Dynamic productivity calculation method and system for production equipment

Abstract

The invention discloses a dynamic productivity calculation method and system for production equipment, and belongs to the technical field of intelligent manufacturing and production operation optimization. The method comprises the following steps: acquiring operation time sequence data of target equipment in a historical statistical period; obtaining a service activity record which is associated with the equipment in the same period and is marked with a special service activity and a time range; cleaning the operation time sequence data based on the business activity record, and removing data in a time period corresponding to the unsteady-state production activity to obtain cleaned time sequence data representing the steady-state production capacity of the equipment; and based on the cleaned time sequence data, calculating to obtain a dynamic capacity reference value of the equipment. According to the method, the business system and the production data are associated, the abnormal time period in the historical data is automatically identified and eliminated, and the steady-state productivity reference of the equipment is continuously calculated and updated in combination with a sliding window mechanism, so that accurate and reliable productivity data input is provided for intelligent production scheduling, and the problems of data distortion and updating lag in the prior art are solved.

Description

Technical Field

[0001] This invention belongs to the field of intelligent manufacturing and production operation optimization technology, specifically relating to a method and system for calculating the dynamic capacity of production equipment. Background Technology

[0002] Against the backdrop of the accelerated integration of intelligent manufacturing and the Industrial Internet, production planning and scheduling optimization is undergoing a transformation from an extensive model relying on historical experience to a refined model driven by data and responding in real time. For discrete manufacturing industries such as cigarettes, the market demand for flexible production of small batches and multiple varieties is becoming increasingly urgent, which places higher demands on accurately assessing the dynamic capacity of production units. Currently, technological practices in this field mainly follow two paths, but both have significant limitations.

[0003] The most common capacity assessment method is static estimation based on fixed empirical parameters. The core of this method is to assign a fixed theoretical speed or rated capacity value to each piece of equipment, derived from human experience. When calculating production scheduling, the planned capacity is simply estimated by multiplying this fixed theoretical speed by the planned production time. The drawback of this method is that the fixed parameters it relies on often originate from theoretical values ​​on the equipment's nameplate or test data from the initial production phase. This fails to reflect the actual performance changes and wear and tear of the equipment during long-term operation and completely ignores dynamic disturbances such as frequent malfunctions, debugging, and line changes during production. The resulting capacity assessment is severely out of sync with the actual workshop capacity, leading to production plans that are often unrealistic, either causing equipment overload and inability to complete tasks or resulting in idle resources, significantly weakening the scientific rigor and reliability of the scheduling system.

[0004] To overcome the shortcomings of static estimation, some improvement schemes attempt to introduce historical data and adopt simple dynamic calculations based on historical averages. This method collects operating data of equipment over a period of time, calculates its average speed or output, and uses this as the basis for future capacity prediction. Although it is an improvement over purely empirical estimation, this method is essentially still a passive historical data backtracking. The fundamental problem is that the historical data used is unprocessed raw mixed data, which includes records of atypical and non-steady-state special production activities such as equipment overhauls, process experiments, and serious failure shutdowns. These outlier data points significantly distort historical averages, making the final calculation results unable to represent the true capacity of the equipment under normal and stable operating conditions. Therefore, this crude "dynamic" capacity data is highly volatile and lacks representativeness, and cannot provide reliable and stable input for intelligent scheduling systems that pursue accuracy and optimization.

[0005] However, whether relying on static experience or rough historical averages, the existing technological paradigm essentially follows the same logic: attempting to summarize the complex and ever-changing output capacity of equipment through a single, fixed value (whether a theoretical value or a historical mixed average). This paradigm exposes a structural contradiction when faced with the high precision requirements of flexible scheduling: to obtain a stable capacity benchmark, non-steady-state anomalies must be eliminated; however, to reflect the true dynamics of the equipment, it must be based on its actual operating data. Those skilled in the art are caught in a dilemma when faced with this contradiction: directly using all historical data containing anomalies results in distorted results; attempting to manually identify and eliminate anomalies faces the problems of exhaustive rules, large workload, and subjective arbitrariness, making automated, reusable, and accurate calculations impossible.

[0006] To address the above problems, this invention is proposed. Summary of the Invention

[0007] To address the shortcomings of existing technologies, the present invention aims to provide a dynamic capacity calculation method and system for production equipment, enabling continuous, automated, and accurate assessment of production equipment capacity. This method and system, through cross-system data association and intelligent cleaning based on business rules, achieves for the first time the ability to automatically generate a steady-state capacity benchmark for equipment from historical data containing known anomalies. Furthermore, by employing a sliding time window mechanism, it ensures that this benchmark value evolves synchronously with the current state of the equipment, ultimately providing a high-quality capacity data source for intelligent scheduling that is unavailable through traditional methods.

