Method and apparatus for processing production equipment performance data

Abstract

The application discloses a production equipment efficiency data processing method and device, comprising: acquiring state signals and action signals of a target equipment in a target statistical period, determining running time information and downtime information, and determining physical action count information; acquiring manually reported production data; acquiring an effective output coefficient of the target equipment; calculating cross-modal verification deviation based on the physical action count information, the manually reported production data and the effective output coefficient; performing corresponding data quality control processing according to a judgment result of the cross-modal verification deviation; and generating equipment efficiency evaluation data. The application performs cross-modal verification on physical action counting and manual reporting, intercepts false reporting and output mismatching abnormalities before data storage, moves data management from post-checking to source blocking, and ensures that only credible data enters OEE statistics, thereby effectively solving the technical problem of efficiency evaluation distortion caused by disconnection between physical data and manual semantic data.

Description

Technical Field

[0001] This application relates to the field of intelligent manufacturing, and in particular to a method and apparatus for processing production equipment performance data. Background Technology

[0002] OEE (Overall Equipment Effectiveness) is an important indicator for manufacturing companies to measure equipment utilization, production cycle time loss, and quality loss. A typical OEE is usually determined by three dimensions: time utilization factor, performance factor, and quality factor. Traditional manufacturing execution systems or OEE statistical systems typically calculate OEE using equipment operating status, manual work reports, quality data, and abnormal downtime records.

[0003] However, in actual factory settings, OEE statistics often suffer from a disconnect between physical data and manually entered semantic data. On one hand, operators can asynchronously input the number of good products, defective products, or reasons for abnormalities at the terminal; on the other hand, the actual physical actions at the equipment's underlying level come from signals such as programmable logic controllers (PLCs), limit switches, sensors, tri-color lights, motor enable signals, or spindle movements. If the system relies solely on manual reporting or a single operating level, the following technical problems are likely to occur: First, manually reported output may exceed the physical limits of equipment movement. Operators can enter output data exceeding the actual production capacity of the equipment at the terminal. Traditional systems only perform format or threshold checks on the reported data at the application layer, making it difficult to detect logical incompatibilities between manually reported output and the actual physical movements of the equipment in a timely manner. Once false output data has been written into the database, subsequent verification will lead to distortions in financial settlements, capacity analysis, and overall equipment efficiency reports.

[0004] Second, it is difficult to identify empty cycles and invalid actions of the equipment. The actual number of equipment actions may far exceed the theoretical number of cycles corresponding to the reported output, which may be caused by empty cycles, trial runs, unregistered output, or invalid actions. Traditional systems lack an automatic identification mechanism for such empty cycle losses, resulting in inaccurate calculation of the performance utilization factor in the overall equipment efficiency.

[0005] Third, it is difficult to detect when equipment is running but lacks effective physical action. Traditional equipment comprehensive efficiency statistics systems usually equate equipment operating level or green light status with effective production time. However, equipment being running does not necessarily mean that it has completed effective physical action; there may be situations of "green light idling" or zero-load operation.

[0006] Fourth, the lack of a mandatory binding mechanism for abnormal downtime causes leads to inaccurate root cause analysis. Traditional systems typically record the time of abnormal downtime, but the causes often rely on voluntary selection by personnel or post-event recording. The lack of a mandatory binding between abnormal downtime time segments and cause codes results in unreliable downtime root cause analysis data.

[0007] Therefore, a technical solution is needed that can perform cross-modal verification of underlying physical motion data, manual reporting semantic data, and anomaly attribution data. This solution can perform quality control on abnormal data at the data writing source, identify empty loops and idling losses during the statistical process, and force attribution binding of abnormal downtime, thereby improving the objectivity, real-time performance, and reliability of equipment performance statistics. Summary of the Invention

[0008] This application provides a method and apparatus for processing production equipment performance data.

[0009] According to a first aspect of this application, this application provides a method for processing production equipment performance data, including: Acquire the status signal and action signal of the target device within the target statistical period, and determine the running time information and downtime information of the target device based on the status signal, and determine the physical action count information corresponding to the target device based on the action signal; Obtain manually reported production data corresponding to the target statistical period; Obtain the effective output coefficient corresponding to a single physical action of the target equipment under the current production task; Based on the physical motion count information, the manually reported production data, and the effective output coefficient, calculate the cross-modal verification deviation; Based on the determination result of the cross-modal verification deviation, the corresponding data quality control process is performed, which includes marking the manually reported production data as valid, invalid, or pending review. Based on the physical motion count information after the data quality control processing, the manually reported production data, and the effective output coefficient, equipment efficiency evaluation data is generated.

