Product two-dimensional code recognition and data tracing system
The automated product QR code recognition and data traceability system has solved the problems of low QR code recognition efficiency and difficult data traceability in semiconductor manufacturing, achieving efficient production management and quality control, and improving production efficiency and the accuracy of data traceability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HUATIAN TECH XIAN
- Filing Date
- 2026-01-07
- Publication Date
- 2026-06-05
Smart Images
Figure CN122155734A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of industrial automation control and semiconductor manufacturing technology, and specifically to a product QR code recognition and data traceability system. Background Technology
[0002] In the semiconductor manufacturing industry, a single chip undergoes hundreds of precise testing processes, and the integrity and traceability of the inspection data directly determine product quality and industry compliance. Faced with increasingly stringent regulatory requirements, traditional manual management models are no longer sufficient. Chip manufacturing is a complex process involving multiple stages, including design, manufacturing, packaging, and testing, each generating massive amounts of data. However, traditional laboratory management methods, such as manual recording, Excel spreadsheets, and isolated systems, result in scattered data, inconsistent formats, and difficulty in sharing, creating "data silos." Under the traditional model, the identification of defective chips relies on manual sampling or offline analysis, which is not only inefficient but also prone to material waste and batch traceability difficulties. This phenomenon not only affects production efficiency but also increases the difficulty of quality traceability, failing to meet the industry's needs for full product lifecycle management.
[0003] In existing technologies, manual recording is prone to errors, and handwritten records are easily missed or tampered with, affecting the reliability of wafer testing results; there is also the problem of data silos, with equipment, personnel, and batch information being stored in a scattered manner, requiring retrieval across multiple systems during traceability; audit risks are exacerbated, as it is impossible to track the sample flow path in real time, and it does not comply with electronic data standards such as GAMP5.
[0004] However, in the precision manufacturing processes of semiconductor chips and electronic components, it is necessary to accurately identify and verify the QR code identifying information of each product in order to distinguish between good and defective products and trace production data. Currently, QR code recognition and whitelist verification rely on manual operation or decentralized systems, resulting in low efficiency, susceptibility to missed detections / false judgments, untimely sorting of defective products, and difficulties in data traceability. This leads to high production management costs and lagging quality control. Therefore, the market needs a more efficient and accurate data traceability system and method. Summary of the Invention
[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a product QR code recognition and data traceability system. This invention is applicable to QR code recognition and whitelist verification systems that rely on manual operation or decentralized systems, which suffer from low efficiency, easy missed detections / false judgments, untimely sorting of defective products, and difficulties in data traceability.
[0006] To achieve the above objectives, the technical solution adopted by this invention is: a product QR code recognition and data traceability system, comprising: a print triggering and information generation module, which triggers the printer in conjunction with the production control system by calling the PWS interface to generate a 2D information file for a single product containing a unique identifiable unit ID; an IT data capture and file transfer module, used to capture the 2D information file for a single product containing unit ID data and a coplanarity Copl parameter information file from the production process and upload it to a preset server path; and a data traceability module, used to automatically aggregate the 2D information file and coplanarity Copl parameter information file data of a single product to the CIM module database, forming a database of multiple products, enabling full-process traceability of information for each product in batch production.
[0007] Optionally, the IT data capture and file transfer module captures a 2D information file containing unitID data for a single product from the production process. The unitID data is named with the work order number and is in the format of a .2D file storing a single QR code information per line.
[0008] Optionally, the IT data capture and file transfer module uploads the 2D information file of the single product and the coplanarity Copl parameter information file corresponding to the single product to the following server path via FTP protocol: http: / / 192.162.9.8.228 / Leadscan / LS-2DMatrix / Cop / Mark-flder / SZ228 / file / preset server path.
[0009] Optionally, the IT data capture and file transfer module needs to capture the product's 2D information file and coplanarity Copl parameter information file at each station where a single product enters or leaves the station, and synchronize them to a preset server path to ensure that the data can be directly accessed when a single product is transferred to the next station, avoiding data gaps in traceability.
