Centralized distribution method, device and medium for line detection

Abstract

The application relates to the technical field of production line automation detection, and provides a centralized distribution method, equipment and medium for production line detection. The method comprises the following steps: obtaining sorting information of each to-be-detected board entering a corresponding sorting machine of a production line, wherein the sorting information at least comprises an entering time corresponding to the to-be-detected board; obtaining idle information corresponding to the sorting machine, wherein the idle information at least comprises a current idle layer number; determining a processing priority of each to-be-detected board according to the sorting information corresponding to the to-be-detected board and the current idle information corresponding to the sorting machine; obtaining a current working state of each maintenance station corresponding to the production line; the current working state at least comprises an idle state and a busy state; and distributing each to-be-detected board to a corresponding maintenance station according to the processing priority corresponding to the to-be-detected board and the current working state of the maintenance station.

Description

Technical Field

[0001] This application relates to the field of automated production line inspection technology, and in particular to a centralized distribution method, equipment and medium for production line inspection. Background Technology

[0002] In the field of PCB production line inspection, existing technologies generally adopt a single-machine centralized judgment mode. This means that each production line's inspection equipment and sorting machine (NGBUFFER) operate independently, and manual review can only process the inspection results of a single machine sequentially. Specifically, after a PCB board completes inspection and flows into the sorting machine, if manual review is required, the operator must process the boards temporarily stored in each sorting machine one by one, and it is impossible to coordinate the production rhythm of multiple production lines. Although existing solutions project the interfaces of each production line's maintenance station to the central control maintenance station via a local area network, this only achieves "centralization" at the information display level and does not perform real-time collection and logical judgment of the board information stored in the sorting machines.

[0003] Therefore, a method is urgently needed to solve at least one of the above problems. Summary of the Invention

[0004] This application provides a centralized distribution method, equipment, and medium for production line inspection, aiming to solve the problem that in the field of PCB production line inspection, the existing technology generally adopts a single-machine centralized judgment mode, that is, each production line inspection equipment and sorting machine (NGBUFFER) operates independently, and the manual review process can only process the inspection results of a single machine sequentially. Specifically, after the PCB board completes inspection and flows into the sorting machine, if manual review is required, the operator must process the boards temporarily stored in each sorting machine one by one, and it is impossible to coordinate the production rhythm of multiple production lines.

[0005] In a first aspect, embodiments of this application provide a centralized distribution method for production line inspection, including: Obtain the sorting information of each board to be tested as it enters the corresponding sorting machine on the production line. The sorting information includes at least the entry time of the board to be tested. Obtain the idle information corresponding to the sorting machine, wherein the idle information includes at least the current idle layer number; Based on the sorting information corresponding to the board to be tested and the current idle information corresponding to the sorting machine, the processing priority of each board to be tested is determined; Obtain the current working status of each maintenance station on the production line; the current working status includes at least an idle state and a busy state; Based on the processing priority of the board to be tested and the current working status of the repair station, each board to be tested is distributed to the corresponding repair station.

[0006] In some embodiments, obtaining the sorting information of each board to be tested entering the corresponding sorting machine of the production line includes: obtaining the entry time, layer number, track position and preliminary judgment result of the board to be tested to generate the sorting information; the preliminary judgment result includes preliminary judgment qualified or preliminary judgment unqualified; wherein, when the board to be tested leaves the sorting machine, the track position is cleared.

[0007] In some embodiments, the method further includes: when a board to be tested that is initially deemed unqualified is detected in the sorting machine, it is identified as a target board; the unique identification information of the target board is obtained, and the board to be tested is removed from the sorting machine according to the unique identification information and sorting information.

[0008] In some embodiments, obtaining the idle information corresponding to the sorting machine includes: collecting the current number of idle layers, the number of preliminarily qualified test boards, and the number of preliminarily unqualified test boards of each sorting machine through a preset timed inspection mechanism, and generating the idle information.

[0009] In some embodiments, determining the processing priority of each board to be tested based on the sorting information corresponding to the board to be tested and the current idle information corresponding to the sorting machine includes: judging based on the number of idle layers of the sorting machine: if the number of idle layers is less, the processing priority of the corresponding board to be tested is higher; if the number of idle layers is the same, sorting according to the entry time of the board to be tested, the earlier the entry time, the higher the priority; for the detection section not connected to the sorting machine, it is set as the highest processing priority.

[0010] In some embodiments, obtaining the current working status of each maintenance station on the production line includes: obtaining the current task queue length, historical average processing speed, and current task difficulty level of the maintenance station; and determining the current working status of the maintenance station based on the current task queue length, historical average processing speed, and current task difficulty level.

[0011] In some embodiments, distributing each board to be tested to a corresponding maintenance station according to the processing priority of the board to be tested and the current working status of the maintenance station includes: distributing the board to be tested to a maintenance station in the idle state according to the processing priority; and / or, when multiple boards to be tested have the same processing priority, allocating the corresponding workload according to the historical processing efficiency of the maintenance station; and / or, distributing the successfully matched board to be tested to the corresponding maintenance station according to the matching relationship between the production line equipment model corresponding to the production line and the maintenance station model corresponding to the maintenance station.

[0012] In some embodiments, the method further includes: obtaining false alarm information corresponding to the production line; establishing a production line data model based on the false alarm information; and adjusting the task allocation quantity of the maintenance station based on the average number of false alarms per unit time.

[0013] Secondly, this application also provides a computer device, comprising: Memory and processor; The memory is used to store computer programs; The processor is configured to execute the computer program and, in executing the computer program, implement the steps of the method described in the first aspect above.

[0014] Thirdly, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the steps of the method described in the first aspect above.