[0010] In the method involved in this application, the step of performing corresponding data quality control processing based on the determination result of the cross-modal verification deviation includes: When the cross-modal verification deviation determination result is that the manually reported production data exceeds the upper limit of the equipment physical action or the excessive reporting deviation is greater than the first tolerance threshold, the manually reported production data is marked as invalid, and the data is directly intercepted before the manually reported production data is submitted to the database, and its write operation to the trusted production dataset is canceled. When the cross-modal verification deviation is determined to be an action-output mismatch or an action redundancy deviation greater than the second tolerance threshold, the non-serious abnormal data of the action redundancy is marked as pending review. The manually reported production data and its corresponding equipment action evidence index are diverted and stored in an independent pending review data area that is completely logically isolated from the trusted production dataset. The data submission link is not directly blocked, but the data is marked as temporarily not participating in the calculation of equipment performance indicators. It will be synchronized to the trusted production dataset after the manual review is completed and confirmed to be valid. When the cross-modal verification deviation is within the preset effective tolerance range, the manually reported production data is marked as valid. The manually reported production data, verification results, and corresponding equipment action evidence index are atomically written into the trusted production dataset as the same submission unit. The transaction rollback operation is triggered only when a write failure occurs during the submission process.

[0011] The methods involved in this application also include: Obtain the device identifier of the target device, the task identifier of the current production task, and the effective start and end times of the current production task; Based on the device identifier, the task identifier, and the valid start and end times, a verification time window is determined; The physical action count information, the manually reported production data, and the verification time window are bound together.

[0012] The method involved in this application, wherein acquiring the status signal and action signal of the target device within the target statistical period, determining the running time information and downtime information of the target device based on the status signal, and determining the physical action count information corresponding to the target device based on the action signal, includes: Multiple action pulses with an adjacent time difference less than the preset de-jitter duration are merged into a single candidate action event; Duplicate candidate action events are deleted based on the signal source identifier and event sequence number of the candidate action event; The candidate action events that occur in maintenance, jogging, trial, changeover, and unbound production task states are excluded to obtain the set of valid action events; The physical action count information is determined based on the number of valid action events in the set of valid action events.

[0013] In the method involved in this application, obtaining the effective output coefficient corresponding to a single physical action of the target equipment under the current production task includes: The effective output coefficient corresponding to the single physical action is determined based on the rated number of cavities, the effective cavity mask, and the product conversion factor corresponding to the occurrence of the effective action event, and satisfies the following: kᵢ =Σ_{j=1}^{Cᵢ}aᵢⱼrⱼ, Where, kᵢ is the effective output coefficient of the i-th action, Cᵢ is the rated number of cavities corresponding to the i-th effective action, aᵢⱼ indicates whether the j-th cavity is in an effective production state when the effective action occurs, and rⱼ is the product conversion coefficient corresponding to the j-th cavity. When a cavity is discontinued, a product is switched, or a task parameter version is changed during the production process, the effective output coefficient corresponding to the subsequent effective actions is calculated using the changed parameters from the time of the change.

[0014] In the method involved in this application, the cross-modal verification deviation includes reporting excess deviation and action redundancy deviation; The method further includes: Based on the physical action count information N_IO and the effective output coefficient kᵢ, the theoretical output of the equipment is determined as Qp = Σ_{i=1}^{N_IO} kᵢ; Based on the total output Qr in the manually reported production data and the theoretical output Qp of the equipment, calculate the output deviation D = Qr − Qp; When the output deviation D is greater than the first tolerance threshold, it is determined that the manually reported production data has the excessive reporting deviation, and the manually reported production data exceeds the upper limit of the physical movement of the equipment. When the output deviation D is less than the negative second tolerance threshold, it is determined that the target equipment has the motion redundancy deviation, and the theoretical output of the equipment is significantly higher than the manually reported production data.

[0015] The method involved in this application also includes an idling loss identification step: Obtain the standard motion cycle duration of the current production task; The theoretical number of actions is calculated based on the running time information and the standard action cycle duration. The idle deviation is calculated based on the theoretical number of actions and the physical action count information. When the idling deviation is greater than the third tolerance threshold, it is determined that the target device is in operation but lacks effective physical action support, and an idling loss record is generated or a control command is issued.

