Intelligent management method and device for lean production of enterprises

Abstract

The invention provides an intelligent management method and device for enterprise lean production, relates to the field of enterprise lean production management, and solves the technical problem of low management efficiency of a production management mode in the prior art. The method comprises the following steps: collecting production total factor data through an intelligent gateway deployed on an edge layer; the production total factor data comprises equipment operation data, process parameter data and enterprise production data; transmitting the collected production total factor data to an industrial data lake of a data middle platform for storage, and calculating a core lean index based on the production total factor data; the core lean indexes comprise the equipment comprehensive efficiency, the production efficiency, the quality loss rate and the inventory turnover rate; generating an intelligent management decision based on the core lean index; the intelligent management decision comprises an equipment management decision, a production scheduling decision and a quality tracing decision. The method and device are used in the enterprise management process.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of enterprise lean production management, and in particular to an intelligent management method and device for enterprise lean production. BACKGROUND

[0002] Under the background of global economic integration, market competition is becoming increasingly fierce, and enterprises are facing unprecedented challenges. In order to stand out in the competition and achieve sustainable development, enterprises must continuously improve production efficiency, optimize cost structure, and ensure product quality. Digital and intelligent transformation has become the core way for enterprises to achieve these goals. However, many enterprises have encountered many significant obstacles in the process of promoting transformation.

[0003] In the current production management mode, management decisions are mostly dependent on manual experience and judgment, lacking systematic decision support based on data, resulting in low equipment utilization, production scheduling lag, and difficulty in tracing quality problems, which restricts the improvement of enterprise lean production level. SUMMARY

[0004] The present application provides an intelligent management method and device for enterprise lean production, which solves the technical problem of low management efficiency of the production management mode of the prior art.

[0005] To achieve the above-mentioned purpose, the present application adopts the following technical solutions: In a first aspect, an intelligent management method for enterprise lean production is provided, comprising: collecting production full-factor data through an intelligent gateway deployed in an edge layer; the production full-factor data includes equipment operation data, process parameter data, and enterprise production data; transmitting the collected production full-factor data to an industrial data lake of a data center for storage, and calculating core lean indicators based on the production full-factor data; the core lean indicators include equipment comprehensive efficiency, production efficiency, quality loss rate, and inventory turnover rate; generating intelligent management decisions based on the core lean indicators; the intelligent management decisions include equipment management decisions, production scheduling decisions, and quality traceability decisions.

[0006] Based on the above technical solutions, the present application can collect production full-factor data in multiple dimensions such as equipment operation data, process parameter data, and enterprise production data through an intelligent gateway deployed in an edge layer, then store the data through a data center and calculate core lean indicators such as equipment comprehensive efficiency, production efficiency, quality loss rate, and inventory turnover rate, and then generate intelligent management decisions based on the calculated core lean indicators, realizing the digitalization and intelligentization of production management, and solving the problem of low management efficiency caused by incomplete data collection, inaccurate indicator calculation, and decision-making relying on experience in the prior art.

[0007] In conjunction with the first aspect mentioned above, in one possible implementation, the method includes: inputting total production factor data into an analysis model to generate corresponding core lean indicators; the analysis model consists of an input layer, a recurrent neural network (RNN) layer, a long short-term memory (LSTM) layer, a fully connected layer, and an output layer; wherein, the input layer is used to receive total production factor data; the RNN layer is used to extract features from the time-series data in the total production factor data; the LSTM layer is used to perform weighted calculations based on the feature data from the previous time step and the feature data extracted by the RNN layer at the current time step to obtain the adjusted feature data at the current time step; the fully connected layer is used to map the feature data output by the LSTM layer to the dimension corresponding to the core lean indicators through a fully connected operation; the output layer is used to calculate and output the core lean indicators based on the mapped feature data output by the fully connected layer.

[0008] In conjunction with the first aspect mentioned above, in one possible implementation, the LSTM layer includes a forget gate, an input gate, an output gate, and a memory unit. The forget gate is used to calculate the forgetting weight based on the feature data output by the LSTM layer at the previous time step and the feature data input at the current time step. The forgetting weight is used to characterize the proportion of information retained from the long-term feature information stored in the memory unit at the previous time step. The input gate is used to calculate the input weight based on the feature data output by the LSTM layer at the previous time step and the input feature data at the current time step. The input weight is used to characterize the proportion of information added from the feature data input at the current time step to the long-term feature information stored in the memory unit. The memory unit is used to update the stored long-term feature information based on the forgetting weight and the input weight. The output gate is used to output the adjusted feature data at the current time step based on the long-term feature information stored in the memory unit.

