Methods, apparatus, electronic equipment and storage media for online hollow wafer production.

By using artificial intelligence technology to cluster and verify the inventory of hollow wafers, the production scheduling tasks are automatically generated, which solves the problems of time delay and low efficiency caused by manual operation in the existing technology and realizes a highly efficient production scheduling process for hollow wafers.

CN123070549APending Publication Date: 2026-11-10ZHAOQING CSG ENERGY SAVING GLASS CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611185171.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-05
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

The existing hollow wafer production scheduling method relies on manual operation, which leads to time delays, low efficiency, frequent errors and omissions in orders, low utilization of raw wafers, and poor inventory communication, making it difficult to meet high-standard project delivery requirements.

Method used

The system uses artificial intelligence technology to acquire a set of sheet-to-be-processed tasks, clusters tasks based on glass process attribute parameters to generate sheet-to-be-processed task clusters, performs original sheet demand prediction and independent layout optimization, and combines inventory verification information to process sheet-to-be-processed production scheduling, thus automatically generating production scheduling tasks.

Benefits of technology

It enables the batch collection and merging of overdue patching tasks, avoiding the time lag and missed orders caused by manual operation, ensuring the sufficiency of raw materials, shortening the delivery cycle, and improving the accuracy and efficiency of patching operations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123070549A_ABST
    Figure CN123070549A_ABST
Patent Text Reader

Abstract

This application provides a method, apparatus, electronic device, and storage medium for online production scheduling of hollow wafers, belonging to the field of glass processing technology. The method includes: acquiring a set of wafer-gathering tasks to be processed, the set containing multiple wafer-gathering tasks, each task including glass process attribute parameters; clustering the wafer-gathering tasks based on the glass process attribute parameters to obtain wafer-gathering task clusters; predicting raw wafer demand based on the wafer-gathering task clusters and the glass process attribute parameters to obtain raw wafer specification demand information; optimizing the layout of the wafer-gathering task clusters independently based on the raw wafer specification demand information to obtain target wafer-gathering production layouts; verifying inventory based on the raw wafer specification demand information to obtain inventory satisfaction information; and performing wafer-gathering scheduling processing on the wafer-gathering task clusters based on the inventory satisfaction information and the target wafer-gathering production layouts to obtain wafer-gathering scheduling tasks. This application embodiment can improve the efficiency of hollow wafer production scheduling.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of glass processing technology, and in particular to a method, apparatus, electronic device and storage medium for online production of hollow sheet glass. Background Technology

[0002] The integrated insulated glass production scheduling system intelligently aggregates, optimizes layouts, and distributes production schedules for damaged glass replacement and scattered orders during the glass deep processing production line. This achieves efficient transformation from the intention to replace damaged glass to production line processing, avoiding the waste of capacity and time caused by cumbersome communication and repeated follow-ups in traditional manual scheduling processes. Taking the production line of large architectural curtain wall glass as an example, integrated insulated glass production scheduling can quickly complete the entire chain of operations, from merging overdue replacement tasks across multiple production lines and accurately predicting original glass specifications to independently optimizing physical layouts. This provides intelligent and automated scheduling support for high-standard project delivery dates, significantly reducing the cost and error rate of unplanned manual scheduling for glass processing companies.

[0003] Currently, the common method for scheduling hollow glass production involves workshop employees reporting broken glass to the system upon discovery. Operators then manually select pending orders in the system backend. Once the orders enter the replacement task pool, manual review is used to re-optimize the arrangement of smaller pieces of the same specifications and thickness onto a single large sheet of original glass. The materials control department is then notified offline for original sheet allocation, and finally, a production schedule is issued. However, this method is time-consuming and labor-intensive, with significant time lags in manually selecting, matching, and re-optimizing broken glass in the backend, resulting in long scheduling cycles. Furthermore, manual cross-departmental communication regarding the status of original sheets is not only inefficient but also prone to human error leading to missed replacements, preventing timely transfer of replacement tasks to the production line. Therefore, improving the efficiency of hollow glass production scheduling has become a pressing technical problem. Summary of the Invention

[0004] The main objective of this application is to provide a method, apparatus, electronic device, and storage medium for connecting hollow wafers for production, aiming to improve the efficiency of hollow wafer production.

[0005] To achieve the above objectives, a first aspect of this application proposes a method for scheduling hollow wafers in a production line, the method comprising: Obtain a set of unprocessed glass task pieces, wherein the set of unprocessed glass task pieces contains multiple unprocessed glass task pieces, and each unprocessed glass task piece contains glass process attribute parameters. Based on the glass process attribute parameters, the set of unprocessed wafer tasks is clustered to obtain wafer task clusters. Based on the aforementioned wafer task cluster and the aforementioned glass process attribute parameters, the raw wafer demand is predicted to obtain raw wafer specification demand information. Based on the original film specification requirements, the layout of the film cluster is optimized independently to obtain the target film production layout. Based on the original film specification requirements information, inventory verification is performed to obtain inventory verification information, wherein the inventory verification information includes inventory satisfaction information; Based on the inventory satisfaction information and the target production layout, the production scheduling of the production task cluster is performed to obtain the production scheduling task.

[0006] In some embodiments, the step of predicting raw wafer demand based on the wafer cluster and the glass process attribute parameters to obtain raw wafer specification demand information includes: Based on the aforementioned glass processing task cluster, the glass processing attribute parameters are filtered to obtain the target processing attribute parameters. Geometric dimension analysis is performed on the target process attribute parameters to obtain the task dimension distribution set; Based on the set of task size distributions, the processing surface is accumulated and estimated to obtain the predicted page area requirement. The predicted plate area requirement is combined with the target process attribute parameters to obtain the original plate requirement feature vector. Based on the original film requirement feature vector, the specification optimization and matching are performed in the preset standard original film specification library to obtain the original film specification requirement information.

[0007] In some embodiments, the step of performing production scheduling on the batch of batching tasks based on the inventory satisfaction information and the target batching production layout to obtain batching production tasks includes: The inventory information is analyzed to obtain the original shelf piece distribution data. Based on the original sheet number distribution data of the material rack, the material rack is screened to obtain the target material rack; Based on the target production layout and the production task cluster, the production scheduling data is reconstructed to obtain the production processing dataset. Based on the pre-cut processing dataset and the target material rack, the production task is encapsulated to obtain the pre-cut production scheduling task.

[0008] In some embodiments, the step of performing inventory verification based on the original wafer specification requirement information to obtain inventory verification information includes: Obtain the attribute data of the original film in stock; The specification requirements of the original film are matched and compared with the attribute data of the original film in stock to obtain the comparison and verification results; Based on the comparison and verification results, verification information is generated to obtain the inventory verification information. If the comparison and verification results indicate that the original film specification requirement information and the inventory original film attribute data are successfully matched, the inventory verification information includes the inventory satisfaction information.

[0009] In some embodiments, the cluster of homogeneous tasks includes multiple homogeneous tasks with the same attributes; The step of optimizing the layout of the whole-piece task cluster independently based on the original film specification requirements information to obtain the target whole-piece production layout includes: Size features are extracted from each of the homogeneous tasks with the same attributes to obtain a set of task geometric dimensions; Based on the set of task geometric dimensions and the original film specification requirements, the layout array is calculated to obtain the initial layout array; The physical layout is reconstructed based on the initial arrangement array to obtain the target uniform production layout.

[0010] In some embodiments, the step of clustering the set of wafer-gluing tasks to be processed based on the glass process property parameters to obtain wafer-gluing task clusters includes: Feature extraction is performed on the glass process property parameters to obtain a glass identification feature vector; Based on the glass identification feature vector, the set of uniform glass tasks to be processed is homogeneously distinguished to obtain uniform glass tasks with the same attributes. The tasks with the same attributes are grouped together to obtain the cluster of tasks.

[0011] In some embodiments, obtaining the set of uniform tasks to be processed includes: Obtain a set of patch tasks to be processed, wherein the set of patch tasks to be processed contains multiple patch tasks to be processed; The duration of each pending patch task is queried to obtain multiple task rolling duration data. The rolling duration data of each task is compared with the preset dynamic patch timeout threshold to obtain duration comparison information, wherein the duration comparison information includes task timeout information. Based on the task timeout information, the set of pending patch tasks is filtered to obtain the set of pending complete patch tasks.

[0012] To achieve the above objectives, a second aspect of this application provides a connecting hollow wafer assembly and dispatching device, the device comprising: The task acquisition module is used to acquire a set of unprocessed glass task pieces, wherein the set of unprocessed glass task pieces contains multiple unprocessed glass task pieces, and the unprocessed glass task pieces contain glass process attribute parameters. The task clustering module is used to cluster the set of unprocessed wafer tasks based on the glass process attribute parameters to obtain wafer task clusters. The raw glass prediction module is used to predict the raw glass demand based on the glass assembly task cluster and the glass process attribute parameters, and obtain raw glass specification demand information. The layout optimization module is used to perform independent layout optimization on the whole-piece task cluster based on the original piece specification requirements information to obtain the target whole-piece production layout. The inventory verification module is used to perform inventory verification based on the original film specification requirement information to obtain inventory verification information, wherein the inventory verification information includes inventory satisfaction information. The production scheduling module is used to perform production scheduling on the production task cluster based on the inventory satisfaction information and the target production layout, so as to obtain the production scheduling task.