[0016] In the method involved in this application, the data quality control processing further includes a quality closed-loop verification step: Calculate the difference between the total output in the manually reported production data and the expected total output calculated based on the physical motion count information; When the absolute value of the difference is greater than the quality statistical tolerance threshold, the generation of quality indicators or equipment performance evaluation data within the current statistical period is blocked.

[0017] The method involved in this application also includes an abnormal shutdown attribution step: When an abnormal shutdown event of the target device is detected, the human-machine interaction terminal is controlled to enter a forced attribution interaction state. In the forced attribution interaction state, users are restricted from continuing to perform subsequent work reporting, production scheduling confirmation or abnormal case closure operations until a valid abnormal reason information is received; The anomaly cause information is bound to the time information of the abnormal downtime event, the device identifier of the target device, and the task identifier of the current production task to generate an anomaly attribution record.

[0018] According to a second aspect of this application, this application provides a production equipment performance data processing device, comprising: The signal acquisition module is used to acquire the status signals and action signals of the target device within the target statistical period, and to determine the running time information and downtime information of the target device based on the status signals, and to determine the physical action count information corresponding to the target device based on the action signals. The work report data acquisition module is used to acquire manually reported production data corresponding to the target statistical period; The theoretical output acquisition module is used to acquire the effective output coefficient corresponding to a single physical action of the target equipment under the current production task; The deviation calculation module is used to calculate the cross-modal verification deviation based on the physical action count information, the manually reported production data, and the effective output coefficient. The data quality control module is used to perform corresponding data quality control processing based on the determination result of the cross-modal verification deviation. The efficiency data generation module is used to generate equipment efficiency evaluation data based on the physical motion count information after the data quality control processing, the manually reported production data, and the effective output coefficient.

[0019] Due to the adoption of the above technical solutions, the beneficial effects of this application are as follows: This application provides a method and apparatus for processing production equipment performance data, including: acquiring status signals and action signals of a target device within a target statistical period, and determining the running time and downtime information of the target device based on the status signals; determining the physical action count information corresponding to the target device based on the action signals; acquiring manually reported production data corresponding to the target statistical period; acquiring the effective output coefficient corresponding to a single physical action of the target device under the current production task; calculating the cross-modal verification deviation based on the physical action count information, manually reported production data, and effective output coefficient; performing corresponding data quality control processing according to the determination result of the cross-modal verification deviation, the data quality control processing including marking the manually reported production data as valid, invalid, or pending review; and generating equipment performance evaluation data based on the physical action count information, manually reported production data, and effective output coefficient after data quality control processing. This application, through cross-modal verification of physical action counts and manually reported work, intercepts false work reports and output mismatch anomalies before data is entered into the database, effectively solving the technical problem of performance evaluation distortion caused by the disconnect between physical data and manually reported semantic data. Attached Figure Description

[0020] Figure 1 A flowchart illustrating one implementation of the method provided in this application; Figure 2 A flowchart of a sub-step in one implementation of the method provided in this application embodiment; Figure 3 A flowchart of another implementation of the method provided in the embodiments of this application; Figure 4 A flowchart of another embodiment of the method provided in this application; Figure 5 This is a schematic diagram of the functional modules of the apparatus provided in one embodiment of the present application. Detailed Implementation

[0021] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of this application. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to this application are not shown or described in the specification. This is to avoid obscuring the core parts of this application with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0022] Furthermore, the features, operations, or characteristics described in the specification can be combined in any suitable manner to form various embodiments. At the same time, the steps or actions in the method description can be rearranged or adjusted in a manner obvious to those skilled in the art. Therefore, the various orders in the specification and drawings are only for the clear description of a particular embodiment and do not imply a necessary order, unless otherwise stated that a particular order must be followed.

[0023] The serial numbers assigned to components in this article, such as "first" and "second", are used only to distinguish the objects being described and have no sequential or technical meaning.

[0024] Furthermore, the technical features and solutions described herein can be combined in any suitable manner in one or more embodiments. It will be readily understood by those skilled in the art that the steps or order of operations related to the embodiments provided herein can also be changed. Therefore, any order in the drawings and embodiments is for illustrative purposes only and does not imply a requirement to follow a particular order, unless explicitly stated otherwise.