[0009] In conjunction with the first aspect mentioned above, in one possible implementation, the method includes: constructing a decision model using a classification tree algorithm; wherein the decision model includes a root node, internal nodes, and leaf nodes; the root node, internal nodes, and leaf nodes are connected through decision branches; the root node is used to store total production factor data and core lean indicators; each internal node corresponds to one parameter in the core lean indicators and related production data; the internal nodes are used to divide decision branches according to the corresponding parameters; each leaf node corresponds to one decision in intelligent management decision-making; and the total production factor data and core lean indicators are input into the decision model to generate intelligent management decisions.

[0010] In conjunction with the first aspect mentioned above, in one possible implementation, equipment management decisions include decisions to generate equipment maintenance work orders and decisions to activate standby equipment; production scheduling decisions include decisions to prioritize urgent orders and decisions to adjust the production sequence; and quality traceability decisions include decisions to locate and alert on problems and decisions to implement corrective actions.

[0011] In conjunction with the first aspect mentioned above, in one possible implementation, the method further includes: performing data preprocessing operations on the full-factor data; the data preprocessing operations include data standardization, data smoothing, and data missing information imputation; wherein, the data standardization operation is used to adjust the data range of different dimensions to a preset range according to the standardization function; the data smoothing operation is used to smooth the data based on the moving average filtering algorithm; and the data missing information imputation operation is used to fit the data of the same dimension through a regression model.

[0012] In conjunction with the first aspect mentioned above, in one possible implementation, the normalized function is as follows: ; in, The data has been standardized. For the data before processing, The mean of the data. The standard deviation of the data; The moving average filtering algorithm is implemented using the following formula: ; in, For time points Data filtered by moving average at that time This represents the number of time points within the time window. For time points Data at that time, This indicates rounding down to the nearest integer.

[0013] In conjunction with the first aspect mentioned above, in one possible implementation, the method further includes: generating visualization charts based on core lean indicators; the visualization charts include: a bar chart for comparing the overall efficiency of different devices, a line chart for showing the trend of quality loss rate changes, and a dashboard for presenting real-time energy consumption data; displaying the visualization charts through a management dashboard; and adjusting the color of the color blocks in the visualization charts based on the values ​​of the core lean indicators.

[0014] Secondly, an intelligent management device for lean production in enterprises is provided, comprising: a communication unit and a processing unit; the communication unit is used to collect all production element data through an intelligent gateway deployed at the edge layer; the all production element data includes equipment operation data, process parameter data, and enterprise production data; the communication unit is used to transmit the collected all production element data to the industrial data lake of the data platform for storage; the processing unit is used to calculate core lean indicators based on the all production element data; the core lean indicators include overall equipment efficiency, production efficiency, quality loss rate, and inventory turnover rate; the processing unit is used to generate intelligent management decisions based on the core lean indicators; the intelligent management decisions include equipment management decisions, production scheduling decisions, and quality traceability decisions.

[0015] In conjunction with the second aspect above, in one possible implementation, the processing unit is used to: input production total factor data into the analysis model to generate corresponding core lean indicators; the analysis model consists of an input layer, a recurrent neural network (RNN) layer, a long short-term memory (LSTM) layer, a fully connected layer, and an output layer; wherein, the input layer is used to receive production total factor data; the RNN layer is used to extract features from the time-series data in the production total factor data; the LSTM layer is used to perform weighted calculations based on the feature data from the previous time step and the feature data extracted by the RNN layer at the current time step to obtain the adjusted feature data at the current time step; the fully connected layer is used to map the feature data output by the LSTM layer to the dimension corresponding to the core lean indicators through a fully connected operation; the output layer is used to calculate and output the core lean indicators based on the mapped feature data output by the fully connected layer.