[0010] Optionally, the IT data capture and file transfer module calculates the captured processing data in real time through software, automatically summarizes and generates processing reports, and automatically uploads them to FTP software to synchronously upload the reports to a fixed path on the cloud drive to avoid data loss or tampering.
[0011] Optionally, the 2D information file and the coplanarity Copl parameter information file specifically include: machine number, processing date, customer code, packaging type, work order, single product serial number, Matrix code, and Copl parameters.
[0012] Optionally, the data traceability module reads reports from the FTP terminal through a scheduled task, captures 2D information and Copy data in fixed rows and columns, summarizes them into an Excel report in a standard format, and embeds it into the web page of the automated CIM software. Users can accurately query by conditions such as equipment number, processing date, customer code, and work order.
[0013] Optionally, the equipment modification and whitelist configuration module is used to modify the equipment hardware and upgrade the software, configure the QR code whitelist and cache it locally; the product surface is illuminated by 11 sets of light sources to eliminate interference such as QR code reflection and blurring, ensuring that the industrial camera can clearly capture 1-20 character information. The whitelist only stores information on homogeneous customers, a total of 15 customer codes, which can be added as needed to avoid invalid data occupying storage resources.
[0014] Optionally, it also includes a comparison and sorting module, which automatically sorts products by recognizing the unit QR code content, extracting the unit ID, and comparing it with a whitelist. In particular, for the characteristics of small size and high batch size of semiconductor products, the sorting algorithm adds a tray coordinate memory function: the machine will record the row and column coordinates of good and defective products in the tray in real time. During the sorting stage, the software drives the sorting mechanism through a customized algorithm to accurately guide good products into the good product track and defective products into the defective product track, avoiding classification errors caused by manual intervention.
[0015] Optionally, the 2D information file contains two main parts: first, a work order identifier, used for file naming association; and second, a 2D code for all individual products under the entire work order, i.e., the unique content of the QR code, which provides a standard data source for subsequent machine identification and comparison, ensuring that the judgment criteria for each product are strongly bound to the production work order, fundamentally avoiding the problem of mixed judgment across work orders, and ensuring the accuracy of traceability data.
[0016] The beneficial effects of this invention are: This invention significantly improves the automation and efficiency of product QR code recognition and data traceability. Data capture and uploading throughout the entire semiconductor production process requires no manual intervention, reducing the time from QR code recognition to defective product sorting and increasing production efficiency by more than 10%. Quality control is precise, with real-time comparison against whitelists, reducing the rate of missed defects and false positives to below 0%, effectively intercepting risky products. A closed-loop data traceability system integrates full-process data through the CIM module, supporting rapid querying and traceability, and facilitating production problem analysis and continuous improvement. Attached Figure Description
[0017] Figure 1 A schematic diagram of the QR code recognition and data traceability system for the product of this invention; Figure 2 This is a schematic diagram of the QR code recognition and data traceability system for the product of this invention; Figure 3This is a schematic diagram of data capture and file transfer in the QR code recognition and data traceability system of the present invention; Figure 4 This is a schematic diagram of the naming of a 2D format file for the product QR code recognition and data traceability system of this invention; Figure 5 This is a schematic diagram of the sorting process of the QR code recognition and data traceability system for the product of this invention; Figure 6 This is a schematic diagram of the traceability interface of the QR code recognition and data traceability system of the present invention.
[0018] Attached reference numerals: 1. Print triggering and information generation module; 2. IT data capture and file transfer module; 3. Equipment modification and whitelist configuration module; 4. Comparison and sorting module; 5. Data traceability module. Detailed Implementation
[0019] The preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby providing a clearer and more explicit definition of the scope of protection of the present invention.
[0020] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.