[0015] This application prioritizes processing tasks by combining the number of idle layers on the sorting machine (reflecting equipment load) with the board inspection time (reflecting task urgency). It prioritizes high-load equipment and earlier-arriving inspection tasks, avoiding board piling up on the sorting machine and improving production line smoothness. Tasks are dynamically distributed based on the real-time working status (idle / busy) of the maintenance station, avoiding uneven workloads caused by fixed allocations and improving maintenance station resource utilization. The system automatically sorts and distributes tasks, reducing manual intervention, improving re-judgment efficiency, and significantly lowering the operational error rate.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic flowchart illustrating the steps of a centralized distribution method for production line inspection provided in an embodiment of this application; Figure 2 This is a schematic diagram of a scenario for a centralized distribution method for production line inspection provided in an embodiment of this application; Figure 3 This is a schematic block diagram of a centralized distribution system for production line inspection provided in one embodiment of this application; Figure 4This is a schematic block diagram of the structure of a computer device provided in an embodiment of this application.

[0019] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Detailed Implementation

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

[0021] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0022] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and the terms "first" and "second" are not necessarily different.

[0023] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0024] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0025] In the field of PCB production line inspection, existing technologies generally adopt a single-machine centralized judgment mode. This means that each production line's inspection equipment and sorting machine (NGBUFFER) operate independently, and manual review can only process the inspection results of a single machine sequentially. Specifically, after a PCB board completes inspection and flows into the sorting machine, if manual review is required, the operator must process the boards temporarily stored in each sorting machine one by one, and it is impossible to coordinate the production rhythm of multiple production lines. Although existing solutions project the interfaces of each production line's maintenance station to the central control maintenance station via a local area network, this only achieves "centralization" at the information display level and does not perform real-time collection and logical judgment of the board information stored in the sorting machines.

[0026] Therefore, a method is urgently needed to solve at least one of the above problems.

[0027] To solve the above problem, please refer to Figure 1 This application provides a centralized distribution method for production line inspection, applied to computer equipment. The computer equipment can be deployed on a single server or a server cluster. It can also be deployed on handheld terminals, laptops, wearable devices, or robots, etc. It should be noted that every piece of information involved in the method provided in this application is extracted with the authorization of the relevant user and in accordance with relevant regulations, and will not infringe on user privacy.

[0028] The provided centralized distribution method for production line inspection includes steps S101 to S105. Details are as follows: Step S101. Obtain the sorting information of each board to be tested entering the corresponding sorting machine on the production line. The sorting information includes at least the entry time of the board to be tested.

[0029] Specifically, in order to collect key information about each PCB board entering the sorting machine (NGBUFFER) in real time, the correspondence between the boards and the production line is established.

[0030] Sorting information may include: Unique board identifier (e.g., QR code, barcode, RFID tag): used to track individual boards. Entry time: a timestamp accurate to the second, reflecting the board's dwell time in the sorting machine. Production line identifier: records the production line to which the board belongs (e.g., Line A, Line B), facilitating subsequent cross-production line coordination. Initial inspection results: such as whether manual re-inspection is required, defect type, etc., to assist in priority calculation.

[0031] Install a QR code scanner or RFID reader at the entrance of the sorting machine to read the board markings; combine this with an industrial-grade timer or the equipment's built-in clock to generate the entry time.

[0032] Data is transmitted in real time to computer devices (such as edge servers) via a PLC (Programmable Logic Controller) or device API interface. The latest data is stored in a real-time database (such as Redis) for quick retrieval; at the same time, it is written to a relational database (such as MySQL) to form historical records.

[0033] Step S102. Obtain the idle information corresponding to the sorting machine, wherein the idle information includes at least the current idle layer number.

[0034] Specifically, to monitor the real-time capacity of the sorting machine's temporary storage area and prevent production line shutdowns due to backlog, the idle information includes: Sorting machine unique identifier: associated with the specific production line's sorting equipment. Total number of layers: the physical number of layers in the sorting machine's temporary storage area (e.g., 10 layers). Number of idle layers: the number of layers currently without boards, directly reflecting the equipment's processing pressure. Status of each layer (optional): whether each layer is occupied, board dwell time, etc., used for fine-tuning scheduling.

[0035] The sorting machine is equipped with photoelectric sensors or mechanical limit switches to detect whether there is a plate on each layer; the layer status is reported in real time through the equipment control system. The computer equipment periodically pulls data (e.g., once per second) via OPC UA protocol or Modbus communication, or the equipment actively pushes data (Push mode).

[0036] Status calculation: Number of idle layers = Total number of layers - Number of occupied layers. If the number of idle layers is lower than the threshold (e.g., ≤2 layers), an early warning mechanism is triggered, and the boards of this sorting machine are scheduled first.

[0037] Step S103. Determine the processing priority of each board to be tested based on the sorting information corresponding to the board to be tested and the current idle information corresponding to the sorting machine.

[0038] Specifically, in order to generate a reasonable processing sequence based on the board dwell time and sorter load, and to avoid production line blockage.

[0039] Boards that enter the sorting machine earlier have higher priority (FIFO strategy). The fewer idle layers the sorting machine has (i.e., the higher the load), the higher the priority of its boards, preventing equipment overload. It also allows boards with serious defects (such as short circuits) to be prioritized over those with minor defects.

[0040] Priority calculation formula: Define priority score: Priority = α × T + β × L. Where: T: the dwell time of the board in the sorting machine (seconds), logarithmic or normalized; α and β: weighting coefficients (e.g., α = 0.6, β = 0.4, which can be optimized through training with production line data). Sort all boards to be processed in descending order of priority to generate a priority queue (e.g., a stack structure).

[0041] The queue is refreshed regularly (e.g., once per minute) to ensure real-time performance; new boards are inserted into the queue in real time.

[0042] Step S104. Obtain the current working status of each maintenance station on the production line; the current working status includes at least an idle state and a busy state.