[0008] The technical solution adopted in this invention is:

[0009] The first aspect of this invention provides a method for calculating the dynamic capacity of production equipment, comprising the following steps:

[0010] S1. Obtain the runtime sequence data of the target device within the historical statistical period, wherein the runtime sequence data includes at least vehicle speed data recorded by timestamp;

[0011] S2. Obtain business activity records associated with the target device within the historical statistical period, wherein at least some business activities are identified as specific types of non-steady-state production activities and have corresponding activity time ranges;

[0012] S3. Based on the business activity records, clean the runtime sequence data: according to the activity time range corresponding to the business activities identified as non-steady-state production activities, remove the data segments whose timestamps fall within the range from the runtime sequence data, thereby obtaining cleaned time sequence data that characterizes the steady-state production capacity of the equipment.

[0013] S4. Based on the time-series data after cleaning, calculate the dynamic production capacity benchmark value of the target equipment.

[0014] In this invention, the specific type of non-steady-state production activity refers to planned activities that cause equipment to deviate from its normal, continuous, and stable production state, thereby preventing its output data from accurately reflecting the equipment's inherent production capacity. These activities are typically planned and recorded as independent work orders or events in an enterprise's production management system (such as MES). Typical characteristics include: during the activity, the equipment may be completely shut down, operate at low speed, frequently start and stop, or be in an unstable state with continuously adjusting process parameters. For example, such activities include, but are not limited to:

[0015] Equipment overhauls, planned maintenance, process debugging, new product trial production, equipment testing and verification, large-scale replacement or adjustment of molds and tooling, and shutdown training or drills are all activities that have in common. The common feature of these activities is that the corresponding production data (vehicle speed, output rate) differs significantly from the "real capacity" represented by the equipment operating under steady-state, full-load, and optimized parameters. Including these activities in capacity calculations will severely distort the results.

[0016] Preferably, step S4 includes: calculating the effective average speed of the target device within the historical statistical period based on the time-series data after cleaning;

[0017] The dynamic production capacity benchmark value is determined based on the effective average vehicle speed.

[0018] Preferably, the calculation of the effective average vehicle speed includes: converting the vehicle speed data in the cleaned time series data into instantaneous production capacity data according to a predetermined output conversion factor;

[0019] The instantaneous production capacity data is integrated over time to obtain the total effective output within the historical statistical period; the total effective production time corresponding to the cleaned time-series data is calculated.

[0020] The effective average speed is obtained by dividing the total effective output by the total effective production time.

[0021] Preferably, the dynamic capacity benchmark value (C) is based on the formula:

[0022]

[0023] The calculation shows that, The effective average vehicle speed is... This is a predetermined standard planned production time.

[0024] Preferably, the historical statistical period is a sliding time window of a preset time length tracing back from the current moment; the method is configured to be executed repeatedly according to the preset period to update the dynamic production capacity benchmark value.

[0025] A second aspect of the present invention provides a dynamic capacity calculation system for production equipment, comprising:

[0026] The data acquisition module is configured to acquire runtime sequence data of the target device within a historical statistical period, and to acquire business activity records associated with the target device within the historical statistical period, wherein at least some of the business activities are identified as specific types of non-steady-state production activities and have corresponding activity time ranges;

[0027] The data cleaning module, which is communicatively connected to the data acquisition module, is configured to clean the runtime sequence data based on the business activity records. According to the activity time range corresponding to the business activities identified as non-steady-state production activities, the module removes data segments whose timestamps fall within the runtime sequence data, thereby obtaining and outputting cleaned time-series data that characterizes the steady-state production capacity of the equipment.

[0028] The capacity calculation module is communicatively connected to the data cleaning module and is configured to calculate the dynamic capacity benchmark value of the target equipment based on the cleaned time-series data.

[0029] Preferably, the data cleaning module is further configured to: extract the activity time range corresponding to the non-steady-state production activities identified as specific types in the business activity records, and use the activity time range as a filtering condition to perform timestamp matching and removal on the runtime sequence data.

[0030] Preferably, the system is configured to: set the historical statistical period to a sliding time window of a preset time length tracing back from the current moment, and trigger the execution of the data acquisition module, data cleaning module and capacity calculation module according to the preset period to update the dynamic capacity benchmark value.

[0031] A third aspect of the present invention provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method described in the first aspect.

[0032] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in the first aspect.