[0021] This invention provides a product QR code identification and data traceability system. Traceability typically refers to the ability to track and trace the origin of each component of a product. Generally, traceability marking is achieved by printing codes or 2D codes on the product or adding RFID. Semiconductor failures can cause severe reputational damage to original equipment manufacturers (OEMs) and Tier 1 and Tier 2 suppliers, while also placing a heavy burden on the supply chain. Therefore, having traceability capabilities for individual products is crucial for chip manufacturing. Once a failure occurs, manufacturers need to quickly and accurately locate the problem. If the failure is related to a semiconductor component, the traceability marking on the component can quickly pinpoint the specific situation of the device during the manufacturing process, enabling faster handling of anomalies, more accurate identification of defective product batches, and timely implementation of measures to reduce quality costs.
[0022] In semiconductor manufacturing, the unique QR code identification information and Coplanarity Copl parameters on the surface of a single product are key process data that affect packaging accuracy, and are also the core data that are most difficult to trace in the traditional model. The absence of both will directly lead to a gap in product quality traceability and make it impossible to locate the point at which defective products are generated.
[0023] In one specific embodiment of this application, as follows: Figures 1 to 6 As shown, the traceability system specifically includes: a print triggering and information generation module 1, which triggers the printing timing by calling the PWS interface and generates a 2D information file containing the unit ID; full-process traceability of a single chip product is achieved by encoding the frame or substrate material through the system, forming the basis for stripe and product uniqueness. Simultaneously, the system automatically monitors and records the processing steps of each process, and finally prints a unique traceability code for each product on the back of the chip based on the process information. Specifically, the PWS interface is linked with the production control system through a specific communication protocol, such as TCP / IP. When the product arrives at the designated workstation, it triggers the printing command, generating a 2D information file containing encoding rules to ensure that the unit ID is uniquely identifiable. The one-piece-one-code solution assigns a unique identification code to each product, which allows traceability of data from various stages, including production information, quality inspection records, and transportation routes. At each stage of product production, data such as production time, production process, and raw material information are recorded, and a unique identification code is generated corresponding to each. This identification code can be in the form of a barcode, QR code, RFID chip, etc., to ensure unique identification.
[0024] In one specific embodiment of this application, after the printing station completes the QR code printing, the machine automatically calls the interface the instant the product leaves the station, reads the 2D code information through dedicated software, and generates a .2D format whitelist file. The file naming strictly follows the "work order number" rule, and the content is presented in the format of "each QR code information occupies a single line"; this .2D format is the only readable format preset by the machine software, ensuring the compatibility and accuracy of subsequent automatic comparison from the source.
[0025] In one specific embodiment of this application, the IT data capture and file transfer module 2 is used to capture data containing unit IDs in the production process, generate .2d files, and transfer them to a preset server path. The files are uploaded to a designated server path via the FTP protocol, requiring permission management to ensure data security. The .2d files are named with work order numbers and are formatted as .2D files storing a single QR code information per line. The fixed server path is: http: / / 192.162.9.8.228 / Leadscan / LS-2DMatrix / Cop / Mark-flder / SZ228 / file / , which can be adjusted as needed. Specifically, a scripting language, such as Python, is developed to periodically scan the production database, capture unitID data, and generate 2D files. FTP is one of the protocols in the TCP / IP protocol suite. The FTP protocol consists of two components: an FTP server and an FTP client. The FTP server is used to store files, and users can use an FTP client to access resources located on the FTP server via the FTP protocol. When developing a website, the FTP protocol is typically used to transfer web pages or programs to a web server.
[0026] In one specific embodiment of this application, the machine calculates processing data in real time through software and automatically generates processing reports; the locally deployed LOG automatic upload FTP software synchronously uploads the reports to a fixed path on the cloud drive, preventing data loss or tampering.