[0025] In this application, cross-modal verification refers to a data processing method that maps equipment physical signals (status signals, action signals) and human-generated semantic data (output quantity, anomaly cause) to the same physical cycle scale for consistency verification. Physical action count information refers to the count information obtained by the edge acquisition gateway based on the accumulated equipment action signals, reflecting the actual number of actions completed by the equipment's actuators. Manually reported production data refers to business data such as output quantity and work order information entered by operators through a human-machine interface terminal. Data quality control processing refers to control operations performed based on the judgment results of cross-modal verification deviations, including but not limited to writing rejection or transaction rollback of manually reported production data exceeding the physical action limit, generating performance loss records for situations where action-output mismatch exists, and triggering mandatory attribution interaction for abnormal downtime events. Mandatory attribution interaction state refers to the operation-restricted state entered by the human-machine interface terminal after detecting an abnormal downtime event. In this state, operators must input valid anomaly cause information before continuing subsequent operations. Anomaly attribution identifier refers to a unique identifier generated after associating anomaly cause information, equipment identifier, statistical period, and the start and end times of the abnormal downtime event.

[0026] Idle loss: refers to the efficiency loss that occurs when equipment is in operation but the actual number of physical actions within the statistical period is significantly less than the theoretical number of actions corresponding to the operating time.

[0027] like Figure 1 As shown in the embodiments of this application, one implementation of the production equipment performance data processing method may include the following steps: Step 110: Obtain the status signal and action signal of the target device within the target statistical period, and determine the running time information and downtime information of the target device based on the status signal, and determine the physical action count information corresponding to the target device based on the action signal.

[0028] In some embodiments, the target equipment may be an injection molding machine, a punch press, a CNC machine tool, a die casting machine, a packaging machine, an assembly line station, or other production equipment with periodic movements.

[0029] The edge acquisition gateway reads and acquires status signals from the target device through an input / output acquisition card or PLC register. Status signals may include operation indication signals (such as green light level), fault indication signals (such as red light level), standby indication signals (such as yellow light level), operation enable signals, emergency stop signals, or device online signals. The edge acquisition gateway accumulates running time and downtime information based on the duration of the status signal levels. Specifically, when a valid operation indication signal is acquired, the running time is accumulated; when a valid fault indication signal is acquired, the downtime is accumulated.

[0030] The edge acquisition gateway can also acquire action signals from the target device through input / output acquisition cards, PLC register readings, or sensor triggers. Action signals can include mold closing completion pulses, stamping stroke pulses, spindle machining completion pulses, workpiece arrival switch signals, output count signals, or controller register count values. The edge acquisition gateway accumulates the number of action signal occurrences within the target statistical period to obtain the physical action count information corresponding to the target device.

[0031] This step utilizes an edge acquisition gateway deployed on the device side to read status and action signals emitted by the target device's controller or sensors in real time. Status signals, such as tri-color light levels and operation enable signals, indicate whether the device is running or stopped. Action signals, such as mold closing pulses and stamping stroke pulses, indicate whether the device's actuators have completed actual physical actions. The edge acquisition gateway accumulates the effective duration of status signals within the target statistical period to obtain running and stopping time information. Simultaneously, it accumulates the number of triggers of action signals to obtain the corresponding physical action count information for the device. This step directly acquires objective physical data from the device's underlying layers, serving as the physical benchmark for all subsequent verifications and providing factual evidence for cross-modal verification.

[0032] like Figure 2 As shown, step 110 may specifically include: Step 1101: Merge multiple action pulses with an adjacent time difference less than the preset de-jitter duration into a single candidate action event; Step 1102: Delete duplicate candidate action events based on the signal source identifier and event sequence number of the candidate action events; Step 1103: Eliminate candidate action events that occur in maintenance, jogging, trial, changeover, and unbound production task states to obtain a set of valid action events; Step 1104: Determine the physical action count information based on the number of valid action events in the valid action event set.

[0033] Step 120: Obtain manually reported production data corresponding to the target statistical period.

[0034] The human-machine interface terminal receives manually reported production data input by operators, corresponding to the target statistical period. The manually reported production data includes at least output quantity information, such as the number of good products and the number of defective products, and may further include work order identifiers, equipment identifiers, personnel identifiers, reasons for scrapping, or work order status.

[0035] This step receives manually reported production data corresponding to the same target statistical period, entered by operators through human-computer interaction terminals such as workstation computers, industrial tablets, or mobile terminals. The manually reported production data must at least include output quantity information, such as the number of good and defective products. It may also further include business semantic information such as work order identifiers, personnel identifiers, and reasons for scrapping. Obtaining manually reported work data carrying management semantics stems from the subjective reporting of operators and may contain errors or falsifications. Therefore, subsequent steps require cross-validation with objective physical data to determine its reliability.