[0043] Specifically, this is to monitor the load of each maintenance station in real time and avoid imbalances in task allocation.

[0044] Idle Status: The repair station has no pending tasks and can immediately receive new boards. Busy Status: The repair station is processing boards; the remaining processing time is recorded (which can be estimated using historical data). Operator Skill Tags (Optional): such as "Proficient in handling missing components," used for task matching.

[0045] The maintenance station is equipped with an industrial tablet terminal or touch screen, allowing operators to manually switch statuses via the interface (e.g., clicking "Busy" to start processing and "Idle" to finish). Combined with IoT devices (such as RFID readers), the station automatically detects operator presence: if an operator leaves their workstation for more than 5 minutes, it is automatically marked as "Idle but Unattended." Status data is pushed to a computer in real time via the WebSocket protocol, maintaining a global status table (e.g., in JSON format).

[0046] Step S105. Based on the processing priority of the board to be tested and the current working status of the repair station, distribute each board to be tested to the corresponding repair station.

[0047] Specifically, the goal is to allocate high-priority boards to the most suitable repair stations to achieve task load balancing and rapid processing.

[0048] Boards are prioritized for allocation to repair stations with an "idle" status. Priority is given to assigning them to repair stations on the same production line to minimize board transfer time (this requires the production line and repair station to be physically linked). Additionally, repair stations with corresponding skills can be matched based on the defect type.

[0049] The allocation algorithm can be implemented by taking the highest priority board from the priority queue; filtering out repair stations with an "idle" status and sorting them by distance (or skill matching); allocating the board to the repair station at the top of the list, updating the repair station's status to "busy", and recording the start time of processing.

[0050] If all repair stations are busy, the station with the earliest estimated availability will be queued. The allocation result is sent to the sorting machine and repair station via an industrial network (such as EtherCAT): the sorting machine, according to instructions, transports the board to the designated conveyor line, guiding it to the corresponding repair station; the repair station terminal displays information about the board to be processed (such as defect location and historical inspection data). After processing, the repair station reports the result back to the system, updating the board status (such as "re-inspection passed" or "rework"), triggering subsequent processes (such as warehousing or re-inspection).

[0051] In some embodiments, obtaining the sorting information of each board to be tested entering the corresponding sorting machine of the production line includes: obtaining the entry time, layer number, track position and preliminary judgment result of the board to be tested to generate the sorting information; the preliminary judgment result includes preliminary judgment qualified or preliminary judgment unqualified; wherein, when the board to be tested leaves the sorting machine, the track position is cleared.

[0052] The sorting information of the board to be tested as it enters the sorting machine is obtained, including the entry time, number of layers, track position, and preliminary judgment result (qualified / unqualified), and the track position is cleared when the board leaves the sorting machine.

[0053] Sensors or scanning devices (such as barcode readers or RFID readers) are installed at the entrance of the sorting machine. When the PCB board enters the sorting machine, the unique identifier of the board (such as a QR code), the entry time, the number of layers assigned (layer position in the multi-layer storage structure of the sorting machine), and the track position (physical position number in the sorting machine) are automatically collected, and the preliminary judgment result (OK / NG) of the detection equipment is obtained simultaneously.

[0054] The above information is stored in the real-time database of the central server to establish a "board ID - sorting information" association record.

[0055] When a board leaves the sorting machine exit (e.g., manually picked up or automatically transported), the system detects the exit sensor trigger signal, automatically deletes the board's track position information from the database, releases storage space, and updates the sorting machine status.

[0056] In some embodiments, the method further includes: when a board to be tested that is initially deemed unqualified is detected in the sorting machine, it is identified as a target board; the unique identification information of the target board is obtained, and the board to be tested is removed from the sorting machine according to the unique identification information and sorting information.

[0057] When a board that is initially deemed unqualified (NG) is detected in the sorting machine, it is identified as the target board, and the board is removed by using its unique identifier and the information of the maintenance station is retrieved.

[0058] The system periodically scans the initial judgment results of the boards in the sorting machine, filters out all boards that are initially judged as NG, and marks them as "target boards".

[0059] Assign a unique identifier (such as laser marking or RFID tag) to each PCB board, and link this identifier to inspection records and repair station information.

[0060] The system generates board retrieval instructions (such as lighting up the corresponding layer indicator light or sending the robot arm coordinate signal) based on the storage layer number and track position of the target board, and notifies the operator or the automated robotic arm to locate and retrieve the board.

[0061] During board retrieval, the system automatically retrieves the corresponding service station record (such as test parameters and fault type) through the board's unique identifier and displays the relevant information on the service station's computer interface, avoiding manual searching.

[0062] In some embodiments, obtaining the idle information corresponding to the sorting machine includes: collecting the current number of idle layers, the number of preliminarily qualified test boards, and the number of preliminarily unqualified test boards of each sorting machine through a preset timed inspection mechanism, and generating the idle information.

[0063] The inspection cycle is set so that the system automatically inspects the status of the sorting machines on each production line every 5 seconds (configurable).

[0064] Data acquisition is performed in real time through an Industrial Internet of Things (IIoT) interface or PLC communication protocol to obtain the following data from the sorting machine: number of idle layers: total number of layers in the sorting machine - number of occupied layers; number of OK boards: number of boards that have passed the initial judgment and have not been shipped out; number of NG boards: number of boards that have failed the initial judgment.

[0065] The status data of each sorting machine is aggregated to the central server to generate a real-time report of "Production Line ID - Sorting Machine Status" for subsequent priority calculation.

[0066] In some embodiments, determining the processing priority of each board to be tested based on the sorting information corresponding to the board to be tested and the current idle information corresponding to the sorting machine includes: judging based on the number of idle layers of the sorting machine: if the number of idle layers is less, the processing priority of the corresponding board to be tested is higher; if the number of idle layers is the same, sorting according to the entry time of the board to be tested, the earlier the entry time, the higher the priority; for the detection section not connected to the sorting machine, it is set as the highest processing priority.