[0027] In one specific embodiment of this application, the equipment modification and whitelist configuration module 3 is used to modify the hardware and upgrade the software of the LS device, configure the QR code whitelist and cache it locally; and set the QR code whitelist using tools such as a dongle. When the system boots up, it automatically downloads the whitelist file from a specified path to local storage. LS devices are automatic IC appearance inspection devices in the back-end semiconductor packaging process. LS is an abbreviation for Lead Scan, and it is a key device for automated detection of appearance defects in integrated circuits in the back-end semiconductor packaging process, covering quality control in multiple stages such as wafer dicing, wire bonding, and chip packaging. This device uses laser scattering and dark-field imaging technology to identify 23nm-level particle contamination, scratches, and bonding parameter anomalies, and the detection runs through the entire process, including silicon wafer certification and production process control inspection. Mainstream equipment models include the ICOS brand LS-7700 and Hitachi High-Tech's LS series, whose technical specifications are deeply related to the cutting wheel specifications and packaging inspection standards. With the increasing proportion of advanced processes below 7nm, the capital expenditure of LS-type equipment has exceeded 20%, becoming a core link in ensuring the reliability of automotive-grade chips. A dongle is a hardware encryption device used for protecting computer software copyrights. It typically connects to a computer's parallel port or USB interface, verifying authorization status through data exchange between hardware and software. Core technologies include microcontroller encryption algorithms, non-volatile storage, and anti-tracking design. The terminology was coined by Rainbow World and has become an industry-standard name. Its main function is to prevent software piracy. A continuous connection to the dongle is required during operation, and some products support remote upgrades and AES encryption technology.
[0028] In one specific embodiment of this application, the comparison and sorting module 4 identifies the QR code through the Vision system and compares it with the whitelist to automatically sort PASS / Fail products; the device host automatically sends the .2D whitelist file to the Vision system. After Vision identifies the unit QR code content, it compares it with the whitelist in real time. If the comparison matches, the product passes and flows into the next process; if the comparison does not match, the product fails and is automatically sorted into the defective product tray and marked for re-inspection.
[0029] In one specific embodiment of this application, the data traceability module 5 is used to aggregate 2D and CO data to the CIM module, realizing full-process traceability of production. It reads reports from the FTP terminal via a scheduled task, captures fixed rows and columns of 2D information and Copl data, aggregates them into an Excel report in a standard format, and embeds it into the automated CIM software web interface. Users can accurately query using conditions such as "equipment number, processing date, customer code, and work order." Users can log in from any computer in the backend to trace the product information of the corresponding work order, achieving systematic management of the entire process data of the LS equipment. The final traceable data includes: machine number, processing date, customer code, packaging form, work order, single product serial number, Matrix code, and Copl parameters. The 2D and CO data of a single product are automatically aggregated to the CIM module and presented through database association and a visual interface, supporting full-process traceability of production, such as work order number, time, equipment parameters, and good / defective product status. The CIM system is an indispensable cornerstone of modern manufacturing, especially the semiconductor industry. It integrates various resources, technologies, processes, and management systems in the manufacturing process through computer technology, forming a highly collaborative digital production system. A Computing Information Modeling (CIM) system is a complex system integrating management, information, automation, and systems engineering technologies, designed to achieve efficient integration and optimization of the manufacturing process. In the semiconductor industry, the CIM system is known as the "commander-in-chief" of the factory, responsible for coordinating complex manufacturing processes. It encompasses multiple areas, from production planning and equipment control to quality management, supporting high levels of automation and intelligence in wafer fabs. The CIM system acts like the brain of a "smart city," ensuring that every department, equipment, process, and quality control within the "city" performs the right task at the right time through the efficient coordination of information flow, data transmission, and wafer logistics.