[0013] To achieve the above objectives, a third aspect of the present application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method of the first aspect described above.

[0014] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method of the first aspect described above.

[0015] The proposed method, apparatus, electronic device, and storage medium for online hollow wafer assembly scheduling first acquire a set of wafer assembly tasks containing multiple tasks to be processed. Then, based on the glass process attribute parameters of each task, task clustering is performed. This clustering aggregates overdue refill tasks with the same or compatible attributes into wafer assembly task clusters, achieving batch collection and merging of cross-production line overdue refill tasks. This lays a data foundation for intensive scheduling and effectively avoids the problems of missed orders and low efficiency associated with traditional manual order selection. Next, based on the wafer assembly task cluster and glass process attribute parameters, raw material demand is predicted. The required raw material specifications can be pre-calculated based on the size distribution and process characteristics of each task after clustering, providing accurate specification matching for inventory verification and layout optimization, avoiding invalid optimization calculations due to specification uncertainty. Subsequently, based on the raw material specification demand information, the wafer assembly task cluster is independently optimized, generating a completely new layout for the merged task cluster. The production layout does not rely on real-time matching or insertion with the current production task, ensuring that overdue backlog tasks can be arranged independently and efficiently. This overcomes the shortcomings of traditional manual optimization, such as time lag and lack of specificity. Based on the original sheet specification requirements, inventory verification is performed. When the verification result includes inventory satisfaction information, the sufficiency of original sheet resources can be automatically confirmed. This effectively ensures that raw materials are available before the production scheduling instruction is issued, preventing production interruptions caused by original sheet shortages. Finally, based on the inventory satisfaction information and the target sheet production layout, sheet production scheduling is performed to generate sheet production scheduling tasks. The optimized layout data, selected original sheet rack information, and order information are comprehensively packaged and directly sent to the production line for execution. This significantly reduces manual intervention, shortens the sheet delivery cycle, and improves the accuracy of patching operations, thereby effectively improving the efficiency of hollow sheet production scheduling. It also solves the technical problems of patching relying on manual intervention and time lag leading to delivery delays in existing technologies. Attached Figure Description

[0016] Figure 1 This is a flowchart of the method for scheduling hollow wafers in connection according to an embodiment of this application; Figure 2 yes Figure 1 The flowchart of step S101 in the text; Figure 3 yes Figure 1 The flowchart of step S102 in the document; Figure 4 yes Figure 1 The flowchart of step S103 in the process; Figure 5 yes Figure 1 The flowchart of step S105 in the process; Figure 6 yes Figure 1The flowchart of step S106 in the process; Figure 7 yes Figure 1 The flowchart of step S106 in the process; Figure 8 This is a schematic diagram of the structure of the inline hollow wafer assembly and discharging device provided in the embodiments of this application; Figure 9 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0018] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0020] First, let's analyze some of the terms used in this application: Raw glass sheet: Raw glass sheet refers to a whole sheet of glass purchased by glass deep processing enterprises that has not undergone any cutting or processing. It has specific dimensions, thickness, and color attributes. In the glass deep processing production process, the raw glass sheet is at the beginning of the entire processing chain. Cutting equipment cuts the raw glass sheet into multiple small pieces of glass that meet the order requirements according to the optimized layout scheme. These small pieces of glass then enter subsequent processes such as edging, tempering, and insulated glass assembly, ultimately forming the insulated glass product delivered to the customer.

[0021] Raw glass storage: Raw glass storage refers to the physical storage area within the glass processing plant used to store raw glass sheets. Within this storage area, raw glass sheets are categorized by different specifications, sizes, thicknesses, and colors and placed on dedicated iron racks. Each rack typically holds multiple sheets of the same specification, and each rack has a fixed location within the storage area. When the production scheduling system completes the layout optimization for the replacement task and determines that raw sheets are needed for cutting, it needs to retrieve a sheet that meets the specifications from a specific rack in this raw glass storage area. Therefore, the production scheduling system needs to know the inventory distribution of each rack in this storage area in advance to determine whether there are available raw sheets and which rack to retrieve them from.

[0022] In-line insulated glass sheet matching scheduling refers to the process in glass deep-processing production lines where tasks requiring replacement sheets due to breakage or other reasons, as well as scattered orders, are aggregated, optimized, and then distributed to the production line for execution. This ensures that all required glass sheets for a single order are complete. In the in-line production of large-scale architectural curtain wall glass, an order often contains dozens to hundreds of glass sheets of different specifications. If any sheet breaks during production, it must be replaced promptly to ensure that all glass sheets for the order are complete before entering the insulated glass assembly process. Therefore, the efficiency of insulated glass sheet matching scheduling directly impacts the on-time delivery rate of orders.

[0023] Currently, the common production scheduling method for hollow glass typically employs a manual operation process. After workshop employees discover broken glass, they report it to the system. Operators then manually search for and select orders requiring replacement sheets in the system backend. Once the orders enter the replacement task pool, operators rely on their experience to meticulously identify each task, selecting multiple replacement sheets of the same specifications and thickness. These smaller sheets are then rearranged onto a single large sheet using manual drawing or basic layout software. The material control department is then notified offline via phone or instant messaging to confirm the availability of matching sheets in the warehouse. Only after confirmation of availability is the production scheduling instruction issued to the production line. However, this method has several problems in actual operation.

[0024] First, the manual selection and subjective verification of each item is time-consuming. When there are dozens of patch tasks from different production lines and orders in the task pool, operators need to check the specifications, thickness, color, and process information of each task one by one. This can easily lead to incorrect or missed selections due to fatigue or negligence, resulting in patch tasks that should be merged for production not being aggregated in time, causing multiple separate production schedules. This wastes raw materials and occupies production line capacity. Second, the efficiency of manual layout optimization heavily depends on the operator's personal experience. For multiple patch tasks with various sizes and specifications, it is difficult for humans to calculate the optimal layout in a short time. Often, they can only rely on experience to make a rough layout, resulting in low raw material utilization and serious material waste. Secondly, manual cross-departmental communication regarding the status of original film inventory is inefficient and involves lengthy information transmission chains. After operators complete production scheduling instructions, they need to contact material control personnel by phone or instant messaging tools to confirm whether there are original films of the correct specifications in the warehouse. This communication process involves waiting time, and the communication content may be unclear or incomplete, leading to errors in material retrieval and thus delaying the execution of replacement tasks. More importantly, in the entire process of relying on manual order selection, matching, optimization, and communication, replacement tasks undergo multiple queuing waits from entering the replacement task pool to the final issuance of production scheduling instructions. Each wait incurs time loss, resulting in the issuance cycle of production scheduling instructions being much longer than the actual required time. Many replacement tasks are often only discovered to be incomplete when the shipment is almost due, at which point it is too late to urgently replace the missing pieces, directly causing order delays and exposing the company to the risk of default. In addition, when operators are absent due to leave or job transfer, the processing of replacement tasks will be forced to stop, further exacerbating the time lag problem.

[0025] In summary, the existing hollow wafer production scheduling method suffers from problems such as cumbersome operation, low efficiency, frequent errors and omissions in order processing, low utilization rate of original wafers, poor inventory communication, and serious time delays due to its reliance on manual order selection and matching, manual layout optimization, and manual cross-departmental communication. This results in the inability to promptly transfer replacement tasks to the production line for processing, ultimately leading to low efficiency in hollow wafer production scheduling and making it difficult to meet high-standard project delivery requirements. Therefore, improving the efficiency of hollow wafer production scheduling has become an urgent technical problem to be solved.

[0026] Based on this, embodiments of this application provide a method and apparatus for connecting hollow wafers for production, an electronic device, and a storage medium, aiming to improve the efficiency of hollow wafer production.

[0027] The connection hollow wafer scheduling method, apparatus, electronic device, and storage medium provided in this application are specifically described through the following embodiments. First, the connection hollow wafer scheduling method in this application is described.

[0028] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0029] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0030] The method for aligning and scheduling hollow glass sheets provided in this application relates to the field of glass processing technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the method for aligning and scheduling hollow glass sheets, but is not limited to the above forms.

[0031] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0032] Figure 1This is an optional flowchart of the hollow wafer alignment scheduling method provided in the embodiments of this application, wherein the hollow wafer alignment scheduling method can be applied to a wafer alignment scheduling system. Figure 1 The method may include, but is not limited to, steps S101 to S106.