[0067] The processing priority is determined based on the number of idle layers in the sorting machine, the time the board enters, and the equipment connection status.

[0068] The priority algorithm includes: Prioritizing idle layers: The fewer idle layers the sorter has, the greater the storage pressure, and the higher the processing priority of the corresponding NG (Not Found) boards (e.g., setting idle layers ≤ 2 to "urgent"). When the number of idle layers is the same, boards are sorted by the time they entered the sorter, with earlier boards processed first (first-in, first-out principle). In inspection sections not connected to the sorter (e.g., due to equipment failure or temporary configuration), NG boards generated therein are directly set to the highest priority to avoid accumulation. After each inspection, the system sorts all pending NG boards according to the priority algorithm, generates a global priority queue, and synchronizes it to the central server.

[0069] In some embodiments, obtaining the current working status of each maintenance station on the production line includes: obtaining the current task queue length, historical average processing speed, and current task difficulty level of the maintenance station; and determining the current working status of the maintenance station based on the current task queue length, historical average processing speed, and current task difficulty level.

[0070] By obtaining the current task queue length, historical average processing speed, and task difficulty level of the repair station, its working status (idle / busy) can be determined.

[0071] Status parameters collected include: Task queue length: the number of NG boards currently pending processing at the repair station; Historical average processing speed: the average processing time for a single board at the repair station in the past 30 minutes (can be dynamically updated); Task difficulty level: based on the type of defect of the NG board (such as short circuit, missing parts, etc.) and historical processing time, the task is divided into three levels: "simple", "medium" and "complex" (e.g., complex task takes ≥5 minutes).

[0072] Status judgment logic: Idle state: queue length = 0 or queue length ≤ 2 and historical average processing speed ≥ 1 block / minute; Busy state: queue length ≥ 5 or there is a "complex" task and the processing progress ≤ 50%.

[0073] In some embodiments, distributing each board to be tested to a corresponding maintenance station according to the processing priority of the board to be tested and the current working status of the maintenance station includes: distributing the board to be tested to a maintenance station in the idle state according to the processing priority; and / or, when multiple boards to be tested have the same processing priority, allocating the corresponding workload according to the historical processing efficiency of the maintenance station; and / or, distributing the successfully matched board to be tested to the corresponding maintenance station according to the matching relationship between the production line equipment model corresponding to the production line and the maintenance station model corresponding to the maintenance station.

[0074] Tasks are dynamically distributed based on priority and maintenance station status, taking into account historical efficiency and equipment model matching.

[0075] When a repair station is detected to be idle, the highest priority NG board task will be assigned to that repair station first.

[0076] If multiple boards have the same priority, the workload will be allocated according to the historical average processing speed of the repair station (e.g., the repair station with the faster processing speed will be allocated more tasks).

[0077] Establish a mapping table of "production line equipment model - maintenance station model" (e.g., model A testing equipment corresponds to model A maintenance station) to ensure that NG boards are distributed to stations with corresponding maintenance capabilities, and avoid inefficiency caused by technical mismatch.

[0078] The central server monitors the progress of maintenance station tasks in real time. When the queue length of a certain station exceeds the threshold, the subsequent tasks are automatically redistributed to other idle stations.

[0079] In some embodiments, the method further includes: obtaining false alarm information corresponding to the production line; establishing a production line data model based on the false alarm information; and adjusting the task allocation quantity of the maintenance station based on the average number of false alarms per unit time.

[0080] A data model is established based on false alarm information, and the number of tasks assigned to the maintenance station is adjusted according to the average number of false alarms per unit time.

[0081] The system collects statistics on false alarm data (the number of boards initially judged as NG but re-judged as OK) for each production line over the past 24 hours and calculates the average number of false alarms per unit time (e.g., the hourly false alarm rate).

[0082] Use statistical models (such as linear regression) or machine learning algorithms to analyze the relationship between false alarm rate and task allocation quantity, and establish a predictive model (such as reducing the task allocation quantity of the production line by 20% when the false alarm rate is ≥5% to avoid ineffective processing).

[0083] The dynamic adjustment strategy includes: if the false alarm rate of a certain production line is consistently below the threshold, increase its task allocation to the maintenance station; if the false alarm rate is too high, trigger an early warning and reduce task allocation, guiding priority to handle high-confidence NG boards.

[0084] In some embodiments, in the existing sorting logic of "number of idle layers in the sorting machine + detection time", the urgency level of the defect type of the board to be re-judged is introduced as an intermediate dimension to ensure that high-urgency defects (such as short circuits) are processed first, avoiding production risks caused by sorting delays (such as short-circuited boards flowing into the next process and causing greater losses).

[0085] The system pre-configures the correspondence between fault types and urgency levels (customizable). For example: short circuit fault (urgency level 1, highest): may cause the board to be scrapped and needs to be re-evaluated immediately; open circuit fault (urgency level 2): ​​affects circuit continuity and is given priority; cold solder joint fault (urgency level 3): has low repair cost and is given the second priority; surface scratches (urgency level 4, lowest): do not affect functionality and are dealt with last.

[0086] When the board enters the sorting machine, the initial defect type is automatically recorded (identified by the error code reported by the detection equipment) and associated with the corresponding emergency level.

[0087] The priority sorting logic is optimized by adjusting the sorting rules in the centralized calculation and priority judgment steps as follows: a. First, compare the number of idle layers in the sorting machine (the fewer idle layers, the higher the priority); b. If the number of idle layers is the same, compare the urgency level of the defect type (the smaller the level value, the higher the priority); c. If the urgency level is also the same, then compare the detection time (the earlier the time, the higher the priority).