[0030] In a specific embodiment of this application, the core functions of CIM can be summarized in the following four aspects: 1. Allocation and management of production materials and human resources: In semiconductor manufacturing, each wafer undergoes hundreds of processes. CIM ensures accurate and efficient resource allocation through data analysis and optimization models. For example, when scheduling orders, MES combines the equipment status collected by EAP to select the optimal path to complete production. 2. Programming and modeling: CIM relies on modeling technology to simulate and optimize the process flow in advance, thereby predicting potential problems. Similar to a "sand table simulation," it helps engineers choose the best process route. 3. Automated manufacturing: The CIM system connects to production equipment, enabling fully automated wafer transfer, such as seamless switching between AGVs, OHT systems, and production tasks. For example, EAP automatically loads production recipes and adjusts equipment parameters in real time, avoiding errors caused by manual intervention. 4. Quality control: CIM monitors quality indicators in real time through modules such as SPC and FDC, and analyzes defect data through YMS. If a batch of wafers deviates, the system can quickly locate the problematic process step, preventing the spread of quality problems.
[0031] In one specific embodiment of this application, the CIM module enables data integration and interaction; the core of the CIM system lies in data interaction. It collects equipment operation data through EAP and uploads it to MES for unified management. The challenge of data integration lies in the compatibility of diverse equipment protocols, requiring CIM to possess high flexibility. Real-time performance and efficiency are crucial; semiconductor production has extremely high time requirements, and CIM needs millisecond-level response capabilities. For example, the RTS scheduling algorithm needs to calculate the optimal equipment allocation scheme in real time to ensure production continuity. Process control and optimization are also essential; CIM's control logic is based on cybernetics, emphasizing "feedback" and "correction." Through the SPC module, CIM can analyze process data fluctuations in real time, trigger abnormal alarms, and automatically adjust process parameters. Knowledge graphs and intelligent decision-making are also important; by labeling and knowledge-based processing of data, CIM can form dynamic process graphs to assist in decision-making. For example, YMS uses historical defect data to train algorithm models to predict the yield of the next process.
[0032] In one specific embodiment of this application, the CIM module can be used in highly complex manufacturing processes. Semiconductor wafer fabs have complex and long manufacturing cycles, and errors in any stage can affect the yield of the entire batch of products. CIM significantly reduces human error through end-to-end data monitoring and automated control. Large-scale production and high-precision requirements mean that each wafer in a 12-inch wafer fab undergoes thousands of processes. CIM ensures the consistency of parameters for each piece of equipment and controls process deviations through modules such as SPC. Equipment utilization optimization is also crucial, as equipment cost is one of the largest investments in a wafer fab. Through EAP and RTD modules, CIM achieves efficient equipment utilization and reduces idle time. Furthermore, global supply chain collaboration is essential, as the semiconductor industry supply chain is geographically dispersed. CIM enables collaboration between production planning and the supply chain by sharing data with upstream and downstream companies.
[0033] In one specific embodiment of this application, the print triggering and information generation module 1 generates a 2D information file containing a unit ID. Specific encoding rules include QR code version and error correction level, with the unit ID being uniquely identifiable. By dynamically recording defect information of each chip on the substrate, such as cracks and misalignments, defective products are automatically skipped in critical processes such as die bonding and wire bonding, avoiding ineffective processing. Statistics show that this function can reduce material waste by 15%-30% while improving production line throughput efficiency. Based on the unique unit ID, the system binds information such as substrate batch, supplier, and process parameters, achieving bidirectional traceability from raw materials to finished products. Once a quality problem is detected, the problematic process, equipment, or even operator can be quickly located, shortening the quality analysis cycle by more than 50%.
[0034] In one specific embodiment of this application, for advanced technologies such as multi-chip packaging and stacked chips, the position of dies from different wafer sources on the substrate is accurately tracked to ensure controllable yield of complex structures. After the substrate enters the production line, a unique unit ID is generated by a 2D marking machine and bound to the substrate's reference information, dimensions, material specifications, etc., to form an initial mapping, which defaults to all good products. Process-level interaction: After each process is completed, the equipment automatically uploads the current chip status, good / defective product coordinates and types, and updates the strip mapping to the database.
