the target factory MES verifies the signature before entering it into the database to prevent the data from being tampered with or stolen during transmission.

[0211] Exception rollback example: When data consistency verification fails or a serious error occurs during the distribution process, the system supports one-click rollback to the data state before migration, restoring the data of the target factory's MES to the snapshot before migration, ensuring rapid recovery in abnormal situations and not affecting production progress.

[0212] During the data migration process for production transition, a segmented, breakpoint-resumption implementation method was adopted for large-volume software packages and large batches of data tables:

[0213] The file to be transferred is divided into contiguous chunks of a fixed size (default 16MB), and each chunk is assigned a unique chunk number;

[0214] The sending end maintains a transmission progress bitmap to record the transmission status of each segment (not sent / sending / acknowledged);

[0215] Each time the receiving end successfully receives and verifies a fragment, it returns an acknowledgment to the sending end, carrying the fragment sequence number;

[0216] When a network interruption or connection timeout causes transmission to stop, the sender retains the current progress bitmap; once the network is restored, transmission continues from the unacknowledged fragments in the bitmap, without having to start from the beginning.

[0217] Each segment is retransmitted independently after a timeout. After five failed retransmissions of a single segment, a manual alarm is triggered to notify the MES engineer for troubleshooting.

[0218] The transmission process uses a segmented verification method, verifying each data segment as it is transmitted, rather than verifying only after the entire data has been transmitted.

[0219] The sending end appends a CRC32 checksum to the end of each data fragment;

[0220] Upon receiving a fragment, the receiving end immediately calculates the checksum and compares it with the additional checksum.

[0221] If the verification passes, the data is written to local storage and an acknowledgment is returned; if the verification fails, the fragment is discarded and a retransmission is requested.

[0222] The above embodiments can detect and retransmit erroneous fragments in real time, avoiding the need for overall retransmission when errors are discovered only after all transmissions are completed, thus greatly saving fault recovery time; at the same time, it does not affect the parallel transmission of other fragments, resulting in higher overall transmission efficiency.

[0223] To ensure reliable compressed transmission, data transmission employs the LZ4 lossless compression algorithm, guaranteeing both compression speed and reliability.

[0224] Compression uses a fixed frame length (64KB) for independent compression, with each frame containing an independent frame header checksum and length identifier;

[0225] The decompression end first verifies the integrity of the frame header, and then decompresses the data body; damage to a single frame only affects that frame and will not spread to subsequent data;

[0226] Calculate the checksum before and after compression to ensure that the decompressed data is completely consistent with the data before compression.

[0227] For extremely small files (less than 1KB), compression is automatically skipped to avoid compression overhead outweighing the benefits.

[0228] The transmitted data also has an integrity verification method. After the migration is completed, the data integrity verification adopts a three-level verification system:

[0229] Level 1: Record count verification. Compare the total number of records in each data table on the source and target sides to quickly determine if any records are missing.

[0230] Level 2: Key Field Hash Verification. For each type of data, calculate a batch hash of the key fields, compare the source and target hash values, and verify data consistency.

[0231] Level 3: Business Link Sampling Verification. Typical engineering vehicle numbers are extracted, and the complete business link is traced layer by layer from vehicle to material to part to software package to configuration word, verifying the correctness of the business logic and the accuracy of the data status.

[0232] The migration can only be considered complete after all three levels of verification have passed.

[0233] Data security measures throughout the migration process:

[0234] The transmission channel uses TLS 1.3 encryption and two-way certificate authentication at both ends to prevent data eavesdropping and tampering;

[0235] Sensitive configuration word data is additionally encrypted at the field level before transmission and remains encrypted when stored.

[0236] The migration operation is fully audited, recording who operated on which data at what time, supporting post-event traceability;

[0237] Before writing data to the target factory, permission verification is performed to ensure that only authorized accounts can perform data writing.

[0238] Automatically back up the original data of the target factory before migration, and roll back to the previous state with one click if the migration fails.

[0239] To improve migration speed, the following performance optimization methods are adopted:

[0240] Multi-threaded parallel transmission: Different types of data files are assigned to different threads for parallel transmission without blocking each other;

[0241] Dynamic flow control: Automatically adjusts the fragment size (default fragment size is 16MB, which can be configured within the range of 8 to 32MB depending on the factory bandwidth) and concurrency based on the current network bandwidth. When the bandwidth is high, the fragment size is increased to improve throughput, and when the bandwidth is low, the fragment size is decreased to avoid packet loss.

[0242] Data deduplication: Before transmission, compare the data with the existing data in the target factory and only transmit the differences to avoid repeatedly transmitting the same version of data;

[0243] Scheduled tasks to avoid peak hours: Large-scale data migrations are scheduled to be performed during off-peak hours at night to avoid consuming production bandwidth.

[0244] The above description is merely an illustration of preferred embodiments of the present invention, and the above technical features can be arbitrarily combined to form multiple embodiments of the present invention.

[0245] The present invention has been described above by way of example with reference to the accompanying drawings. Obviously, the specific implementation of the present invention is not limited to the above-described manner. Any non-substantial improvements made using the concept and technical solution of the present invention, or the direct application of the concept and technical solution of the present invention to other occasions without modification, are all within the protection scope of the present invention.