[0088] The central server distributes tasks according to the adjusted priority queue, ensuring that high-urgency defects (such as short circuits) are prioritized for allocation to idle repair stations, thereby reducing production risks.

[0089] In some embodiments, by establishing skill profiles for repair stations (including operators) and recording information such as the types of defects they are good at, the number of layers on the board, and their historical processing efficiency, when tasks are distributed, the skill profiles are matched with the parameters of the board to be reviewed (such as the type of defect and the number of layers) to improve the review efficiency (e.g., repair stations that are good at handling multi-layer boards will be given priority in handling multi-layer board tasks).

[0090] Skill profiles are created by recording the following information for each repair station through the system: supported production line equipment models, number of board layers that can be processed (e.g., single-layer boards, 4-8 layer multi-layer boards, >8 layer high-layer boards); operator skills are statistically analyzed based on historical data, including the types of problems they excel at (e.g., operator A has a 95% accuracy rate in handling short circuits, far higher than the average of 85%), and historical processing speed (e.g., an average of 2 minutes to handle short circuits on multi-layer boards, lower than the average of 3 minutes). Historical performance: Completion rate and misjudgment rate of the past 30 days (e.g., the misjudgment rate of repair station B is 2%, which is lower than the factory average of 5%).

[0091] Task matching logic: When a task to be reviewed is generated, the system executes the following steps: a. Filter repair stations with matching hardware capabilities (e.g., for multi-layer board tasks, filter repair stations that support multi-layer boards); b. Calculate the skill matching score (example formula): Matching score = (Weight of proficiency in defect type × 0.4) + (Weight of layer processing capability × 0.3) + (Weight of historical efficiency × 0.3) (e.g., "Proficiency in defect type" gets 1 point if matched, otherwise 0 points; "Weight of layer processing capability" gets 1 point if matched, otherwise 0 points; "Historical efficiency" is the ratio of the average time the repair station takes to handle this type of task to the factory's average time, the smaller the ratio, the higher the score); c. Select the repair station with the highest matching score (if the scores are the same, select the currently available repair station). The system updates the skill profile monthly based on the latest review data (e.g., if the operator's accuracy in handling new types of defects improves, adjust their proficiency type) to ensure matching accuracy.

[0092] In some embodiments, by changing the existing fixed inspection method, the inspection interval is dynamically adjusted according to the real-time status of the production line sorting machine (such as the rate of change of the number of NG boards and the rate of reduction of the number of idle layers), so as to reduce the consumption of system resources while ensuring response speed (such as increasing the inspection frequency of busy lines and decreasing the frequency of idle lines).

[0093] The inspection interval configuration allows setting an adjustable range for the inspection interval for each production line (e.g., minimum 2 seconds, maximum 30 seconds), with the initial value set to 5 seconds (consistent with the existing solution).

[0094] Busyness index calculation: The system monitors the following indicators in real time through inspection: NG board quantity change rate: The increase in the number of NG boards in the past minute (e.g., from 5 boards to 15 boards, the change rate is 10 boards / minute). Idle floor reduction rate: The amount of time the number of idle floors has decreased in the past minute (e.g., from 10 floors to 4 floors, the reduction rate is 6 floors / minute).

[0095] Inspection frequency adjustment logic: Set trigger thresholds (e.g., NG board change rate ≥ 5 boards / minute, or idle layer reduction rate ≥ 3 layers / minute): a. When a production line reaches the threshold, shorten the inspection interval (e.g., from 5 seconds to 2 seconds) to increase the monitoring frequency of that line; b. When the indicator recovers to below the threshold, extend the inspection interval (e.g., from 2 seconds back to 5 seconds) to reduce system resource consumption; c. For production lines that have not reached the threshold, maintain the initial 5-second interval.

[0096] In some embodiments, in order to automate the triggering and information synchronization of the NG board retrieval process, when the number of NG boards in the sorting machine reaches a threshold, the system automatically reminds the operator and sends detailed information about the NG boards (such as location and defect type), reducing manual search time (existing solutions require manual information retrieval).

[0097] Threshold setting involves setting a threshold for the number of NG boards for each sorter (e.g., if the total number of sorter layers is 20, the threshold is set to 15 layers). When the number of NG boards reaches 15, board removal is triggered.

[0098] Board Retrieval Trigger and Information Synchronization: a. The system monitors the number of NG boards in the sorting machine through inspection. When the threshold is reached, a board retrieval request is sent to the central server; b. The central server retrieves information on all NG boards in the sorting machine, including unique identifiers (such as QR codes), track positions (such as the 2nd position of the 3rd layer), initial defect type (such as short circuit), and detection equipment number; c. The system synchronously sends the above information to the operator terminal (such as a PDA). The terminal displays a prominent board retrieval reminder (pop-up window + sound alarm) and marks the specific location of the NG board (such as "Sorting Machine A, 2nd position of the 3rd layer, NG board: short circuit fault").

[0099] If the factory is equipped with a robotic arm, the system can send control commands to the robotic arm to automatically remove the NG board based on its track position and place it in the maintenance area. Simultaneously, it sends a signal indicating completion of board removal and updates the NG board count record in the sorting machine. After removal, the system automatically clears the track position record of the NG board in the sorting machine, but retains maintenance station information (such as initial judgment results and re-judgment records) for traceability.

[0100] In some embodiments, by using historical production data (such as the number of NG boards and the number of re-judgment tasks in the past 7 days) to predict the changes in the workload in the next hour, the workstation allocation can be adjusted in advance to avoid uneven workstation workload caused by a sudden increase in workload (the existing solution is real-time allocation, which cannot cope with emergencies).

[0101] The system collects hourly data for each production line over the past 7 days, including the number of NG boards, re-judgment tasks, number of repair station assignments, and processing time. Time series forecasting algorithms (such as the ARIMA model) are used to predict the task volume for each production line in the next hour (for example, it is predicted that the number of NG boards in production line B will reach 20 between 10:00 and 11:00, an increase of 150% compared to the current number).