[0033] The beneficial effects of this invention are:

[0034] 1. This invention creatively introduces and integrates business activity records from upper-level management systems (such as MES) with clear semantics (e.g., overhaul, trial production) and precise time ranges. For the first time, this invention achieves automated and rule-based cleaning of historical equipment runtime sequence data. This directly transforms the data foundation used for calculations from raw, mixed data containing abnormal interference in the background technology into clean data characterizing the steady-state production capacity of the equipment, solving the fundamental problem of distorted calculation results caused by abnormal data contamination.

[0035] 2. By introducing a sliding time window as the historical statistical period and configuring the method to repeat the process according to a preset cycle, the production capacity benchmark can be continuously updated to track the latest equipment status. This abandons the rigid model that relies on fixed theoretical values ​​and overcomes the lag of using outdated historical data. The output is a key performance indicator that evolves in sync with the current actual capacity of the equipment, meeting the stringent requirements of flexible scheduling for data timeliness.

[0036] 3. Existing technological improvements are limited to mathematical processing (such as averaging and filtering) within a single device performance data stream. This invention breaks free from this framework, innovatively linking the business event timeline of the MES system with the performance data timeline of the data acquisition system, utilizing the management semantics of the former to cleanse the physical signals of the latter. This cross-domain data fusion and collaboration method solves the problem of not being able to identify abnormal business contexts within a single system.

[0037] 4. The dynamic capacity benchmark calculated based on cleaned and purified data exhibits significantly reduced volatility and enhanced representativeness and reliability. This provides high-quality input for advanced planning and scheduling systems to make refined and flexible scheduling decisions, directly empowering enterprises to cope with the market challenges of small-batch, multi-variety production, reducing planning delays and resource misallocation, and improving overall operational efficiency and resource utilization.

[0038] 5. Compared with existing technologies that rely on simple dynamic calculations based on historical averages, this invention introduces business rules for intelligent cleaning, proactively identifying and eliminating known non-steady-state anomalies in historical data, thereby actively generating effective data that represents the equipment's inherent stable production capacity. This ensures that the final calculation result is no longer a simple arithmetic average based on mixed data, but rather a benchmark of the equipment's ability to continuously and stably perform under normal conditions, fundamentally overcoming the shortcomings of historical averaging methods where results are distorted by outliers and cannot represent true steady-state capabilities.

[0039] Instruction manual illustrations

[0040] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0041] Figure 1 This is a flowchart illustrating a method for calculating the dynamic production capacity of production equipment according to an embodiment of the present invention;

[0042] Figure 2 This is a structural block diagram of a dynamic capacity calculation system for production equipment provided in an embodiment of the present invention. Detailed Implementation

[0043] The present invention will be further described in detail below with reference to embodiments, but this is not intended to limit the invention. Any modifications or improvements made based on the teachings of the present invention fall within the protection scope of the present invention. Where specific techniques or conditions are not specified in the embodiments, they shall be performed in accordance with the techniques or conditions described in the literature in the art or in accordance with the product manual.

[0044] Example 1

[0045] This embodiment takes the cigarette rolling and packaging machine unit in the cigarette rolling and packaging workshop of a cigarette factory as the target equipment, and elaborates on the implementation process of the dynamic capacity calculation method of the present invention.

[0046] The execution logic of this embodiment and Figure 1 The process is exactly the same, and its purpose is to calculate the dynamic capacity baseline value of the equipment at the current moment. This value will serve as a key input parameter for the intelligent scheduling system to formulate the production plan for the coming week. Specifically:

[0047] S1. Obtain the runtime sequence data of the target device within the historical statistical period.

[0048] Target equipment: Wrapping machine, equipment number CBJ-2023-05.

[0049] Determination of the historical statistical period: In this embodiment, the sliding time window configured by the system has a length of 30 calendar days. Assuming that the current system is performing the calculation at time T_now (e.g., November 1, 2023, 00:05), the historical statistical period for this calculation is automatically determined to be [T_now - 30 days, T_now), that is, from October 2, 2023, 00:05 to November 1, 2023, 00:05.

[0050] Data Acquisition: The system automatically acquires all runtime sequence data of equipment CBJ-2023-05 within the aforementioned 30-day period through the application programming interface (API) or database of the workshop's Supervisory Control and Data Acquisition (SCADA) system. This data is stored in timestamp order, and each record contains at least:

[0051] timestamp: The time of the data point, accurate to the second (e.g., 2023-10-15 08:30:15).

[0052] speed: The equipment speed at this moment, measured in "packs per minute". This data directly reflects the instantaneous output rate of the equipment.