[0033] Step S101: Obtain the set of blanking tasks to be processed, wherein the set of blanking tasks to be processed contains multiple blanking tasks to be processed, and the blanking tasks to be processed contain glass process attribute parameters. Step S102: Based on the glass process attribute parameters, perform task clustering on the set of whole-wafer tasks to be processed to obtain whole-wafer task clusters; Step S103: Based on the wafer task cluster and glass process attribute parameters, predict the raw wafer demand to obtain raw wafer specification demand information. Step S104: Based on the original film specification requirements, perform independent layout optimization on the aligned film task cluster to obtain the target aligned film production layout; Step S105: Based on the original film specification requirements information, perform inventory verification to obtain inventory verification information, which includes inventory fulfillment information. Step S106: Based on the inventory satisfaction information and the target production layout, perform production scheduling on the production task cluster to obtain the production scheduling task.

[0034] Steps S101 to S106 as illustrated in this embodiment first involve acquiring a set of unprocessed wafer-gathering tasks containing multiple unprocessed wafer-gathering tasks. Then, based on the glass process attribute parameters of each task, task clustering is performed. This clustering aggregates overdue wafer-gathering tasks with the same or compatible attributes into wafer-gathering task clusters, achieving batch collection and merging of cross-production line overdue wafer-gathering tasks. This lays a data foundation for intensive production scheduling and effectively avoids the problems of missed orders and low efficiency associated with traditional manual order selection. Next, based on the wafer-gathering task cluster and glass process attribute parameters, raw wafer demand is predicted. The required raw wafer specification information can be pre-calculated based on the size distribution and process characteristics of each clustered task, providing accurate specification matching for inventory verification and layout optimization. This avoids invalid optimization calculations due to specification uncertainty. Subsequently, based on the raw wafer specification requirement information, the wafer-gathering task cluster is independently optimized, generating a completely new production layout for the merged task cluster. This new technology, independent of real-time matching or insertion with current production tasks, ensures that backlogged tasks can complete layout independently and efficiently. It overcomes the shortcomings of traditional manual optimization, such as time lag and lack of specificity. Based on the original sheet specifications and requirements, inventory verification is performed. When the verification result includes inventory availability information, the sufficiency of original sheet resources can be automatically confirmed. This effectively ensures that raw materials are available before the production scheduling instruction is issued, preventing production interruptions due to a shortage of original sheets. Finally, based on the inventory availability information and the target sheet production layout, sheet production scheduling is performed to generate sheet production scheduling tasks. The optimized layout data, selected original sheet rack information, and order information are comprehensively packaged and directly sent to the production line for execution. This significantly reduces manual intervention, shortens the sheet delivery cycle, and improves the accuracy of patching operations. As a result, it effectively improves the efficiency of hollow sheet production scheduling and solves the technical problems of patching relying on manual intervention and time lag leading to delivery delays in existing technologies.

[0035] In some application scenarios, taking the production line of a large building curtain wall glass order as an example, the order requires delivery within 72 hours. However, during the production process, glass breakage occurred continuously on multiple production lines, resulting in dozens of pending replacement tasks accumulating in the dynamic replacement pool. Some of these replacement tasks have been waiting for more than 48 hours without being matched with a suitable production schedule. If these overdue replacement tasks continue to roll in the dynamic replacement pool, it will seriously threaten the timely delivery of the order. After detecting the overdue tasks in the dynamic replacement pool, the production scheduling system retrieves all pending replacement tasks from the pool, queries the duration of each task, obtains the time spent rolling for each task, and then compares this data with the system's preset dynamic replacement timeliness threshold. The overdue replacement tasks are then filtered out from the pending replacement task set to form a set of pending replacement tasks. Next, the glass assembly scheduling system extracts features from the glass process attribute parameters of each glass assembly task to obtain the glass identification feature vector of each task. Tasks with identical glass specifications, thickness, color and process type in the glass identification feature vector are classified into the same category. The classified glass assembly tasks with the same attributes are then grouped together to generate multiple glass assembly task clusters. Each glass assembly task cluster consists of patching tasks with identical glass process attributes. Furthermore, the sheet-aligning scheduling system takes each sheet-aligning task cluster as a unit, and selects the specific length, width, and thickness data of each patch task from the glass process attribute parameters corresponding to each sheet-aligning task cluster. It then obtains the task size distribution set of each cluster through geometric dimension analysis. Based on the task size distribution set of each cluster, it accumulates and estimates the processing surface. On the basis of accumulating the area of ​​each small piece, it adds the layout gap loss to obtain the predicted layout area requirement of each cluster. Then, it combines the predicted layout area requirement with thickness and color information to form the original sheet requirement feature vector. It searches in the preset standard original sheet specification library for the original sheet specification that can meet the area requirement and minimize material waste, thus obtaining the original sheet specification requirement information corresponding to each sheet-aligning task cluster. Next, the glass sheet scheduling system extracts the size features of multiple glass sheet tasks with the same attributes in each cluster based on the original sheet specification requirements. This yields a set of task geometric dimensions. Under the boundary constraints determined by the standard original sheet specifications, the system performs layout array calculations on each small glass sheet to generate an initial layout array. The system then performs physical layout reconstruction on the initial layout array, adding actual process constraints such as cutting kerf width, edge grinding allowance, and glass edge distance to the initial layout array. This generates target glass sheet production layouts corresponding to each glass sheet task cluster. Each target glass sheet production layout contains the final position coordinates, cutting path, and cutting sequence data of each small glass sheet on the original sheet.Furthermore, the production scheduling system obtains real-time inventory raw material attribute data from the raw material warehouse, which includes information such as the specifications, thickness, color, remaining number of raw materials on each shelf, and shelf location. It then performs specification matching and comparison between the raw material specification requirements of each production scheduling task cluster and the real-time inventory raw material attribute data. When a match is successful, it generates inventory verification information containing inventory satisfaction information. This inventory satisfaction information carries the shelf number of the successfully matched shelf and the distribution data of the remaining number of raw materials on each shelf. Finally, assuming the inventory satisfies the information, the production scheduling system performs scheduling processing on each batch of batching tasks. This involves parsing the distribution of the number of original pieces on each rack to obtain the distribution data of the number of original pieces on each rack, determining the target rack according to the preset filtering rule of prioritizing the rack with the fewest remaining original pieces, and then associating and binding the cut pieces in each target batching production panel with the corresponding patching task orders to generate a batching processing dataset. The batching processing dataset is then packaged and integrated with the target rack number and task priority identifier to generate batching scheduling tasks corresponding to each batching task cluster. These multiple batching scheduling tasks are then sent to the connected production line to perform patching and cutting operations, achieving batch clearing of overdue patching tasks. Through the above-mentioned fully automated processing, the production scheduling system can quickly aggregate dozens of overdue replacement tasks backed up in the dynamic replacement pool into multiple production task clusters and perform independent layout optimization and inventory verification. This allows each overdue replacement task to complete its production scheduling and issuance independently without being constrained by the current production task layout, effectively avoiding further delays caused by waiting for real-time matching opportunities, thereby ensuring the timely delivery of large building curtain wall glass orders.

[0036] In step S101 of some embodiments, the set of pending piece-fitting tasks refers to a collection of multiple pending piece-fitting tasks that have entered the piece-fitting scheduling process after screening. A pending piece-fitting task refers to a piece-fitting task whose task rollover time has exceeded the dynamic patching timeliness threshold, thus meeting the precondition for entering the piece-fitting scheduling process. In the glass deep processing field, this precondition typically manifests as the patching task's waiting time in the dynamic patching pool exceeding a system-preset time threshold. In the scenario of batch processing of timed-out tasks in the glass deep processing field, when multiple timed-out patching tasks in the dynamic patching pool need to be processed centrally, the set of pending piece-fitting tasks is the collection of these timed-out patching tasks.

[0037] Glass processing attribute parameters refer to data items used to describe the inherent properties of glass in terms of physical dimensions and processing technology. It should be noted that glass processing attribute parameters include at least glass specifications, glass thickness, glass color, and processing type.

[0038] In this embodiment, the production scheduling system first retrieves all pending patch tasks from the dynamic patch pool. It then queries the duration of each pending patch task to obtain the accumulated scrolling time data of each task in the dynamic patch pool. The dynamic patch pool is a temporary data storage area used to temporarily store all pending patch tasks that have not yet completed patch processing. Next, the accumulated scrolling time data of each pending patch task is compared with a preset dynamic patch timeout threshold to determine if any task has timed out. If it has, the timed-out tasks are filtered out from the pending patch task set to form a set of pending patch tasks. As an optional scenario in this implementation, when there are no timed-out pending patch tasks, the production scheduling system can continue monitoring and wait for new pending patch tasks to enter the dynamic patch pool until a timed-out pending patch task is detected, triggering the filtering of the set of pending patch tasks.

[0039] For details, please refer to Figure 2 In some embodiments, step S101 may include, but is not limited to, steps S201 to S204: Step S201: Obtain the set of patch tasks to be processed, wherein the set of patch tasks to be processed contains multiple patch tasks to be processed. Step S202: Query the duration of each pending patch task to obtain multiple task rolling duration data; Step S203: Compare the rolling duration data of each task with the preset dynamic patch timeout threshold to obtain duration comparison information, wherein the duration comparison information includes task timeout information. Step S204: Based on the task timeout information, the task set to be processed is filtered to obtain the complete set of tasks to be processed.