[0036] Step 130: Obtain the effective output coefficient corresponding to a single physical action of the target equipment under the current production task.

[0037] This step obtains the effective output coefficient corresponding to a single physical action of the target equipment under the current production task from the process configuration table or the work order parameters of the MES system. That is, how many products should the equipment theoretically produce after completing one effective physical action? For example, a four-cavity injection molding machine can produce 4 products after one mold closing. This parameter is the key conversion coefficient connecting physical modal data and human semantic data. Without this parameter, it is impossible to compare how many times the equipment has acted and how many products have been reported by the human on the same scale, and cross-modal verification cannot be carried out.

[0038] In some embodiments, step 130 may specifically include: The effective output coefficient for a single physical action is determined based on the rated number of cavities, the effective cavity mask, and the product conversion factor corresponding to the occurrence of the effective action event, and satisfies the following: kᵢ =Σ_{j=1}^{Cᵢ}aᵢⱼrⱼ, Where, kᵢ is the effective output coefficient of the i-th action, Cᵢ is the rated number of cavities corresponding to the i-th effective action, aᵢⱼ indicates whether the j-th cavity is in an effective production state when the effective action occurs, and rⱼ is the product conversion coefficient corresponding to the j-th cavity. When a cavity is discontinued, a product is switched, or a task parameter version is changed during the production process, the effective output coefficient corresponding to the subsequent effective actions is calculated using the changed parameters from the time of the change.

[0039] Step 140: Calculate the cross-modal verification deviation based on the physical motion count information of the target equipment, manually reported production data, and effective output coefficient.

[0040] In some embodiments, cross-modal verification deviations include reporting excess deviations and action redundancy deviations; The production equipment performance data processing method provided in this application embodiment may further include: Based on the physical action count information N_IO and the effective output coefficient kᵢ, determine the theoretical output of the equipment Qp = Σ_{i=1}^{N_IO} kᵢ; based on the total output Qr in the manually reported production data and the theoretical output of the equipment Qp, calculate the output deviation D = Qr − Qp; When the output deviation D is greater than the first tolerance threshold, it is determined that there is an overreporting deviation in the manually reported production data, and the manually reported production data exceeds the upper limit of the physical movement of the equipment. When the output deviation D is less than the negative second tolerance threshold, it is determined that the target equipment has motion redundancy deviation, and the theoretical output of the equipment is significantly higher than the manually reported production data. The standard for the theoretical output of the equipment to be significantly higher than the manually reported production data can be set based on experience.

[0041] This step integrates the set of effective action events obtained in step 110, the manually reported production data obtained in step 120, and the effective output coefficients of each effective action event obtained in step 130. First, the theoretical output of the target equipment within the target statistical period is calculated as Qp = Σ_{i=1}^{N_IO} kᵢ. Then, the output deviation D = Qr − Qp between the total reported output and the theoretical output is calculated. When D > 0, it indicates that the reported output is higher than the theoretical output, corresponding to an over-reporting deviation; when D < 0, it indicates that the theoretical output is higher than the reported output, corresponding to an action redundancy deviation. This step unifies the physical world (theoretical equipment output) and the business world (manually reported output) onto the common scale of "output quantity" for quantitative comparison, objectively determining whether there is a logical conflict between manual data and physical facts.

[0042] Step 150: Based on the determination result of cross-modal verification deviation, perform the corresponding data quality control processing.

[0043] In some embodiments, data quality control processing may include marking manually reported production data as valid, invalid, or pending review. Appropriate data quality control processing is then performed: severe anomalies (excessive reporting) are directly intercepted before being entered into the database; non-severe anomalies (action-output mismatch) are diverted to an independent pending review area for manual confirmation, thus achieving differentiated data quality control.