Claims

1. A method for migrating software data during factory production transition, characterized in that, include: Receive production conversion application information, which includes the vehicle model to be converted and the target factory identifier; Determine whether the target factory is a brand new factory based on the target factory identifier; If it is a brand new factory, the data retrieval task of the original factory is suspended, the software data of the original factory is completely migrated to the target factory, and the data retrieval channel of the target factory is enabled after the data consistency verification is passed. If it is an existing factory, the corresponding data distribution logic is executed according to the type of transferred material, and the engineering view data is mapped to the factory view data and then sent to the target factory.

2. The factory conversion software data migration method according to claim 1, characterized in that, The suspension of the original factory data retrieval task includes: Send a pull pause command to the manufacturing execution system of the original factory, the pull pause command carrying the conversion application number and vehicle model identifier; The newly released software data received by the original factory during the suspension period is temporarily stored in the cache queue of the upstream platform. Once the data retrieval channel of the target factory is enabled, all data in the cache queue will be resent from the target factory.

3. The factory conversion software data migration method according to claim 2, characterized in that, The complete migration of software data from the original factory to the target factory includes: By copying the underlying database tables, the software data tables in the original factory's manufacturing execution system are fully copied to the target factory's manufacturing execution system. The software data tables include a table showing the relationship between the engineering vehicle number and the material number, a table showing the relationship between the material number and the controller part number and the software package number, a software details table, and a software configuration word data table.

4. The factory conversion software data migration method according to claim 1, 2, or 3, characterized in that, The step of executing the corresponding data distribution logic based on the type of transferred material includes: If the target factory is an existing factory and the transferred material is a new material, then data distribution will be automatically triggered according to the release process number, and the engineering view data corresponding to the new material will be automatically added to the factory view of the target factory and distributed. If the transferred material is part or all of the material, the specified software data is exported from the manufacturing execution system of the original plant, imported into the upstream platform, and then the upstream platform matches the engineering view data according to the imported list, generates the plant view data, and distributes it.

5. The factory conversion software data migration method according to claim 4, characterized in that, The process of mapping engineering view data to factory view data and then distributing it includes: The upstream platform, based on the factory identifier of the target factory, filters software data corresponding to the model being converted for production from the engineering view, and generates six types of JSON data files according to a predefined interface protocol. These six types of JSON data files include: Data files containing the relationship between the engineering vehicle number and the vehicle part number, the relationship between the engineering vehicle number and the part number, the relationship between the part number and the software package number and the factory, the software details data file, the relationship between the engineering vehicle number and the configuration word short code, and the complete configuration word data file; The six types of JSON data files are pushed to the manufacturing execution system of the target factory via a scheduled task.

6. The factory conversion software data migration method according to claim 5, characterized in that, Also includes: When there are multiple versions of the same part number or the same software package number, the data of each version shall be distributed in the order of release time. The target factory's manufacturing execution system automatically assigns status identifiers to each version of data according to the order of receipt: the first received version is marked as release status, the second received version is marked as trial assembly status, and the third and subsequent received versions are marked as pending release status.

7. The factory conversion software data migration method according to claim 1 or 6, characterized in that, After enabling the data retrieval channel for the target factory, the following is also included: Determine whether the original factory should continue producing the model to be converted; If the original factory no longer produces the model to be converted, then the factory group status of the original factory will be set to invalid, or the overall status of the original factory will be set to shutdown.

8. A software data migration device for factory production transition, characterized in that, include: The application receiving module is used to receive production conversion application information, which includes the vehicle model to be converted and the target factory identifier; The scene discrimination module is used to determine whether the target factory is a brand new factory based on the target factory identifier; The data processing module is used to pause the data retrieval task of the original factory and completely migrate the software data of the original factory to the target factory when the target factory is a brand new factory. When the target factory is an existing factory, the corresponding data distribution logic is executed according to the type of transferred material, and the engineering view data is mapped to the factory view data before being distributed. The verification module is used to perform data consistency verification after the data migration is completed. The channel control module is used to enable the data retrieval channel of the target factory after the data consistency verification is passed.

9. The factory conversion software data migration device according to claim 8, characterized in that, Also includes: The version sorting module is used to sort the version data in chronological order of release time when there are multiple versions of the same part number or the same software package number. The status allocation module is used to automatically assign release status, trial installation status, or pending release status to each version of data according to the order of receipt.

10. The factory conversion software data migration device according to claim 9, characterized in that, Also includes: The view mapping module is used to filter software data corresponding to the production model from the engineering view based on the factory identifier of the target factory, and generate six types of JSON data files according to a predefined interface protocol.

11. The factory conversion software data migration device according to claim 10, characterized in that, Also includes: The channel pause module is used to send a pull pause command to the manufacturing execution system of the original factory, and temporarily store newly released software data received during the pause in a cache queue; The data resending module is used to resend all data in the cache queue from the target factory after the data retrieval channel of the target factory is enabled.

12. The factory conversion software data migration device according to claim 8, 9, 10 or 11, characterized in that, Also includes: The status cleanup module is used to set the factory group status of the original factory to invalid or the overall status of the original factory to shutdown when the original factory no longer produces the model to be converted.

13. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the factory conversion software data migration method according to any one of claims 1 to 7.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the factory conversion software data migration method as described in any one of claims 1 to 7.

Citation Information

Patent Citations

  • A Multi-Factory Capacity Coordination Method Based on MES System

    CN116976653B