[0040] In step S201 of some embodiments, the set of patch tasks to be processed refers to a data set consisting of multiple patch tasks to be processed. This data set is the original input data obtained by the splicing and scheduling system from the dynamic patch pool. For example, in the scenario of screening overdue tasks in the field of glass deep processing, when the splicing and scheduling system needs to screen overdue tasks from the dynamic patch pool, the set of patch tasks to be processed can be the set of all patch tasks currently existing in the dynamic patch pool.

[0041] A pending patch task refers to a patch task in the dynamic patch pool that has not yet been processed.

[0042] In this embodiment of the application, the production scheduling system can periodically obtain all pending patch tasks from the dynamic patch pool according to a preset time period, or it can trigger the acquisition operation each time a new patch task enters the dynamic patch pool. All the pending patch tasks obtained can constitute a set of pending patch tasks.

[0043] In step S202 of some embodiments, the task rolling duration data refers to the value of the time each pending patch task has stayed in the dynamic patch pool after querying the task duration. For example, in the timeout judgment scenario in the field of glass deep processing, the task rolling duration data can be 2.5 hours that have passed from the time a pending patch task enters the dynamic patch pool to the current time.

[0044] In this embodiment, the production scheduling system reads the timestamp of each pending patch task entering the dynamic patch pool, and calculates the difference between the current system time and each timestamp to obtain the cumulative time that each pending patch task has stayed in the dynamic patch pool, which is used as the task rolling time data corresponding to each pending patch task.

[0045] In step S203 of some embodiments, the dynamic patching timeliness threshold refers to a time threshold parameter preset by the production scheduling system to determine whether a patching task to be processed has timed out. For example, in the timeout judgment scenario in the glass deep processing field, when the system administrator sets the timeout judgment criteria based on the urgency of the order and the production line load, the dynamic patching timeliness threshold can be set to 48 hours.

[0046] Duration comparison information refers to the comparison results output after duration comparison, which includes whether each pending patch task has timed out. For example, in the timeout judgment scenario in the glass deep processing field, this duration comparison information can be a data record containing the task identifier of each pending patch task and the corresponding timeout or non-timeout marker.

[0047] Task timeout information refers to the identification data in the duration comparison information used to indicate whether each pending patch task has exceeded the dynamic patch timeout threshold.

[0048] In this embodiment, the production scheduling system determines whether the task rolling duration data of each pending patch task is greater than the dynamic patch timeliness threshold, thereby obtaining duration comparison information containing task timeout information corresponding to each pending patch task. Specifically, if the task rolling duration data is greater than the dynamic patch timeliness threshold, the duration comparison result of the pending patch task is that it has timed out, and the corresponding duration comparison information is task timeout information. If the task rolling duration data is less than or equal to the dynamic patch timeliness threshold, the duration comparison result of the pending patch task is that it has not timed out, and the corresponding duration comparison information is task not timed out information.

[0049] In step S204 of some embodiments, after the production scheduling system traverses each pending patch task in the pending patch task set, it can extract pending patch tasks marked as expired according to the task timeout information, and keep pending patch tasks marked as not expired in the dynamic patch pool. Then, all the pending patch tasks that have expired extracted by the production scheduling system can jointly constitute the pending patch task set.

[0050] Steps S201 to S204 of this embodiment involve obtaining a set of pending patch tasks from the dynamic patch pool, containing multiple pending patch tasks, and querying the task duration for each pending patch task to obtain the corresponding task rolling duration data. This allows for real-time and accurate monitoring of the waiting time of each pending patch task in the dynamic patch pool, providing a precise time data basis for subsequent timeout judgment. Then, by comparing the rolling duration data of each task with a preset dynamic patch validity threshold, time comparison information including task timeout information is obtained. This enables automatic identification of whether each pending patch task has exceeded the allowed waiting time using a unified and objective time standard, avoiding subjective bias and judgment delays that may occur with manual judgment. Finally, based on the task timeout information, the pending patch tasks are processed... The processing of patch task sets involves task filtering. Timed-out patch tasks are extracted from the dynamic patch pool to form a patch set for subsequent patch scheduling. Non-timed-out patch tasks remain in the dynamic patch pool awaiting other matching opportunities. This process accurately identifies timed-out tasks that require batch processing while ensuring that non-timed-out tasks are not prematurely included in the batch merging process, thus disrupting their original dynamic matching opportunities. Furthermore, since the patch scheduling system automatically completes the filtering process for patch sets based on time thresholds, there is no need for manual review of the waiting status of each patch task. This avoids time delays and errors caused by manual judgment, enabling timed-out patch tasks to be filtered out promptly and accurately and enter the subsequent clustering and scheduling processes. This effectively improves the efficiency of hollow patch scheduling.

[0051] In step S102 of some embodiments, the wafer task cluster refers to the task grouping formed after the task clustering operation is completed. Each wafer task contained in the group has the same or compatible glass process attribute parameters and can be jointly laid out on the same original wafer board.

[0052] In this embodiment, the sheet-forming scheduling system first extracts features from the glass process attribute parameters of each sheet-forming task to obtain a glass identification feature vector for identifying the glass attributes of each task. This glass identification feature vector contains information such as glass specifications, thickness, color, and process type. Then, based on the glass identification feature vector of each sheet-forming task, the sheet-forming scheduling system distinguishes the sheet-forming task set to be processed, obtaining the classification results of sheet-forming tasks with the same attributes. Finally, the sheet-forming scheduling system groups the sheet-forming tasks with the same attributes obtained from the classification to obtain a sheet-forming task cluster.

[0053] For details, please refer to Figure 3 In some embodiments, step S102 may include, but is not limited to, steps S301 to S303: Step S301: Extract features from the glass process property parameters to obtain the glass identification feature vector; Step S302: Based on the glass identification feature vector, homogeneous distinction is made in the set of homogeneous tasks to be processed to obtain homogeneous tasks with the same attributes. Step S303: Group tasks with the same attributes together to obtain clustered task groups.

[0054] In step S301 of some embodiments, the glass identification feature vector refers to multi-dimensional vector data used to identify the glass attributes of each piece-fitting task to be processed. For example, in a task clustering scenario in the field of glass deep processing, when the piece-fitting scheduling system needs to determine whether multiple piece-fitting tasks to be processed meet the conditions for merging and layout, the glass identification feature vector can be a multi-dimensional vector composed of the glass specifications, thickness, color, and process type of a piece-fitting task, wherein the glass specifications correspond to the first dimension of the vector, the thickness corresponds to the second dimension, the color corresponds to the third dimension, and the process type corresponds to the fourth dimension.

[0055] In this embodiment, the glass assembly scheduling system first extracts attribute information of four dimensions—glass specifications, glass thickness, glass color, and process type—from the task data of each glass assembly task to be processed. Then, it combines the attribute information of the above four dimensions into a data structure in the form of a multi-dimensional vector to obtain a glass identification feature vector used to identify the glass attributes of each task. This glass identification feature vector can be used as the basis for comparing the similarity between tasks in the subsequent homogeneity differentiation operation.

[0056] In step S302 of some embodiments, the same attribute piece-fitting task refers to multiple piece-fitting tasks to be processed that have been classified into the same category after homogeneity differentiation. For example, in the task aggregation scenario in the field of glass deep processing, after the piece-fitting scheduling system completes homogeneity differentiation, the same attribute piece-fitting task can be multiple overdue piece-fitting tasks where the glass specifications are all 2440 mm x 3660 mm, the thickness is all 5 mm, the color is all clear glass, and the process type is all tempered.

[0057] In this embodiment, the glass alignment scheduling system compares the glass identification feature vectors of each glass alignment task to be processed pairwise. Tasks with identical values ​​in all four dimensions of the glass identification feature vector or with process compatibility are grouped into the same category, thus obtaining the classification result for glass alignment tasks with the same attributes. As an optional scenario in the implementation of this application, when two glass alignment tasks of tempered and semi-tempered processes can be merged for layout due to process compatibility, the glass alignment scheduling system can group two tasks whose glass identification feature vectors differ only in the process type dimension, but which falls within the compatibility range, into the same category.

[0058] In step S303 of some embodiments, the production scheduling system can obtain a batch of batch tasks by merging and grouping multiple batch tasks with the same attributes belonging to the same category into a task group.