[0035] In one specific embodiment of this application, the naming rule "work order number_timestamp.2D" enables one-click association between "work order" and "file". When the machine calls the file, no additional matching is required; it directly reads the corresponding file according to the work order, significantly reducing operation time. Secondly, the format generated by the machine software after recognizing the product's QR code is completely consistent with the .2D file format, providing a unified format foundation for automatic comparison and avoiding process delays caused by format incompatibility. The .2D file content contains two core elements: first, the work order identifier, used for file naming association; and second, the 2D code of all individual products under the entire work order, i.e., the unique content of the QR code. This provides a standard data source for subsequent machine identification and comparison, ensuring that the judgment criteria for each product are strongly bound to the production work order, fundamentally avoiding the problem of mixed judgment across work orders and ensuring the accuracy of traceability data.
[0036] In one specific embodiment of this application, the equipment modification and whitelist configuration module 3 is used to modify the hardware of the LS device, specifically including adding a barcode scanner and a communication module. A brand-name industrial barcode scanner is installed on the LS device, communicating with the device's main control board via an I / O interface.
[0037] In one specific embodiment of this application, the device modification and whitelist configuration module 3 is used to upgrade the software of the LS device, specifically including optimizing the identification algorithm and data interaction logic. The device firmware is optimized by adding a "whitelist verification" function module, supporting encryption algorithms such as AES decryption of whitelist files. AES is a block cipher with a fixed plaintext length of 128 bits (bits). 1 bit equals 1 byte, so 128 bits equals 16 bytes. The key length can be 128, 192, or 256 bits. When the key is 128 bits, it requires 10 rounds of encryption. For every 64 bits increase in the key, 2 more rounds are needed: 192 bits require 12 rounds, and 256 bits require 14 rounds. AES is a widely used symmetric block cipher algorithm, where encryption and decryption use the same key. It is primarily used to protect data confidentiality, and decryption is the reverse process of encryption. AES decryption is resistant to known attacks, such as differential / linear cryptanalysis, and has not been substantially broken to date. Hardware acceleration is highly efficient, making it suitable for resource-constrained devices.
[0038] In one specific embodiment of this application, the device modification and whitelist configuration module 3 is used for whitelist configuration: when the system boots up, it automatically executes a command, such as C:\huatian_2d\download.bat, to download the latest whitelist from the server to the local cache.
[0039] In one specific embodiment of this application, the comparison and sorting module 4 uses the Vision system to call an image recognition algorithm, such as an improved version of OpenCV, to recognize the QR code, extract the unitID, and compare it line by line with a local whitelist. Products that fail the comparison trigger device I / O signals, controlling the sorting mechanism to push the product into the defective product tray. OpenCV is a cross-platform computer vision and machine learning software library released under the Apache 2.0 license, which can run on Linux, Windows, Android, and Mac OS operating systems. It is lightweight and efficient—consisting of a series of C functions and a small number of C++ classes, while also providing interfaces for languages such as Python, Ruby, and MATLAB, implementing many general-purpose algorithms in image processing and computer vision. OpenCV provides a rich set of visual processing algorithms, and since it is partly written in C, coupled with its open-source nature, it can be fully compiled and linked to generate an executable program without adding new external support, making it suitable for algorithm porting. With appropriate modifications, OpenCV code can run normally on DSP systems and ARM embedded systems. OpenCV is dedicated to real-world real-time applications. Its optimized C code significantly improves execution speed, and even faster processing speeds can be achieved by purchasing Intel's IPP high-performance multimedia function library.
[0040] In one specific embodiment of this application, for products automatically sorted to the defective product tray that do not match the whitelist, the reason for the comparison failure is marked for further review. If the comparison fails, a device I / O signal is triggered to control the sorting mechanism, such as a pneumatic pusher, to push the product into the defective product tray, while simultaneously recording the reason for the failure, such as "ID mismatch" or "QR code damage". Considering the small size and high batch characteristics of semiconductor products, the sorting algorithm adds a "Tray coordinate memory" function: the machine records the row and column coordinates of good and defective products in the tray in real time; during the sorting stage, the software drives the sorting mechanism through a customized algorithm to accurately guide good products into the good product track and defective products into the defective product track, avoiding classification errors caused by manual intervention.