[0053] This step provides the raw data foundation for the entire method based on the actual operating state of the device.

[0054] S2: Obtain the business activity records associated with the target device within the historical statistical period.

[0055] Data source and acquisition: The system simultaneously accesses the factory's Manufacturing Execution System (MES) to query all work orders or event records that are closed or completed and bound to equipment CBJ-2023-05 within the same 30-day historical statistical period.

[0056] Record Content: The business activity records returned from MES are structured data. In this embodiment, assuming there are two related records within the cycle, their key fields are shown in the table below:

[0057]

[0058] Identification: In the system configuration of this invention, equipment overhaul and new product trial production have been predefined as specific types of non-steady-state production activities. Therefore, the system can automatically identify these two records as business activities requiring special attention. This step introduces external knowledge from the management system, with clear semantics and time context.

[0059] S3: Clean the runtime sequence data based on the business activity records.

[0060] Cleaning rule application: The system reads the two business activity records obtained in step S2. For each record identified as a non-steady-state production activity, the system extracts its activity time range (planned start time to planned end time).

[0061] For WO-20231001 (overhaul): the time range is from 08:00 on 2023-10-10 to 18:00 on 2023-10-12.

[0062] For WO-20231025 (trial production): the time range is from 09:00 on October 26, 2023 to 17:00 on October 26, 2023.

[0063] Data Removal: The system iterates through all runtime sequence data acquired in step S1. For each data point, it checks whether its timestamp falls within any of the aforementioned time ranges. Data points whose timestamps fall within these time periods are identified as abnormal data segments and removed from the dataset. For example, all data with vehicle speeds of 0 or extremely low during the major overhaul period from October 10th to 12th.

[0064] Output: After the removal process, the remaining data constitutes the cleaned time-series data. This data eliminates known non-steady-state production periods that would severely distort capacity calculations, thus providing a better representation of the equipment's production capacity under normal and stable operating conditions.

[0065] S4: Based on the time-series data after cleaning, calculate the dynamic production capacity benchmark value of the target equipment.

[0066] Calculate the effective average vehicle speed:

[0067] Data conversion: The factory's predefined output conversion factor is 1 standard box = 20 bags.

[0068] First, based on the coefficient "1 standard box = 20 bags", the vehicle speed unit is converted from bags / minute to standard boxes / minute, that is: vehicle speed_std (standard boxes / minute) = vehicle speed (bags / minute) × (1 / 20).

[0069] Then, the time unit is converted to the final required "standard containers / minute", that is: instantaneous capacity (standard containers / hour) = vehicle speed_std (standard containers / minute) × 60 (minutes / hour).

[0070] Combining the two steps above, we get the complete conversion formula: Instantaneous capacity (standard containers / hour) = Vehicle speed (packages / minute) × 3. The multiplier 3 comes from the product of the output and time conversion factors: (1 / 20) × 60 = 3.

[0071] Time integration to calculate effective total output: Integrate the converted instantaneous capacity data over time (in practice, the output of each time interval is accumulated). Assume that, according to calculations, within a 30-day historical statistical period (excluding overhaul and trial production periods), the effective total output of the equipment is P_total = 720,000 standard containers.

[0072] Calculate the total effective production time: directly calculate the total time covered by the cleaned time series data. Assume the total effective production time T_eff = 25,000 minutes (approximately 416.67 hours).

[0073] Calculate the effective average vehicle speed: according to the formula The effective average speed of the equipment was calculated as follows: Standard containers / 416.67 hours ≈ 1728 standard containers / hour. This value differs from a simple arithmetic average; it is an equivalent average speed calculated based on clean data, representing the equipment's steady-state output capacity.

[0074] Determine the dynamic capacity benchmark value:

[0075] The standard planned production time used in factory production scheduling The time for a standard production shift is 8 hours.

[0076] The dynamic capacity baseline value is calculated based on the formula C = V_ae × T_sp:

[0077] C = 1728 standard containers / hour × 8 hours = 13,824 standard containers.

[0078] Conclusion and Output: Based on the calculations performed using the method described in this embodiment, the system calculates the dynamic production capacity baseline value for equipment CBJ-2023-05 at the current time (November 1, 2023) to be 13,824 standard cases / shift. This value will be automatically output to the Intelligent Production Scheduling System (APS) via the system interface, serving as the basis for formulating the production capacity plan for each shift of this equipment for the coming week (November 1st to November 7th).