[0059] Steps S301 to S303, as illustrated in this embodiment, involve using a wafer-gathering scheduling system to extract features from the glass process attribute parameters of each wafer-gathering task to obtain glass identification feature vectors that identify the glass attributes of each task. This quantifies the attribute information of each wafer-gathering task in four dimensions—glass specifications, thickness, color, and process type—into multi-dimensional vector data that can be processed by a computer, providing a standardized data foundation for subsequent automated clustering. Then, based on the glass identification feature vectors corresponding to each task, homogeneity differentiation is performed on the set of wafer-gathering tasks to be processed. This automatically and objectively identifies tasks with identical glass identification feature vectors or process compatibility through vector similarity comparison and assigns them to appropriate groups. By categorizing tasks within the same type, potential classification errors and judgment biases arising from subjective human identification are avoided. Ultimately, tasks with the same attributes obtained after homogeneous differentiation are clustered to generate clustered task clusters. This allows multiple overdue patching tasks with the same or compatible glass process attribute parameters to be aggregated into a single processing unit. This lays the data foundation for subsequent joint layout optimization of all tasks in the cluster on the same original sheet. Furthermore, since the task clustering process is automatically completed by the clustering production system based on glass process attribute parameters, there is no need for manual comparison of the specifications, thickness, color, and process type of each patching task. This significantly reduces the degree of human involvement and comparison time, thereby effectively improving the efficiency of hollow sheet clustering production.

[0060] In step S103 of some embodiments, the original film specification requirement information refers to the set of information output after the original film requirement prediction processing, which includes parameters such as the required original film size, thickness, and color.

[0061] In this embodiment, the sheet-aligning scheduling system first selects the target process attribute parameters corresponding to the sheet-aligning task cluster from the glass process attribute parameters. These target process attribute parameters at least include the specific length, width, and thickness data of each task in the sheet-aligning task cluster. Then, the sheet-aligning scheduling system performs geometric dimension analysis on the target process attribute parameters to obtain a task size distribution set of each task's size data. Next, the sheet-aligning scheduling system performs processing surface accumulation and estimation based on the task size distribution set. That is, it estimates the total raw sheet area required to complete all tasks in the sheet-aligning task cluster based on the area of ​​each small piece and the loss of layout gaps, and obtains the predicted layout area. Then, it combines the predicted layout area with the thickness and color information in the target process attribute parameters to generate a raw sheet requirement feature vector. Finally, the sheet-aligning scheduling system performs specification optimization matching in a preset standard raw sheet specification library based on the raw sheet requirement feature vector to obtain raw sheet specification requirement information containing the required raw sheet size, thickness, and color.

[0062] For details, please refer to Figure 4 In some embodiments, step S103 may include, but is not limited to, steps S401 to S405: Step S401: Based on the glass processing task cluster, perform parameter filtering on the glass processing attribute parameters to obtain the target processing attribute parameters; Step S402: Perform geometric dimension analysis on the target process attribute parameters to obtain the task dimension distribution set; Step S403: Based on the task size distribution set, accumulate and estimate the processing surface area to obtain the predicted page area requirement. Step S404: Combine the predicted plate area requirement with the target process attribute parameters to obtain the original plate requirement feature vector; Step S405: Based on the original film requirement feature vector, perform specification optimization matching in the preset standard original film specification library to obtain the original film specification requirement information.

[0063] In step S401 of some embodiments, the target process attribute parameter refers to the process parameter used for geometric dimension analysis selected from all glass process attribute parameters corresponding to the assembling task cluster. The target process attribute parameter typically includes the specific length, width, and thickness data of each assembling task to be processed in the assembling task cluster. For example, in the raw sheet demand prediction scenario in the field of glass deep processing, when the assembling task cluster contains three overdue sheet replenishment tasks, the target process attribute parameter can be a set of size and thickness data selected from the glass process attribute parameters of the three tasks respectively.

[0064] In this embodiment, the sheet-forming scheduling system first queries the glass process attribute parameters to find the target parameters corresponding to each sheet-forming task in the sheet-forming task cluster. Then, it extracts the specific length, width, thickness, color, glass type, and other data of each sheet-forming task from the target parameters, while ignoring other attribute information that is not related to geometric dimension analysis. Finally, it gathers the extracted specific length, width, thickness, color, glass type, and other data to obtain the target process attribute parameters.

[0065] In step S402 of some embodiments, the task size distribution set refers to the distribution list of the size data of each piece-fitting task to be processed in the piece-fitting task cluster obtained after geometric size analysis of the target process attribute parameters. For example, in the area estimation scenario in the glass deep processing field, when the piece-fitting scheduling system needs to know the specific size of each piece-fitting task in the piece-fitting task cluster for area accumulation, the task size distribution set can be the set of glass length and width size data corresponding to each of the five piece-fitting tasks in a certain piece-fitting task cluster, which reflects the size distribution of the piece-fitting task cluster.

[0066] In this embodiment, the production scheduling system first reads the specific length and width dimensions of each piece-to-be-processed task from the target process attribute parameters, and then organizes and summarizes all the read dimension data according to the corresponding task identifier to form a task dimension distribution set.

[0067] In step S403 of some embodiments, the predicted panel area requirement refers to the total original sheet area required to complete all tasks in the aligning task cluster, obtained by summing the glass sheet areas corresponding to each size data in the task size distribution set, and considering the layout gap loss and cutting kerf loss based on the summation result. For example, in the original sheet specification matching scenario in the glass deep processing field, when the task size distribution set contains multiple glass sheets of different sizes such as 1000 mm × 1200 mm and 800 mm × 1000 mm, the predicted panel area requirement can be the total area obtained by summing the glass sheets and adding 5% layout gap loss to the original summation value. It should be noted that the predicted panel area requirement reflects the lower limit of the original sheet area required to complete all the aligning tasks to be processed in the aligning task cluster.

[0068] In this embodiment, the sheet-making system first calculates the glass sheet area corresponding to each size data in the task size distribution set, then adds up the areas of each glass sheet, and adds the allowance for layout gap loss and cutting kerf loss to the accumulated result, so as to obtain the predicted layout area required to complete all tasks of the sheet-making task cluster.

[0069] In step S404 of some embodiments, the original sheet demand feature vector refers to vector data obtained by combining the predicted sheet area demand with the thickness and color information in the target process attribute parameters. For example, in the specification matching scenario in the glass deep processing field, when the sheet scheduling system needs to find a suitable original sheet specification in the standard original sheet specification library, the original sheet demand feature vector can be a vector composed of data from three dimensions: a thickness of 5 mm, a color category of clear glass, and a demand area of ​​7.5 square meters.

[0070] In this embodiment, the production scheduling system first extracts the thickness and color information of the alignment task cluster from the target process attribute parameters. Then, it merges the predicted plate requirement area, thickness value and color information of the alignment task cluster as the various dimensions of a multi-dimensional vector to obtain the original plate requirement feature vector.

[0071] In step S405 of some embodiments, the standard raw glass specification library refers to a database that stores the standard size specifications of raw glass commonly used by glass processing enterprises. For example, in the specification optimization and matching scenario in the field of glass deep processing, when the sheet production scheduling system needs to find the optimal matching specification among the preset raw glass specifications, the standard raw glass specification library may be a data table containing commonly used standard specifications such as 2440 mm × 3660 mm and 2440 mm × 1830 mm, as well as their corresponding thickness and color information.

[0072] In this embodiment, the film scheduling system uses the thickness and color dimensions in the original film demand feature vector as filtering conditions. It can filter a set of candidate specifications from the standard original film specification library that match both thickness and color. Then, it searches this candidate specification set for the specification whose original film area is greater than or equal to the predicted layout area and has the smallest area difference, selecting this as the optimal matching specification. This yields the original film specification requirement information, including the required original film size, thickness, and color. For example, when the predicted layout area in the original film demand feature vector is 7.5 square meters, the thickness is 5 millimeters, and the color is clear glass, the film scheduling system first filters all specifications with a thickness of 5 millimeters and a clear glass color from the standard original film specification library. Then, it selects the 2440 mm × 3660 mm specification, which has an area greater than 7.5 square meters and is the closest, as the optimal matching result, thus obtaining the original film specification requirement information.

[0073] Steps S401 to S405, as illustrated in this embodiment, involve the assembly scheduling system filtering all glass process attribute parameters corresponding to the assembly task cluster based on the cluster. This allows for the precise extraction of length, width, and thickness data directly related to size calculations from various process attributes of each assembly task to be processed, and the aggregation of these data into target process attribute parameters. This provides accurate data input covering the entire assembly task cluster for subsequent area estimation. Next, geometric dimension analysis of the target process attribute parameters yields a task size distribution set. Based on this task size distribution set, processing surface accumulation and estimation are performed to obtain the predicted plate area requirement. This allows for the accurate calculation of the total raw sheet area required to complete all tasks in the assembly task cluster, based on the actual size data of each patching task within the cluster. This provides precise data for subsequent raw sheet specification matching. Based on accurate area data, the predicted layout area is then combined with the thickness and color information in the target process attribute parameters to generate a raw material requirement feature vector. This integrates area requirements and physical attribute constraints into unified vector data, enabling subsequent specification matching to simultaneously consider area sufficiency and attribute consistency. Finally, based on this raw material requirement feature vector, specification optimization matching is performed in a preset standard raw material specification library. This automatically finds the optimal specification from the company's commonly used standard raw material specifications that meets both area requirements and minimizes material waste. The process of obtaining raw material specification requirement information does not require manual querying of the specification library or manual calculation of area differences, significantly reducing manual intervention and matching time. This provides accurate specification data for subsequent independent layout optimization and inventory verification, effectively improving the efficiency of hollow sheet production scheduling.