[0102] Advance scheduling logic: Set a prediction threshold (e.g., the workload increases by 50% compared to the current workload). When a production line reaches the threshold, trigger the following operations: a. Count the number of currently available maintenance stations (e.g., if production line B currently has 2 maintenance stations, and it is predicted that 4 are needed, then 2 more need to be added); b. Transfer available maintenance stations from production lines with less workload (e.g., production line C) and assign them to production line B; c. Remind the administrator to arrange standby operators to production line B (if there are not enough available maintenance stations).

[0103] After the prediction period ends, the system adjusts the parameters of the prediction model (such as the lag order of the ARIMA model) based on the deviation between the actual workload and the predicted value (e.g., 18 blocks actually, 20 blocks predicted, deviation -10%), to improve the accuracy of future predictions.

[0104] In some embodiments, the prior art adopts a single-machine centralized judgment mode. That is, for the inspection section of multiple production lines, after the PCB board is inspected, it flows from the inspection equipment into the sorting machine (also known as NGBUFFER). If manual re-judgment is required, the operator can only re-judge the inspection result of one inspection equipment at a time, and then re-judge and process the boards temporarily stored in another sorting machine.

[0105] After a period of time, the operator needs to remove several actual NG boards. When removing the boards, the operator needs to use a computer to retrieve the repair station information and mark the physical boards as defective.

[0106] The following technical defects exist: 1. Lack of centralized multi-line processing: Current technology only processes boards awaiting judgment from one testing device in the re-judgment stage, failing to comprehensively coordinate the production status of multiple production lines or automatically adjust the re-judgment processing rhythm. The current method simply projects the maintenance station interfaces of each production line to the central control maintenance station via a local area network, without collecting and logically judging the board storage information of the sorting machines. Therefore, on some production lines, due to high production speeds or numerous false alarms, the sorting machine storage bins easily become overflowing with boards.

[0107] 2. Lack of intelligent distribution processing: The allocation method for re-judgment repair stations is fixed, meaning a single repair station is responsible for multiple sorting machines for an extended period. Due to varying board complexity, some repair stations may be relatively unattended due to simpler boards, while others may be overworked due to more complex boards or a heavier workload, sometimes even requiring additional repair stations. This imbalance frequently occurs on-site, affecting re-judgment efficiency.

[0108] 3. Lack of automated repair station retrieval capability: After a period of production, defective boards must be removed for processing. However, the current method requires the operator to manually press the board removal button while simultaneously searching for the repair station record for that board on the repair station computer before proceeding with further operations. This process requires manual confirmation and lacks the ability to automatically identify board removal needs and simultaneously retrieve repair station information, thus impacting work efficiency.

[0109] This embodiment aims to achieve centralized processing of multiple production lines, optimize re-judgment scheduling, solve the problem of uneven workload in the maintenance station, improve the automation level of NG board retrieval, and reduce manual operation.

[0110] The proposed improvements include: 1. Automatic recording: When each board enters the sorting machine after inspection, relevant information (time, number of layers, track position, preliminary judgment result, etc.) is automatically recorded, and some records are automatically cleared when the board leaves the sorting machine.

[0111] 2. Inspection mechanism: Automatically inspect each production line at regular intervals, such as once every 5 seconds, to collect information such as the number of idle layers, the number of OK boards, and the number of NG boards for each sorting machine.

[0112] 3. Centralized Calculation and Priority Determination: After each inspection, the system automatically queues the boards to be processed and determines the priority based on the number of idle layers in each sorting machine. For inspection sections not connected to sorting machines, the system automatically sets the highest priority and sends the results to the central server.

[0113] 4. Intelligent distribution: The central server distributes tasks to one or more maintenance stations, and the specific number can be dynamically adjusted based on the average number of false alarms per unit time.

[0114] The effectiveness of reducing manpower is judged based on actual production conditions. Typically, a factory has 10 production lines, each with 2 testing devices and 2 machines. Under normal production conditions and personnel arrangements, before adding the centralized control software, 8-10 people were needed to handle the verification work. After adding the software, only 2-3 people will be needed to handle it.

[0115] For example, if there are 7 production lines, each with 2 inspection devices at the front and back, originally one person was required per production line to handle the re-inspection work. After adding centralized control software, a pilot test was conducted on 4 of the lines. Only 2 operators are needed to complete the task.

[0116] Traditionally, 10 production lines (each with 2 inspection devices and 2 sorting machines) typically require 8-10 operators to handle the re-judgment work. With the method described in this embodiment, only 2-3 operators are needed to complete the same task. A pilot program was implemented on 7 production lines (each with 2 inspection devices), and after implementing the new solution on 4 of them, only 2 operators were required to fully handle the re-judgment tasks. Intelligent scheduling ensures a balanced workload at the repair stations, preventing some stations from being overworked while others are idle. Automatic identification of NG board retrieval needs and simultaneous retrieval of repair station information improve operational efficiency.

[0117] like Figure 2As shown, the collection and sorting process involves gathering error records to the server when the AOI generates an error, and sorting the errors gathered on the server according to the text information received from AO1. This text is sent from the sorting machine to the AOI (the content of the text is a comparison of the number of idle layers of each sorting machine, with the fewer layers ranked first. When no sorting machine is installed, the AOI automatically generates the fewest texts, which are equivalent to the text of one idle layer generated by the sorting machine. If they are the same, they are sorted according to the time when the AOI generated the error).