[0044] Step 150 may specifically include: When the cross-modal verification deviation is determined to be that the manually reported production data exceeds the upper limit of the equipment physical action or the excessive reporting deviation is greater than the first tolerance threshold, the manually reported production data is marked as invalid. Before the manually reported production data is submitted to the database, the data is directly intercepted and its write operation to the trusted production dataset is canceled. When the cross-modal verification deviation is determined to be a mismatch between action and output or an action redundancy deviation greater than the second tolerance threshold, the non-serious abnormal data of the action redundancy is marked as pending review. The manually reported production data and its corresponding equipment action evidence index are diverted and stored in an independent pending review data area that is completely logically isolated from the trusted production dataset. The data submission link is not directly blocked, but the data is marked as temporarily not participating in the calculation of equipment performance indicators. It will be synchronized to the trusted production dataset after the manual review is completed and confirmed to be valid. When the cross-modal verification deviation is within the preset effective tolerance range, the manually reported production data is marked as valid. The manually reported production data, verification results, and corresponding equipment action evidence index are atomically written into the trusted production dataset as the same submission unit. The transaction rollback operation is triggered only when a write failure occurs during the submission process.

[0045] A trusted production dataset refers to a persistent storage area used solely for storing production data that has been verified as valid through cross-modal validation and can be used in equipment performance evaluation calculations. The data in this dataset has the following characteristics: it has passed dual-source consistency verification of physical action counts and manual work reports; its equipment identifiers, production task identifiers, and verification time windows are all bound and traceable; and it includes a corresponding equipment action evidence index at the time of writing, supporting subsequent review and auditing. In contrast to the trusted production dataset, this application may also set up a logically isolated pending verification data area to store data that has not yet passed verification and is awaiting manual confirmation. Data in this area does not directly participate in the calculation of equipment performance indicators.

[0046] Based on the judgment result pointed to by the output deviation D calculated in step 140, this step performs corresponding data quality control processing: if the deviation D is greater than the first tolerance threshold, it means that the reported output exceeds the theoretical output of the equipment, so write rejection or transaction rollback is performed before the data is written to the database; if the absolute value of the deviation D is greater than the second tolerance threshold, it means that the theoretical output of the equipment exceeds the reported output, so an efficiency loss record is generated.

[0047] Step 160: Based on the physical motion count information after data quality control processing, manually reported production data, and effective output coefficient, generate equipment efficiency evaluation data.

[0048] Equipment performance evaluation data may include at least one of the following: time utilization factor, performance factor, quality factor, work reporting anomaly records, empty cycle loss records, idling loss records, and anomaly attribution analysis data. The production equipment performance data processing method provided in this application embodiment may further include: Aggregate anomaly attribution data by dimension, such as cause of anomaly, equipment, production task, shift, personnel, or material, to generate performance loss analysis reports or root cause location reports.

[0049] This step integrates the physical motion data (such as running time, downtime, and physical motion counts) processed by the data quality control in step 150, the reported production data (such as the number of good products and the number of defective products), and the anomaly attribution data (such as the binding records of anomaly causes and downtime), to generate equipment performance evaluation data. This includes outputs such as time utilization factors, performance factors, quality factors, and records of idle cycle loss, idling loss, and anomaly attribution analysis. This step ensures that only reliable data that has undergone physical verification, quality control processing, and mandatory attribution binding can enter the performance evaluation stage, guaranteeing that the final report accurately reflects the objective situation on the production floor and provides a reliable basis for management decisions.

[0050] The production equipment performance data processing method provided in this application embodiment may further include: Obtain the device identifier of the target device, the task identifier of the current production task, and the effective start and end times of the current production task; The verification time window is determined based on the device identifier, task identifier, and valid start and end times; The physical motion count information, manually reported production data, and verification time windows are linked.

[0051] The production equipment performance data processing method provided in this application embodiment may further include an abnormal shutdown attribution step independent of the three-state data splitting logic: When an abnormal shutdown event of the target device is detected, the human-machine interface terminal is controlled to enter a forced attribution interaction state. In the forced attribution interaction state, users are restricted from continuing to perform subsequent work reporting, production scheduling confirmation or abnormal case closure operations until a valid abnormal reason information is received. The anomaly cause information is bound to the downtime event time information, the device identifier of the target device, and the task identifier of the current production task to generate an anomaly attribution record. In some embodiments, this anomaly attribution record can be synchronized to a trusted production dataset as supplementary data for subsequent root cause analysis of equipment performance loss.

[0052] like Figure 3 As shown, between steps 140 and 150, an idling loss identification step may also be included: Step 310: Obtain the standard motion cycle duration of the current production task; Step 320: Calculate the theoretical number of actions based on the runtime information and the standard action cycle duration; Step 330: Calculate the idling deviation based on the theoretical number of movements and the physical movement count information; Step 340: When the idling deviation is greater than the third tolerance threshold, it is determined that the target equipment is in operation but lacks effective physical action support, and an idling loss record is generated or a control command is issued.