[0041] In one specific embodiment of this application, the data traceability module 5 stores the entire product process data in a database. This data includes work order number, time, equipment parameters, and good / defective product status. A CIM module data interface is developed, storing the entire product process data in a database such as SQL Server, and associating it with work order number, equipment ID, timestamp, comparison results, etc. The data traceability module presents this data through database association and a visual interface. The front end uses visualization tools, such as Power BI, to build a query interface, supporting product status traceability by work order number and time range, and generating traceability reports. SQL Server is a relational database management system developed by Microsoft Corporation. This system provides data storage and transaction processing. Its core components include graphical tools such as SQL Server Management Studio. It uses a table structure to store data and implements access control through login names and roles. SQL Server supports deployment on Windows and Linux platforms and features Always On high availability, memory-resident technology, and transparent data encryption. Its high-performance design fully leverages the advantages of Windows NT. Advanced system management supports Windows graphical management tools, local and remote system management and configuration. Robust transaction processing capabilities ensure data integrity through various methods. Supporting symmetric multiprocessor architecture, stored procedures, ODBC, and its own SQL language, SQL Server provides an outstanding database platform for users, developers, and system integrators with its built-in data replication capabilities, powerful management tools, tight integration with the Internet, and open system architecture.
[0042] In one specific embodiment of this application, based on the PWS production execution system, each product entering and leaving the station must be verified through the system. At key nodes, the "print station exit" interface is automatically invoked to synchronize the .2D file to the FTP, ensuring that data can be directly accessed when the product flows to the LS station, avoiding traceability gaps. Using our proprietary 2D system, after the printing station prints a unique QR code for each product, the 2D file of the entire work order is automatically stored in the system, forming a closed-loop management of "production-storage-access," ensuring data is searchable and traceable. After the LS station completes comparison, sorting, and batch closing, the machine automatically generates a processing report; the CIM system periodically reads and captures the report content, summarizes it into a standard format report, and embeds it into the CIM system. After entering query conditions, users can intuitively view full-dimensional information such as equipment number, date, customer code, and packaging form, achieving "one-click traceability."
[0043] This invention significantly improves automation and efficiency, eliminating the need for manual intervention throughout the entire process. From QR code recognition to defective product sorting, time is reduced, increasing production efficiency by more than 10%. Quality control is precise, with real-time comparison against the whitelist, reducing the rate of missed defects and false positives to below 0%, effectively intercepting risky products. A closed-loop data traceability system integrates full-process data through the CIM module, supporting rapid querying and traceability, aiding in production problem analysis and continuous improvement. Batch processing replaces manual labor: For batches of tens of thousands of semiconductor products, each with a unique QR code, manual labor is neither efficient in recognizing QR code content nor effective in preventing mixed-up or incorrect interception. The Vision system can complete sorting and unloading within hours, completely solving the efficiency and accuracy pain points of manual operation. Targeted data management: The whitelist only stores information on 15 customer codes (XXX homogeneous customers), allowing for subsequent additions as needed, avoiding invalid data consuming storage resources. Simultaneously, the .2D file is automatically associated with work orders and uploaded to FTP after generation, requiring no manual intervention throughout the process. High-definition recognition assurance: By illuminating the product surface with a combination of 11 light sources, interference such as QR code reflection and blurring is eliminated, ensuring that the industrial camera clearly captures 1-20 character information, raising the comparison accuracy to an industry-leading level. Secondary verification of defective products: After automatic sorting, the machine retains high-definition images of defective products. Workers can quickly locate the defective product and perform a secondary review using these images, reducing the false positive rate and further improving production efficiency.
[0044] The above are merely embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.