[0118] When a repair station becomes available (repair stations 1, 2, and 3 indicate they are available if they have 1 error record, and busy if they have 2), the error message (the one at the top of the sorted error list) is distributed to that repair station for review. (Repair stations 1, 2, and 3 are busy if they have 2 error records, and no further messages are distributed, ensuring that each repair station can display a maximum of two pending records). The server prioritizes distributing error messages to repair stations with fewer pending records.

[0119] Please see Figure 3 As shown, Figure 3 This is a schematic diagram of the structure of a centralized distribution system 200 for production line inspection provided in an embodiment of this application. The centralized distribution system 200 for production line inspection is used to execute the steps of the centralized distribution method for production line inspection shown in the above embodiments. The centralized distribution system 200 for production line inspection can be a single server or a server cluster, or it can be a terminal, such as a handheld terminal, a laptop computer, a wearable device, or a robot.

[0120] like Figure 3 As shown, the centralized distribution system 200 for production line inspection includes: The information acquisition unit 201 is used to acquire sorting information of each board to be tested entering the corresponding sorting machine of the production line, wherein the sorting information includes at least the entry time of the board to be tested; The idle acquisition unit 202 is used to acquire idle information corresponding to the sorting machine, wherein the idle information includes at least the current idle layer number; Priority determination unit 203 is used to determine the processing priority of each board to be tested based on the sorting information corresponding to the board to be tested and the current idle information corresponding to the sorting machine; The status acquisition unit 204 is used to acquire the current working status of each maintenance station on the production line; the current working status includes at least an idle status and a busy status; The board distribution unit 205 is used to distribute each board to be tested to the corresponding maintenance station according to the processing priority of the board to be tested and the current working status of the maintenance station.

[0121] In some embodiments, obtaining the sorting information of each board to be tested entering the corresponding sorting machine of the production line includes: obtaining the entry time, layer number, track position and preliminary judgment result of the board to be tested to generate the sorting information; the preliminary judgment result includes preliminary judgment qualified or preliminary judgment unqualified; wherein, when the board to be tested leaves the sorting machine, the track position is cleared.

[0122] In some embodiments, the method further includes: when a board to be tested that is initially deemed unqualified is detected in the sorting machine, it is identified as a target board; the unique identification information of the target board is obtained, and the board to be tested is removed from the sorting machine according to the unique identification information and sorting information.

[0123] In some embodiments, obtaining the idle information corresponding to the sorting machine includes: collecting the current number of idle layers, the number of preliminarily qualified test boards, and the number of preliminarily unqualified test boards of each sorting machine through a preset timed inspection mechanism, and generating the idle information.

[0124] In some embodiments, determining the processing priority of each board to be tested based on the sorting information corresponding to the board to be tested and the current idle information corresponding to the sorting machine includes: judging based on the number of idle layers of the sorting machine: if the number of idle layers is less, the processing priority of the corresponding board to be tested is higher; if the number of idle layers is the same, sorting according to the entry time of the board to be tested, the earlier the entry time, the higher the priority; for the detection section not connected to the sorting machine, it is set as the highest processing priority.

[0125] In some embodiments, obtaining the current working status of each maintenance station on the production line includes: obtaining the current task queue length, historical average processing speed, and current task difficulty level of the maintenance station; and determining the current working status of the maintenance station based on the current task queue length, historical average processing speed, and current task difficulty level.

[0126] In some embodiments, distributing each board to be tested to a corresponding maintenance station according to the processing priority of the board to be tested and the current working status of the maintenance station includes: distributing the board to be tested to a maintenance station in the idle state according to the processing priority; and / or, when multiple boards to be tested have the same processing priority, allocating the corresponding workload according to the historical processing efficiency of the maintenance station; and / or, distributing the successfully matched board to be tested to the corresponding maintenance station according to the matching relationship between the production line equipment model corresponding to the production line and the maintenance station model corresponding to the maintenance station.

[0127] In some embodiments, the method further includes: obtaining false alarm information corresponding to the production line; establishing a production line data model based on the false alarm information; and adjusting the task allocation quantity of the maintenance station based on the average number of false alarms per unit time.

[0128] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the centralized distribution system and its modules for production line inspection described above can be found in the corresponding contents of the various embodiments of the centralized distribution method for production line inspection described above, and will not be repeated here.

[0129] The aforementioned centralized distribution method for production line inspection can be implemented as a computer program, which can be used in, for example... Figure 3 It runs on the device shown.

[0130] Please see Figure 4 , Figure 4 This is a schematic block diagram of the structure of a computer device provided in an embodiment of this application. The computer device includes a processor, a memory, and a network interface connected via a device bus, wherein the memory may include a storage medium and internal memory.

[0131] The storage medium may store operating devices and computer programs. The computer program includes program instructions that, when executed, cause the processor to perform any centralized distribution method for production line inspection.

[0132] The processor provides computing and control capabilities, supporting the operation of the entire computer device.

[0133] Internal memory provides an environment for the execution of computer programs in non-volatile storage media. When executed by a processor, the computer program can enable the processor to perform any centralized distribution method for production line inspection.

[0134] This network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the terminal to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0135] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0136] In one embodiment, the processor is configured to run a computer program stored in memory to perform the following steps: Obtain the sorting information of each board to be tested as it enters the corresponding sorting machine on the production line. The sorting information includes at least the entry time of the board to be tested. Obtain the idle information corresponding to the sorting machine, wherein the idle information includes at least the current idle layer number; Based on the sorting information corresponding to the board to be tested and the current idle information corresponding to the sorting machine, the processing priority of each board to be tested is determined; Obtain the current working status of each maintenance station on the production line; the current working status includes at least an idle state and a busy state; Based on the processing priority of the board to be tested and the current working status of the repair station, each board to be tested is distributed to the corresponding repair station.

[0137] In some embodiments, obtaining the sorting information of each board to be tested entering the corresponding sorting machine of the production line includes: obtaining the entry time, layer number, track position and preliminary judgment result of the board to be tested to generate the sorting information; the preliminary judgment result includes preliminary judgment qualified or preliminary judgment unqualified; wherein, when the board to be tested leaves the sorting machine, the track position is cleared.