[0053] like Figure 4 As shown, a quality data closed-loop verification step may also be included between steps 140 and 150: Step 410: Calculate the difference between the total output in the manually reported production data and the expected total output calculated based on the physical motion count information; Step 420: When the absolute value of the difference exceeds the quality statistical tolerance threshold, the generation of quality indicators or equipment performance evaluation data for the current statistical period is blocked. The quality statistical tolerance threshold can be set as needed.

[0054] like Figure 5 As shown in the embodiment of this application, the production equipment efficiency data processing device may include, in one embodiment, a signal acquisition module 510, a work report data acquisition module 520, a theoretical output acquisition module 530, a deviation calculation module 540, a data quality control module 550, and an efficiency data generation module 560.

[0055] The signal acquisition module 510 is used to acquire the status signals and action signals of the target device within the target statistical period, and to determine the running time information and downtime information of the target device based on the status signals, and to determine the physical action count information of the target device based on the action signals. The work data acquisition module 520 is used to acquire manually reported production data corresponding to the target statistical period; The theoretical output acquisition module 530 is used to acquire the effective output coefficient corresponding to a single physical action of the target equipment under the current production task; The deviation calculation module 540 is used to calculate the cross-modal verification deviation based on physical motion count information, manually reported production data and effective output coefficient. The data quality control module 550 is used to perform corresponding data quality control processing based on the judgment result of cross-modal verification deviation. The efficiency data generation module 560 is used to generate equipment efficiency evaluation data based on physical motion count information after data quality control processing, manually reported production data, and effective output coefficient.

[0056] The electronic device provided in this application includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the production equipment performance data processing method described above.

[0057] Those skilled in the art will understand that all or part of the steps of the various methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: read-only memory, random access memory, disk or optical disk, etc.

[0058] The above description, in conjunction with specific embodiments, provides a further detailed explanation of this application and should not be construed as limiting the specific implementation of this application to these descriptions. Those skilled in the art to which this application pertains can make several simple deductions or substitutions without departing from the concept of this application.

Claims

1. A method for processing production equipment efficiency data, characterized in that, include: Acquire the status signal and action signal of the target device within the target statistical period, and determine the running time information and downtime information of the target device based on the status signal, and determine the physical action count information corresponding to the target device based on the action signal; Obtain manually reported production data corresponding to the target statistical period; Obtain the effective output coefficient corresponding to a single physical action of the target equipment under the current production task; Based on the physical motion count information, the manually reported production data, and the effective output coefficient, calculate the cross-modal verification deviation; Based on the determination result of the cross-modal verification deviation, the corresponding data quality control process is performed, which includes marking the manually reported production data as valid, invalid, or pending review. Based on the physical motion count information after the data quality control processing, the manually reported production data, and the effective output coefficient, equipment efficiency evaluation data is generated.

2. The method according to claim 1, characterized in that, The step of performing corresponding data quality control processing based on the determination result of the cross-modal verification deviation includes: When the cross-modal verification deviation determination result is that the manually reported production data exceeds the upper limit of the equipment physical action or the excessive reporting deviation is greater than the first tolerance threshold, the manually reported production data is marked as invalid, and the data is directly intercepted before the manually reported production data is submitted to the database, and its write operation to the trusted production dataset is canceled. When the cross-modal verification deviation is determined to be an action-output mismatch or an action redundancy deviation greater than the second tolerance threshold, the non-serious abnormal data of the action redundancy is marked as pending review. The manually reported production data and its corresponding equipment action evidence index are diverted and stored in an independent pending review data area that is completely logically isolated from the trusted production dataset. The data submission link is not directly blocked, but the data is marked as temporarily not participating in the calculation of equipment performance indicators. It will be synchronized to the trusted production dataset after the manual review is completed and confirmed to be valid. When the cross-modal verification deviation is within the preset effective tolerance range, the manually reported production data is marked as valid. The manually reported production data, verification results, and corresponding equipment action evidence index are atomically written into the trusted production dataset as the same submission unit. The transaction rollback operation is triggered only when a write failure occurs during the submission process.

3. The method according to claim 1, characterized in that, Also includes: Obtain the device identifier of the target device, the task identifier of the current production task, and the effective start and end times of the current production task; Based on the device identifier, the task identifier, and the valid start and end times, a verification time window is determined; The physical action count information, the manually reported production data, and the verification time window are bound together.