Claims
1. A product QR code recognition and data traceability system, characterized in that, include: The print triggering and information generation module triggers the printer in conjunction with the production control system by calling the PWS interface, generating a 2D information file for a single product containing a unique and identifiable unit ID. The IT data capture and file transfer module is used to capture 2D information files of a single product containing unit ID data and coplanarity Copl parameter information files from the production process and upload them to a preset server path. The data traceability module is used to automatically aggregate the 2D information file and coplanarity Copl parameter information file of a single product into the CIM module database, forming a database of multiple products, so that the information of each product in the entire process of mass production can be traced.
2. The product QR code recognition and data traceability system according to claim 1, characterized in that, The IT data capture and file transfer module captures a 2D information file containing unit ID data for a single product from the production process. The unit ID data is named with the work order number and is in the format of a .2D file that stores a single QR code information per line.
3. The product QR code recognition and data traceability system according to claim 1, characterized in that, The IT data capture and file transfer module uploads the 2D information file of the single product and the Coplanarity Copl parameter information file corresponding to the single product to the following server path via FTP protocol: http: / / 192.162.9.8.228 / Leadscan / LS-2DMatrix / Cop / Mark-flder / SZ228 / file / preset server path.
4. The product QR code recognition and data traceability system according to claim 1, characterized in that, The IT data capture and file transfer module needs to capture the product's 2D information file and coplanarity Copl parameter information file at each station when a single product enters or leaves the station, and synchronize them to the preset server path to ensure that the data can be directly accessed when a single product is transferred to the next station, avoiding data gaps in traceability.
5. A product QR code recognition and data traceability system according to claim 1 or 3, characterized in that, The IT data capture and file transfer module calculates the captured processing data in real time through software, automatically summarizes and generates processing reports, and automatically uploads them to FTP software. The reports are then synchronously uploaded to a fixed path on the cloud drive to prevent data loss or tampering.
6. The product QR code recognition and data traceability system according to claim 1, characterized in that, The 2D information file and the coplanarity Copl parameter information file specifically include: machine number, processing date, customer code, packaging type, work order, single product serial number, Matrix code, and Copl parameters.
7. The product QR code recognition and data traceability system according to claim 1, characterized in that, The data traceability module reads reports from the FTP client via scheduled tasks, captures 2D information and Copy data in fixed rows and columns, summarizes them into an Excel report in a standard format, and embeds it into the web interface of the automated CIM software. Users can accurately query data by equipment number, processing date, customer code, work order, and other criteria.
8. The product QR code recognition and data traceability system according to claim 1, characterized in that, Also includes: The device modification and whitelist configuration module is used for hardware modification and software upgrade of devices, configuring QR code whitelists and caching them locally; among them, By illuminating the product surface with a combination of 11 light sources, interference such as QR code reflection and blurring is eliminated, ensuring that the industrial camera can clearly capture 1-20 character information. The whitelist stores information for only homogeneous customers, totaling 15 customer codes. Additional customers can be added as needed to avoid invalid data consuming storage resources.
9. A product QR code recognition and data traceability system according to claim 1 or 8, characterized in that, Also includes The comparison and sorting module automatically sorts products by recognizing the unit QR code content, extracting the unit ID, and comparing it with a whitelist; among which, In response to the characteristics of small size and high batch size of semiconductor products, the sorting algorithm has added a tray coordinate memory function: the machine will record the row and column coordinates of good and defective products in the tray in real time; during the sorting stage, the software drives the sorting mechanism through a customized algorithm to accurately guide good products into the good product track and defective products into the defective product track, avoiding classification errors caused by manual intervention.
10. A product QR code recognition and data traceability system according to claim 2, characterized in that, The 2D information file contains two main parts: first, the work order identifier, used for file naming and association; and second, the 2D code of all individual products under the entire work order, i.e., the unique content of the QR code, which provides a standard data source for subsequent machine identification and comparison, ensuring that the judgment criteria of each product are strongly bound to the production work order, avoiding cross-work order mixed judgment, and ensuring the accuracy of traceability data.