[0074] In step S104 of some embodiments, the target glass sheet production layout refers to a layout scheme that includes the position coordinates, cutting paths, cutting order, etc. of each small glass piece on the original sheet and can be directly sent to the cutting machine for execution.

[0075] In this embodiment, the glass sheet alignment and scheduling system first extracts the size features of each glass sheet alignment task belonging to the same alignment task cluster to obtain the specific length and width dimensions of the glass sheets required for each glass sheet alignment task, forming a set of task geometric dimensions. Then, based on the set of task geometric dimensions and the standard original sheet size specifications determined in the original sheet specification requirements information, the glass sheet alignment and scheduling system performs layout array calculations to generate an initial layout array containing the preliminary position information of each small piece. Finally, the glass sheet alignment and scheduling system performs physical layout reconstruction on the initial layout array to obtain a target glass sheet alignment production layout containing the final position coordinates, cutting paths, and cutting order data of each small glass piece on the original sheet.

[0076] For details, please refer to Figure 5 In some embodiments, step S104 may include, but is not limited to, steps S501 to S505: Step S501: Extract size features from each homogeneous task with the same attribute to obtain a set of task geometric dimensions; Step S502: Based on the task geometry set and the original film specification requirements, perform layout array calculation to obtain the initial layout array; Step S503: Based on the initial layout array, perform physical layout reconstruction to obtain the target uniform production layout.

[0077] In step S501 of some embodiments, homogeneous tasks of the same attribute refer to each unprocessed homogeneous task in a homogeneous task cluster.

[0078] The task geometry set refers to the collection of specific length and width dimensions of the glass slides required for each task with the same properties.

[0079] In this embodiment of the application, the glass sheet scheduling system can read the specific length and width dimensions of the glass sheets required for each glass sheet task with the same attribute from the target process attribute parameters corresponding to each glass sheet task with the same attribute in the glass sheet task cluster. At the same time, by summing up all the length and width dimensions read, the set of task geometric dimensions can be obtained.

[0080] In step S502 of some embodiments, the initial layout array refers to the layout scheme that includes the preliminary position information of each small piece of glass on the original sheet, generated after the layout array calculation.

[0081] In this embodiment, the production scheduling system first obtains the size specifications of the target original sheet from the original sheet specification requirements information to determine the original sheet boundary constraints. Under these boundary constraints, the system performs two-dimensional arrangement position calculations on each glass sheet in the task geometry set with the optimization objective of maximizing the original sheet utilization rate. This yields an initial arrangement array containing the preliminary position information of the small pieces corresponding to each same-attribute production scheduling task on the original sheet.

[0082] In step S503 of some embodiments, the sheet-forming production system, based on the initial layout array, can call preset cutting slit width parameters, edge grinding allowance parameters, and glass edge distance parameters to add cutting slit widths to adjacent positions between small glass pieces, add edge grinding allowances to the edges of each small glass piece, and add glass edge distances to the four boundaries of the original sheet. This allows the coordinates of each small piece in the initial layout array, which only contains geometric position information, to be corrected by the above-mentioned process constraints, forming a target sheet-forming production layout that includes the final position coordinates, cutting paths, and cutting sequence data of each small glass piece on the original sheet.

[0083] Steps S501 to S503 of this embodiment involve extracting the dimensional features of each uniform glass task with the same attribute through the uniform glass production system to obtain a set of task geometric dimensions. This allows for accurate acquisition of the actual length and width dimensions of the glass sheets required for each replacement task from the uniform glass task cluster, providing precise dimensional input for subsequent layout calculations. Next, based on the task geometric dimension set and the original sheet specification requirements, an initial layout array is generated through layout array calculation. This allows for the calculation of the layout positions of each small glass sheet within the original sheet boundary with the optimization goal of maximizing the utilization rate of the original sheet. This avoids unreasonable layout and waste of original sheets caused by subjective judgment during manual layout. Finally, a physical layout is reconstructed based on the initial layout array. By adding actual process constraints such as cutting kerf width, grinding allowance, and glass edge distance to the initial layout array, the theoretical layout result with only geometric position information is converted into an executable layout that meets the actual processing requirements of the cutting equipment. This allows the optimized target uniform glass production layout to be directly sent to the cutting machine for execution without the occurrence of cutting defects due to lack of process compensation.

[0084] In step S105 of some embodiments, inventory verification information refers to the data object containing the verification results output after the inventory verification operation is completed. When at least one rack successfully matches the required specifications, the inventory verification information includes inventory satisfaction information and data of the successfully matched racks. When no racks successfully match, the inventory verification information includes inventory non-satisfaction information and data of the missing original sheet specifications.

[0085] Inventory fulfillment information refers to the information in the inventory verification results that indicates the existence of successfully matched racks. The inventory fulfillment information also includes the rack number of the successfully matched rack and the number of original pieces remaining on each rack.

[0086] In this embodiment, the production scheduling system first obtains inventory raw material attribute data from the raw material warehouse. This inventory raw material attribute data refers to the specifications, thickness, color, remaining quantity, and shelf location information of the raw materials on each shelf. Then, the system performs a specification matching comparison between the required raw material specification parameters in the raw material specification requirement information and the raw material attribute information of each shelf in the inventory raw material attribute data. This yields a comparison verification result indicating whether each shelf has successfully matched or failed to match. Finally, the system generates verification information based on this comparison verification result. Specifically, when at least one shelf in the comparison verification result successfully matches the required specifications, the inventory verification information includes inventory satisfaction information, which carries the shelf number of the successfully matched shelf and the distribution data of the remaining quantity of raw materials on each shelf. When no shelf in the comparison verification result successfully matches, the inventory verification information includes inventory dissatisfaction information, which carries the specification data of the missing raw materials.

[0087] For details, please refer to Figure 6 In some embodiments, step S105 may include, but is not limited to, steps S601 to S603: Step S601: Obtain the attribute data of the original film in stock; Step S602: The original film specification requirements information is compared with the inventory original film attribute data to obtain the comparison and verification results; Step S603: Based on the comparison and verification results, verification information is generated to obtain inventory verification information. If the comparison and verification results show that the original film specification requirements information and the inventory original film attribute data are successfully matched, the inventory verification information includes inventory fulfillment information.

[0088] In step S601 of some embodiments, the inventory raw film attribute data refers to the attribute information data related to the raw films stored on each shelf, which is obtained from the raw film warehouse. It should be noted that the inventory raw film attribute data includes at least the specifications, thickness, color, remaining number of films and shelf location information of the raw films on each shelf.

[0089] In this embodiment of the application, the whole-piece scheduling system can send a data read request to the raw material warehouse. In response to the data read request, the raw material warehouse returns the specifications, thickness, color, remaining number of pieces and shelf location information of the raw materials on each shelf. Then, the whole-piece scheduling system can receive the above-mentioned raw material information and use it as inventory raw material attribute data. It should be noted that the inventory raw material attribute data can reflect the real-time inventory status in the raw material warehouse when performing inventory verification operations.

[0090] In step S602 of some embodiments, the comparison verification result refers to the comparison result data output after specification matching and comparison, indicating whether each rack is successfully matched or not matched with the required specifications. For example, in the inventory verification scenario in the glass deep processing field, after the sheet production scheduling system completes specification matching and comparison of the inventory raw sheet attribute data of all racks, the comparison verification result may be the comparison result data of the 3rd rack being successfully matched, the 5th rack being successfully matched, and the remaining racks failing to match.

[0091] In this embodiment, the production scheduling system can first extract the required thickness, size, and color of the original sheet from the original sheet specification requirement information as matching conditions. Then, it sequentially reads the thickness, size, and color information of the original sheets in each rack from the inventory original sheet attribute data. Next, it checks the consistency between the thickness of the original sheets in each rack and the required thickness, the color of the original sheets in each rack and the required color, and the size of the original sheets in each rack and the required size. When the thickness, color, and size of the original sheet in a rack are completely consistent with the above matching conditions, the rack is marked as successfully matched. When any one of the thickness, color, and size of the original sheet in a rack is inconsistent, the rack is marked as unmatched. Finally, after the production scheduling system completes the above comparison operation for all racks one by one, it can obtain the comparison verification results of whether each rack is successfully matched or unmatched.

[0092] In step S603 of some embodiments, the wafer scheduling system first traverses and compares the matching marks of each rack in the verification results. When the traversal results show that at least one rack is marked as a successful match, the wafer scheduling system determines that there is available wafer inventory in the wafer warehouse and can generate inventory verification information containing inventory satisfaction information. This inventory satisfaction information carries the rack number of the successful match and the distribution data of the remaining wafers on each rack. When the traversal results show that there is no rack marked as a successful match, the wafer scheduling system determines that there is no available wafer inventory in the wafer warehouse and can generate inventory verification information containing inventory dissatisfaction information. This inventory dissatisfaction information carries the specification data of the missing wafer.