[0138] In some embodiments, the method further includes: when a board to be tested that is initially deemed unqualified is detected in the sorting machine, it is identified as a target board; the unique identification information of the target board is obtained, and the board to be tested is removed from the sorting machine according to the unique identification information and sorting information.

[0139] In some embodiments, obtaining the idle information corresponding to the sorting machine includes: collecting the current number of idle layers, the number of preliminarily qualified test boards, and the number of preliminarily unqualified test boards of each sorting machine through a preset timed inspection mechanism, and generating the idle information.

[0140] In some embodiments, determining the processing priority of each board to be tested based on the sorting information corresponding to the board to be tested and the current idle information corresponding to the sorting machine includes: judging based on the number of idle layers of the sorting machine: if the number of idle layers is less, the processing priority of the corresponding board to be tested is higher; if the number of idle layers is the same, sorting according to the entry time of the board to be tested, the earlier the entry time, the higher the priority; for the detection section not connected to the sorting machine, it is set as the highest processing priority.

[0141] In some embodiments, obtaining the current working status of each maintenance station on the production line includes: obtaining the current task queue length, historical average processing speed, and current task difficulty level of the maintenance station; and determining the current working status of the maintenance station based on the current task queue length, historical average processing speed, and current task difficulty level.

[0142] In some embodiments, distributing each board to be tested to a corresponding maintenance station according to the processing priority of the board to be tested and the current working status of the maintenance station includes: distributing the board to be tested to a maintenance station in the idle state according to the processing priority; and / or, when multiple boards to be tested have the same processing priority, allocating the corresponding workload according to the historical processing efficiency of the maintenance station; and / or, distributing the successfully matched board to be tested to the corresponding maintenance station according to the matching relationship between the production line equipment model corresponding to the production line and the maintenance station model corresponding to the maintenance station.

[0143] In some embodiments, the method further includes: obtaining false alarm information corresponding to the production line; establishing a production line data model based on the false alarm information; and adjusting the task allocation quantity of the maintenance station based on the average number of false alarms per unit time.

[0144] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to implement the steps of the centralized distribution method for production line inspection as provided in any embodiment of this application.

[0145] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiments, such as the hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, SmartMedia Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the computer device.

[0146] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A centralized distribution method for production line inspection, characterized in that, The method includes: Obtain the sorting information of each board to be tested as it enters the corresponding sorting machine on the production line. The sorting information includes at least the entry time of the board to be tested. Obtain the idle information corresponding to the sorting machine, wherein the idle information includes at least the current idle layer number; Based on the sorting information corresponding to the board to be tested and the current idle information corresponding to the sorting machine, the processing priority of each board to be tested is determined; Obtain the current working status of each maintenance station on the production line; the current working status includes at least an idle state and a busy state; Based on the processing priority of the board to be tested and the current working status of the repair station, each board to be tested is distributed to the corresponding repair station.

2. The method according to claim 1, characterized in that, The step of obtaining the sorting information of each board to be tested as it enters the corresponding sorting machine on the production line includes: The sorting information is generated by obtaining the entry time, layer number, track position, and preliminary judgment result of the board to be tested; the preliminary judgment result includes preliminary judgment of pass or preliminary judgment of fail. Specifically, when the board to be tested leaves the sorting machine, the track position is cleared.

3. The method according to claim 2, characterized in that, The method further includes: When a board that initially failed to pass inspection is detected in the sorting machine, it is confirmed as the target board; Obtain the unique identifier information of the target board, and take the board to be tested out of the sorting machine according to the unique identifier information and sorting information.

4. The method according to claim 1, characterized in that, The step of obtaining the idle information corresponding to the sorting machine includes: The idle information is generated by collecting the current number of idle layers, the number of boards that have passed the initial test and the number of boards that have failed the initial test from each sorting machine through a preset timed inspection mechanism.

5. The method according to claim 1, characterized in that, The step of determining the processing priority of each board to be tested based on the sorting information corresponding to the board to be tested and the current idle information corresponding to the sorting machine includes: The processing priority of the board to be tested is determined based on the number of idle layers of the sorting machine: if the number of idle layers is less, the processing priority of the board to be tested is higher; if the number of idle layers is the same, the boards to be tested are sorted according to their entry time, and the earlier the entry time, the higher the priority; for the testing section that is not connected to the sorting machine, it is set to the highest processing priority.

6. The method according to claim 1, characterized in that, The process of obtaining the current working status of each maintenance station on the production line includes: Get the current task queue length, historical average processing speed, and current task difficulty level for the corresponding repair station; The current working status of the maintenance station is determined based on the current task queue length, historical average processing speed, and current task difficulty level.

7. The method according to claim 1, characterized in that, The step of distributing each board to be tested to a corresponding repair station based on the processing priority of the board and the current working status of the repair station includes: The boards to be tested will be distributed to service stations that are in an idle state according to the processing priority; and / or, When multiple boards to be tested have the same processing priority, the corresponding workload is allocated based on the historical processing efficiency of the repair station; and / or, Based on the matching relationship between the production line equipment model and the repair station model, the successfully matched boards to be tested are distributed to the corresponding repair stations.

8. The method according to claim 1, characterized in that, The method further includes: Obtain the false alarm information corresponding to the production line; A production line data model is established based on the false alarm information, and the task allocation quantity of the maintenance station is adjusted according to the average number of false alarms per unit time.

9. A computer device, characterized in that, The computer device includes a memory and a processor; The memory is used to store computer programs; The processor is configured to execute the computer program and, in executing the computer program, implement the method as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, causes the processor to implement the method as described in any one of claims 1 to 8.

Images