[0079] Implementation of the dynamic update mechanism:

[0080] The entire calculation process described above is encapsulated as an automated job, repeatedly executed at a preset cycle (e.g., 2 AM daily). Each execution automatically updates the historical statistical period (e.g., when calculating on November 2nd, the window updates to October 3rd to November 2nd). The system automatically retrieves data from the new window and re-executes S1 to S4, thereby generating an updated dynamic capacity baseline value. This ensures that the capacity baseline always evolves in sync with the latest operating status of the equipment, achieving true dynamic capacity calculation.

[0081] System Implementation:

[0082] The above method can be completely derived from Figure 2The system implementation is shown below. The data acquisition module is responsible for pulling data from the SCADA and MES systems; the data cleaning module executes cleaning logic based on business activity records; and the capacity calculation module completes the calculation of the effective average vehicle speed and dynamic capacity benchmark value. This system can be deployed on a server, and the software program implementing the method of this invention is stored on the server's hard drive.

[0083] Finally, it should be noted that the above preferred embodiments are only used to illustrate the technical solutions of this application and are not intended to limit it. Although this application has been described in detail through the above preferred embodiments, those skilled in the art should understand that various changes can be made to it in form and detail without departing from the scope defined by the claims of this application; the dimensions described in the drawings and embodiments are not related to the specific physical object and are not used to limit the protection scope of this application. The physical dimensions can be selected and changed according to actual needs.

Claims

1. A method for calculating the dynamic production capacity of production equipment, characterized in that, Includes the following steps: S1. Obtain the runtime sequence data of the target device within the historical statistical period, wherein the runtime sequence data includes at least vehicle speed data recorded by timestamp; S2. Obtain business activity records associated with the target device within the historical statistical period, wherein at least some business activities are identified as specific types of non-steady-state production activities and have corresponding activity time ranges; S3. Based on the business activity records, clean the runtime sequence data: according to the activity time range corresponding to the business activities identified as non-steady-state production activities, remove the data segments whose timestamps fall within the range from the runtime sequence data, thereby obtaining cleaned time sequence data that characterizes the steady-state production capacity of the equipment. S4. Based on the time-series data after cleaning, calculate the dynamic production capacity benchmark value of the target equipment.

2. The method according to claim 1, characterized in that, Step S4 includes: Based on the time-series data after cleaning, the effective average speed of the target device within the historical statistical period is calculated. The dynamic production capacity benchmark value is determined based on the effective average vehicle speed.

3. The method according to claim 2, characterized in that, The calculation of the effective average vehicle speed includes: converting the vehicle speed data in the cleaned time series data into instantaneous production capacity data according to a predetermined output conversion factor; The instantaneous production capacity data is integrated over time to obtain the total effective output within the historical statistical period; the total effective production time corresponding to the cleaned time-series data is calculated. The effective average speed is obtained by dividing the total effective output by the total effective production time.

4. The method according to claim 2 or 3, characterized in that, The dynamic capacity benchmark value (C) is based on the formula: The calculation shows that, The effective average vehicle speed is... This is a predetermined standard planned production time.

5. The method according to any one of claims 1 to 4, characterized in that, The historical statistical period is a sliding time window of a preset time length that traces backward from the current moment; the method is configured to be executed repeatedly according to the preset period to update the dynamic production capacity benchmark value.

6. A dynamic capacity calculation system for production equipment, characterized in that, include: The data acquisition module is configured to acquire runtime sequence data of the target device within a historical statistical period, and to acquire business activity records associated with the target device within the historical statistical period, wherein at least some of the business activities are identified as specific types of non-steady-state production activities and have corresponding activity time ranges; The data cleaning module, which is communicatively connected to the data acquisition module, is configured to clean the runtime sequence data based on the business activity records. According to the activity time range corresponding to the business activities identified as non-steady-state production activities, the module removes data segments whose timestamps fall within the runtime sequence data, thereby obtaining and outputting cleaned time-series data that characterizes the steady-state production capacity of the equipment. The capacity calculation module is communicatively connected to the data cleaning module and is configured to calculate the dynamic capacity benchmark value of the target equipment based on the cleaned time-series data.

7. The system according to claim 6, characterized in that, The data cleaning module is further configured to: extract the activity time range corresponding to the non-steady-state production activities identified as specific types in the business activity records, and use the activity time range as a filtering condition to perform timestamp matching and removal on the runtime sequence data.

8. The system according to claim 6, characterized in that, The system is configured to: set the historical statistical period to a sliding time window of a preset time length that traces backward from the current moment, and trigger the execution of the data acquisition module, data cleaning module and capacity calculation module according to the preset period to update the dynamic capacity benchmark value.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 5.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1 to 5.

Images