[0093] In steps S601 to S603 as illustrated in some embodiments, the raw material attribute data of the inventory is obtained from the raw material warehouse through the production scheduling system. This allows for real-time acquisition of key information such as the specifications, thickness, color, remaining quantity, and shelf location of the raw materials on each rack, providing an accurate data foundation for subsequent inventory verification. This avoids discrepancies between verification results and actual inventory caused by using expired or cached data. Next, by comparing the raw material specification requirements with the real-time inventory raw material attribute data, the system automatically and objectively identifies all racks in the warehouse system that can meet the current patch production requirements, based on the consistency of thickness, color, and size specifications across each rack. This avoids potential omissions and judgment errors that might occur from manually checking inventory ledgers one by one. Finally, based on the comparison and verification results, verification information is generated. When a matching rack exists, inventory verification information containing inventory satisfaction information and matching rack data is generated to trigger subsequent production scheduling. When no matching rack exists, inventory verification information containing inventory insufficiency information is generated to trigger material control allocation notification. This allows the whole-piece scheduling system to automatically divert to different processing paths according to the type of inventory verification information, eliminating the need for manual confirmation of inventory status and manual notification to the material control department before production scheduling. This significantly reduces cross-departmental communication time and manual confirmation steps, thereby effectively improving the efficiency of hollow whole-piece scheduling.

[0094] In step S106 of some embodiments, the wafer alignment scheduling task refers to a complete production instruction data package generated after the wafer alignment scheduling process is completed, whose main content is CNC code that can be directly read and executed by the cutting machine controller. In the task issuance scenario in the field of glass deep processing, when the production execution system needs to transmit the scheduling instruction to the connected production line, the wafer alignment scheduling task can be a production execution instruction that includes the cutting coordinate path and the target material rack number.

[0095] In this embodiment, the production scheduling system first analyzes the shelf piece distribution of the inventory satisfaction information to obtain the shelf piece distribution data. Then, the system selects the target shelf from multiple available shelf pieces based on the shelf piece distribution data. It should be noted that the selection rule prioritizes the shelf with the fewest remaining original pieces to reduce scattered inventory backlog. Next, the system associates and binds each cut piece in the target production panel with the corresponding replacement task order in the production task cluster, generating a production processing dataset containing cutting position data and corresponding order identifiers. Finally, the system packages and integrates the production processing dataset with the target shelf number information and assigns a task priority identifier to obtain a production scheduling task containing the cutting coordinate path and the target shelf number.

[0096] For details, please refer to Figure 7In some embodiments, step S106 may include, but is not limited to, steps S701 to S704: Step S701: Analyze the shelf piece distribution of the inventory information to obtain the original shelf piece distribution data; Step S702: Based on the original sheet number distribution data of the material rack, the material rack is screened to obtain the target material rack; Step S703: Based on the target production layout and production task cluster, reconstruct the production scheduling data to obtain the production processing dataset. Step S704: Based on the whole-piece processing dataset and the target material rack, the production task is encapsulated to obtain the whole-piece production scheduling task.

[0097] In step S701 of some embodiments, the original sheet number distribution data of the rack refers to the distribution data of the number of original sheets remaining on each successfully matched rack. For example, in the rack screening scenario in the glass deep processing field, when the inventory information includes three successfully matched racks (rack 3, rack 5, and rack 7), the original sheet number distribution data of the rack can be the distribution data of 3 sheets remaining on rack 3, 8 sheets remaining on rack 5, and 12 sheets remaining on rack 7.

[0098] In this embodiment of the application, the production scheduling system can extract the number of each successfully matched rack and the number of original wafers remaining in each rack from the inventory satisfaction information. Then, the rack number and the number of remaining wafers are organized into corresponding relationship data to obtain the distribution data of the number of original wafers in the rack.

[0099] In step S702 of some embodiments, the target rack refers to the rack selected from multiple available racks for this replacement production based on the rack sheet number distribution data and a preset screening rule. For example, in the rack screening scenario in the glass deep processing field, when the rack sheet number distribution data shows that rack 3 has 3 sheets remaining, rack 5 has 8 sheets remaining, and rack 7 has 12 sheets remaining, and the preset screening rule is to prioritize the rack with the fewest remaining sheets, the target rack can be rack 3, which is selected because it has the fewest remaining sheets, in order to reduce the accumulation of scattered inventory by prioritizing the consumption of the rack with the least inventory.

[0100] In this embodiment, the production scheduling system first calls the preset rack screening rules, then iterates through each rack in the original sheet number distribution data, compares the remaining sheet number of each rack one by one, and selects the rack that meets the rack screening rules as the target rack. Commonly, the rack screening rules prioritize the rack with the fewest remaining original sheets to reduce scattered inventory backlog.

[0101] In step S703 of some embodiments, the wafer processing dataset refers to a data set containing cutting position data and corresponding task order identifiers generated after the production scheduling data is reconstructed. For example, in the production task packaging scenario of the glass deep processing field, after the production scheduling data is reconstructed and the association between each cutting wafer and the order is bound, the wafer processing dataset can be a set of data records containing the coordinate position of each cutting wafer and its corresponding order number.

[0102] In this embodiment of the application, the target uniform production layout contains the coordinate positions and cutting order information of multiple cutting pieces. Each cutting piece corresponds to a uniform task with the same attribute in the uniform task cluster. Therefore, the uniform scheduling system can obtain the mapping relationship between each cutting piece and its corresponding uniform task with the same attribute in the target uniform production layout, and write the mapping relationship into the attribute information of each cutting piece in a data association manner, thereby generating a uniform processing dataset containing cutting position data and corresponding task order identifier.

[0103] In step S704 of some embodiments, the assembly scheduling system can first query the target rack number information and the task urgency information of the assembly task cluster. Then, it packages and integrates the cutting position data, task order identifier association data, target rack number, and task priority identifier in the assembly processing dataset to form an assembly scheduling task. It should be noted that the assembly scheduling task can be directly sent to the connected production line to perform the patch cutting operation.

[0104] Steps S701 to S704, as illustrated in this embodiment, involve parsing the rack piece distribution data of the inventory satisfaction information using a rack-based production scheduling system. This allows for the accurate extraction of the remaining piece distribution of each successfully matched rack from the inventory satisfaction information, providing quantitative data for subsequent rack selection. Based on this rack piece distribution data, rack selection is performed, automatically determining the target rack from multiple available racks according to a preset rule that prioritizes racks with the fewest remaining pieces. This reduces scattered inventory buildup and optimizes the company's inventory structure by prioritizing racks with the lowest inventory levels. Next, production scheduling data is reconstructed based on the target rack-based production layout and rack-based task clusters to generate a rack-based processing dataset. Each cut piece in the standard sheet production panel is precisely associated with its corresponding replacement sheet task order, ensuring that each cut piece of glass has a clear order affiliation. This avoids confusion caused by mismatches between cut pieces and orders after production. Finally, based on the sheet processing dataset and target material rack, production tasks are encapsulated to generate sheet scheduling tasks. This integrates cutting data, material rack numbers, and priority identifiers into complete production instructions that can be directly executed by production line equipment. The sheet scheduling tasks can directly drive production line equipment to perform replacement sheet cutting operations without manual re-entry or adjustment. This achieves automated encapsulation of scheduling data from optimization results to production line execution instructions, significantly reducing manual data entry time and the risk of entry errors, thereby effectively improving the efficiency of hollow sheet scheduling.

[0105] This application first obtains a set of pending sheet-fitting tasks containing multiple tasks, and then performs task clustering based on the glass process attribute parameters of each task. This clustering aggregates overdue sheet-fitting tasks with the same or compatible attributes into a sheet-fitting task cluster, realizing the batch collection and merging of overdue sheet-fitting tasks across production lines. This lays the data foundation for intensive production scheduling and effectively avoids the problems of missed orders and low efficiency in traditional manual order selection. Next, based on this sheet-fitting task cluster and glass process attribute parameters, raw sheet demand is predicted. The required raw sheet specifications can be pre-calculated based on the size distribution and process characteristics of each task after clustering, providing accurate specification matching for inventory verification and layout optimization, avoiding invalid optimization calculations due to specification uncertainty. Subsequently, based on the raw sheet specification requirement information, the sheet-fitting task cluster is independently optimized for layout. This generates a completely new production layout for the merged task cluster, independent of the current production layout. Real-time task matching or insertion ensures that overdue backlog tasks can complete layout independently and efficiently, overcoming the shortcomings of traditional manual optimization, such as time lag and lack of specificity. Inventory verification is then performed based on original sheet specifications. When the verification result includes inventory availability, the sufficiency of original sheet resources is automatically confirmed, effectively ensuring that raw materials are ready for execution before production scheduling instructions are issued, and preventing production interruptions due to original sheet shortages. Finally, based on inventory availability and the target sheet production layout, sheet production scheduling is performed to generate sheet production tasks. The optimized layout data, selected original sheet rack information, and order information are comprehensively packaged and directly sent to the production line for execution, significantly reducing manual intervention, shortening sheet delivery cycles, and improving the accuracy of patching operations. This effectively improves the efficiency of hollow sheet production scheduling and solves the technical problems of delayed delivery caused by manual intervention and time lag in existing technologies.