4. The method according to claim 1, characterized in that, The process of acquiring the status signal and action signal of the target device within the target statistical period, determining the running time information and downtime information of the target device based on the status signal, and determining the physical action count information corresponding to the target device based on the action signal includes: Multiple action pulses with an adjacent time difference less than the preset de-jitter duration are merged into a single candidate action event; Duplicate candidate action events are deleted based on the signal source identifier and event sequence number of the candidate action event; The candidate action events that occur in maintenance, jogging, trial, changeover, and unbound production task states are excluded to obtain the set of valid action events; The physical action count information is determined based on the number of valid action events in the set of valid action events.

5. The method according to claim 1, characterized in that, The step of obtaining the effective output coefficient corresponding to a single physical action of the target equipment under the current production task includes: The effective output coefficient corresponding to the single physical action is determined based on the rated number of cavities, the effective cavity mask, and the product conversion factor corresponding to the occurrence of the effective action event, and satisfies the following: kᵢ =Σ_{j=1}^{Cᵢ}aᵢⱼrⱼ, Where, kᵢ is the effective output coefficient of the i-th action, Cᵢ is the rated number of cavities corresponding to the i-th effective action, aᵢⱼ indicates whether the j-th cavity is in an effective production state when the effective action occurs, and rⱼ is the product conversion coefficient corresponding to the j-th cavity. When a cavity is discontinued, a product is switched, or a task parameter version is changed during the production process, the effective output coefficient corresponding to the subsequent effective actions is calculated using the changed parameters from the time of the change.

6. The method according to claim 1, characterized in that, The cross-modal verification deviation includes excessive reporting deviation and motion redundancy deviation; The method further includes: Based on the physical action count information N_IO and the effective output coefficient kᵢ, the theoretical output of the equipment is determined as Qp = Σ_{i=1}^{N_IO} kᵢ; Based on the total output Qr in the manually reported production data and the theoretical output Qp of the equipment, calculate the output deviation D = Qr − Qp; When the output deviation D is greater than the first tolerance threshold, it is determined that there is an excessive reporting deviation in the manually reported production data, and the manually reported production data exceeds the upper limit of the physical movement of the equipment. When the output deviation D is less than the negative second tolerance threshold, it is determined that the target equipment has the motion redundancy deviation, and the theoretical output of the equipment is significantly higher than the manually reported production data.

7. The method according to claim 1, characterized in that, It also includes a step for identifying idling loss: Obtain the standard motion cycle duration of the current production task; The theoretical number of actions is calculated based on the running time information and the standard action cycle duration. The idle deviation is calculated based on the theoretical number of actions and the physical action count information. When the idling deviation is greater than the third tolerance threshold, it is determined that the target device is in operation but lacks effective physical action support, and an idling loss record is generated or a control command is issued.

8. The method according to claim 1, characterized in that, The data quality control process also includes a quality closed-loop verification step: Calculate the difference between the total output in the manually reported production data and the expected total output calculated based on the physical motion count information; When the absolute value of the difference is greater than the quality statistical tolerance threshold, the generation of quality indicators or equipment performance evaluation data within the current statistical period is blocked.

9. The method according to claim 1, characterized in that, It also includes abnormal shutdown attribution steps: When an abnormal shutdown event of the target device is detected, the human-machine interaction terminal is controlled to enter a forced attribution interaction state. In the forced attribution interaction state, users are restricted from continuing to perform subsequent work reporting, production scheduling confirmation or abnormal case closure operations until a valid abnormal reason information is received; The anomaly cause information is bound to the time information of the abnormal downtime event, the device identifier of the target device, and the task identifier of the current production task to generate an anomaly attribution record.

10. A production equipment efficiency data processing device, characterized in that, include: The signal acquisition module is used to acquire the status signals and action signals of the target device within the target statistical period, and to determine the running time information and downtime information of the target device based on the status signals, and to determine the physical action count information corresponding to the target device based on the action signals. The work report data acquisition module is used to acquire manually reported production data corresponding to the target statistical period; The theoretical output acquisition module is used to acquire the effective output coefficient corresponding to a single physical action of the target equipment under the current production task; The deviation calculation module is used to calculate the cross-modal verification deviation based on the physical action count information, the manually reported production data, and the effective output coefficient. The data quality control module is used to perform corresponding data quality control processing based on the determination result of the cross-modal verification deviation. The efficiency data generation module is used to generate equipment efficiency evaluation data based on the physical motion count information after the data quality control processing, the manually reported production data, and the effective output coefficient.

Images