[0106] Please see Figure 8 This application also provides an inline hollow wafer assembly and scheduling device, which can realize the above-mentioned inline hollow wafer assembly and scheduling method. The device includes: The task acquisition module 801 is used to acquire the set of unprocessed glass task pieces, which contains multiple unprocessed glass task pieces, and each unprocessed glass task piece contains glass process attribute parameters. The task clustering module 802 is used to cluster the set of whole-piece tasks to be processed based on glass process attribute parameters to obtain whole-piece task clusters. The raw material prediction module 803 is used to predict the raw material demand based on the uniform task cluster and glass process attribute parameters, and obtain the raw material specification demand information. The layout optimization module 804 is used to perform independent layout optimization on the aligned task cluster based on the original film specification requirements information to obtain the target aligned film production layout. The inventory verification module 805 is used to perform inventory verification based on the original film specification requirement information to obtain inventory verification information, which includes inventory fulfillment information. The production scheduling module 806 is used to perform production scheduling on the production task cluster based on inventory satisfaction information and target production layout, and obtain production scheduling tasks.

[0107] The specific implementation of this inline hollow wafer assembly production device is basically the same as the specific implementation of the above-mentioned inline hollow wafer assembly production method, and will not be repeated here.

[0108] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method for aligning hollow wafers in production. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0109] Please see Figure 9 , Figure 9 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 901 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 902 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 902 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 902 and called and executed by the processor 901 using the wiring hollow wafer alignment method of the embodiments of this application. The input / output interface 903 is used to implement information input and output; The communication interface 904 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 905 transmits information between various components of the device (e.g., processor 901, memory 902, input / output interface 903, and communication interface 904); The processor 901, memory 902, input / output interface 903, and communication interface 904 are connected to each other within the device via bus 905.

[0110] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for arranging hollow wafers in interconnected circuits.

[0111] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0112] The online hollow wafer alignment scheduling method, online hollow wafer alignment scheduling device, electronic device, and storage medium provided in this application embodiment first acquire a set of wafer alignment tasks to be processed, wherein the set of wafer alignment tasks to be processed contains multiple wafer alignment tasks, and each wafer alignment task contains glass process attribute parameters. Then, based on the glass process attribute parameters, the set of wafer alignment tasks to be processed is clustered to obtain wafer alignment task clusters. Next, based on the wafer alignment task clusters and the glass process attribute parameters, the raw wafer demand is predicted to obtain raw wafer specification demand information. Finally, based on the raw wafer specification demand information, the wafer alignment task clusters are optimized for independent layout. Upon reaching the target production layout, inventory verification is performed based on the original sheet specifications and requirements to obtain inventory verification information that includes inventory fulfillment information. Finally, based on the inventory fulfillment information and the target production layout, the production scheduling of the production task cluster is processed to obtain the production scheduling task. This achieves fully automated processing from batch collection of overdue replacement tasks, original sheet demand forecasting, independent layout optimization, automatic inventory verification to automatic generation of scheduling instructions. It avoids the tedious operations of manual order selection, manual matching, and manual layout optimization, shortens the processing cycle of replacement tasks, and improves the efficiency of hollow sheet production scheduling.

[0113] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0114] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0115] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0116] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0117] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0118] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0119] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. The coupling or direct coupling or communication connection between the shown or discussed units may be through some interfaces, or indirect coupling or communication connection between the apparatus or units, and may be electrical, mechanical, or other forms.

[0120] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0121] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0122] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0123] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for producing hollow wafers in a continuous line, characterized in that, The method includes: Obtain a set of unprocessed glass task pieces, wherein the set of unprocessed glass task pieces contains multiple unprocessed glass task pieces, and each unprocessed glass task piece contains glass process attribute parameters. Based on the glass process attribute parameters, the set of unprocessed wafer tasks is clustered to obtain wafer task clusters. Based on the aforementioned wafer task cluster and the aforementioned glass process attribute parameters, the raw wafer demand is predicted to obtain raw wafer specification demand information. Based on the original film specification requirements, the layout of the film cluster is optimized independently to obtain the target film production layout. Based on the original film specification requirements information, inventory verification is performed to obtain inventory verification information, wherein the inventory verification information includes inventory satisfaction information; Based on the inventory satisfaction information and the target production layout, the production scheduling of the production task cluster is performed to obtain the production scheduling task.

2. The method according to claim 1, characterized in that, The process of predicting raw wafer demand based on the wafer task cluster and the glass process attribute parameters to obtain raw wafer specification demand information includes: Based on the aforementioned glass processing task cluster, the glass processing attribute parameters are filtered to obtain the target processing attribute parameters. Geometric dimension analysis is performed on the target process attribute parameters to obtain the task dimension distribution set; Based on the set of task size distributions, the processing surface is accumulated and estimated to obtain the predicted page area requirement. The predicted plate area requirement is combined with the target process attribute parameters to obtain the original plate requirement feature vector. Based on the original film requirement feature vector, the specification optimization and matching are performed in the preset standard original film specification library to obtain the original film specification requirement information.

3. The method according to claim 1, characterized in that, Based on the inventory satisfaction information and the target production layout, the process of scheduling production for the batch of batch production tasks is performed to obtain batch production scheduling tasks, including: The inventory information is analyzed to obtain the original shelf piece distribution data. Based on the original sheet number distribution data of the material rack, the material rack is screened to obtain the target material rack; Based on the target production layout and the production task cluster, the production scheduling data is reconstructed to obtain the production processing dataset. Based on the pre-cut processing dataset and the target material rack, the production task is encapsulated to obtain the pre-cut production scheduling task.

4. The method according to claim 1, characterized in that, The inventory verification based on the original wafer specification requirements information, to obtain inventory verification information, includes: Obtain the attribute data of the original film in stock; The specification requirements of the original film are matched and compared with the attribute data of the original film in stock to obtain the comparison and verification results; Based on the comparison and verification results, verification information is generated to obtain the inventory verification information. If the comparison and verification results indicate that the original film specification requirement information and the inventory original film attribute data are successfully matched, the inventory verification information includes the inventory satisfaction information.

5. The method according to claim 1, characterized in that, The cluster of homogeneous tasks contains multiple homogeneous tasks with the same attribute; The step of optimizing the layout of the whole-piece task cluster independently based on the original film specification requirements information to obtain the target whole-piece production layout includes: Size features are extracted from each of the homogeneous tasks with the same attributes to obtain a set of task geometric dimensions; Based on the set of task geometric dimensions and the original film specification requirements, the layout array is calculated to obtain the initial layout array; The physical layout is reconstructed based on the initial arrangement array to obtain the target uniform production layout.

6. The method according to any one of claims 1-5, characterized in that, Based on the glass process attribute parameters, the set of tasks to be processed for uniform lamination is clustered to obtain uniform lamination task clusters, including: Feature extraction is performed on the glass process property parameters to obtain a glass identification feature vector; Based on the glass identification feature vector, the set of uniform glass tasks to be processed is homogeneously distinguished to obtain uniform glass tasks with the same attributes. The tasks with the same attributes are grouped together to obtain the cluster of tasks.

7. The method according to any one of claims 1-5, characterized in that, The process of obtaining the set of tasks to be processed includes: Obtain a set of patch tasks to be processed, wherein the set of patch tasks to be processed contains multiple patch tasks to be processed; The duration of each pending patch task is queried to obtain multiple task rolling duration data. The rolling duration data of each task is compared with the preset dynamic patch timeout threshold to obtain duration comparison information, wherein the duration comparison information includes task timeout information. Based on the task timeout information, the set of pending patch tasks is filtered to obtain the set of pending complete patch tasks.

8. A connecting hollow wafer straightening and discharging device, characterized in that, The device includes: The task acquisition module is used to acquire a set of unprocessed glass task pieces, wherein the set of unprocessed glass task pieces contains multiple unprocessed glass task pieces, and the unprocessed glass task pieces contain glass process attribute parameters. The task clustering module is used to cluster the set of unprocessed wafer tasks based on the glass process attribute parameters to obtain wafer task clusters. The raw glass prediction module is used to predict the raw glass demand based on the glass assembly task cluster and the glass process attribute parameters, and obtain raw glass specification demand information. The layout optimization module is used to perform independent layout optimization on the whole-piece task cluster based on the original piece specification requirements information to obtain the target whole-piece production layout. The inventory verification module is used to perform inventory verification based on the original film specification requirement information to obtain inventory verification information, wherein the inventory verification information includes inventory satisfaction information. The production scheduling module is used to perform production scheduling on the production task cluster based on the inventory satisfaction information and the target production layout, so as to obtain the production scheduling task.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the wire-connected hollow wafer scheduling method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for arranging hollow wafers according to any one of claims 1 to 7.