[0016] In conjunction with the second aspect mentioned above, in one possible implementation, the LSTM layer includes a forget gate, an input gate, an output gate, and a memory unit. The forget gate is used to calculate the forgetting weight based on the feature data output by the LSTM layer at the previous time step and the feature data input at the current time step. The forgetting weight is used to characterize the proportion of information retained from the long-term feature information stored in the memory unit at the previous time step. The input gate is used to calculate the input weight based on the feature data output by the LSTM layer at the previous time step and the input feature data at the current time step. The input weight is used to characterize the proportion of information added from the feature data input at the current time step to the long-term feature information stored in the memory unit. The memory unit is used to update the stored long-term feature information based on the forgetting weight and the input weight. The output gate is used to output the adjusted feature data at the current time step based on the long-term feature information stored in the memory unit.

[0017] In conjunction with the second aspect above, in one possible implementation, the processing unit is used to: construct a decision model using a classification tree algorithm; wherein the decision model includes a root node, internal nodes, and leaf nodes; the root node, internal nodes, and leaf nodes are connected through decision branches; the root node is used to store total production factor data and core lean indicators; each internal node corresponds to one parameter in the core lean indicators and related production data; the internal nodes are used to divide decision branches according to the corresponding parameters; each leaf node corresponds to one decision in intelligent management decision-making; and the total production factor data and core lean indicators are input into the decision model to generate intelligent management decisions.

[0018] In conjunction with the second aspect above, in one possible implementation, equipment management decisions include decisions to generate equipment maintenance work orders and decisions to activate standby equipment; production scheduling decisions include decisions to prioritize urgent orders and decisions to adjust the production sequence; and quality traceability decisions include decisions to locate and alert on problems and decisions to implement corrective actions.

[0019] In conjunction with the second aspect above, in one possible implementation, the processing unit is used to: perform data preprocessing operations on all-factor data; the data preprocessing operations include data standardization, data smoothing, and data missing information imputation; wherein, the data standardization operation is used to adjust the data range of different dimensions to a preset range according to the standardization function; the data smoothing operation is used to smooth the data based on the moving average filtering algorithm; and the data missing information imputation operation is used to fit the data of the same dimension through a regression model.

[0020] In conjunction with the second aspect above, in one possible implementation, the normalized function is as follows: ; in, The data has been standardized. For the data before processing, The mean of the data. The standard deviation of the data; The moving average filtering algorithm is implemented using the following formula: ; in, For time points Data filtered by moving average at that time This represents the number of time points within the time window. For time points Data at that time, This indicates rounding down to the nearest integer.

[0021] In conjunction with the second aspect above, in one possible implementation, the processing unit is used to: generate visualization charts based on core lean indicators; the visualization charts include: a bar chart for comparing the overall efficiency of different devices, a line chart for showing the trend of quality loss rate changes, and a dashboard for presenting real-time energy consumption data; display the visualization charts through a management dashboard; the color of the color blocks in the visualization charts is adjusted based on the values ​​of the core lean indicators.

[0022] Thirdly, this application provides an intelligent management device for lean manufacturing in enterprises, comprising: a processor and a storage medium; the storage medium includes instructions, and the processor is used to execute the instructions to implement the method described in any of the above embodiments. This intelligent management device for lean manufacturing can be an electronic device or a chip within an electronic device.

[0023] Fourthly, this application provides a computer-readable storage medium storing instructions that, when executed on an intelligent management device for lean manufacturing, cause the intelligent management device for lean manufacturing to perform the methods described in any of the above embodiments.

[0024] Fifthly, this application provides a computer program product containing instructions that, when run on an intelligent management device for lean manufacturing, cause the intelligent management device for lean manufacturing to perform the method described in any of the above embodiments.

[0025] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description

[0026] Figure 1 A system architecture diagram of an intelligent management system for lean production in an enterprise, provided as an embodiment of this application; Figure 2 A flowchart illustrating an intelligent management method for lean production in enterprises, provided as an embodiment of this application; Figure 3 A flowchart illustrating another intelligent management method for lean production in enterprises, provided as an embodiment of this application; Figure 4 A flowchart illustrating another intelligent management method for lean production in enterprises, provided as an embodiment of this application; Figure 5 A flowchart illustrating another intelligent management method for lean production in enterprises, provided as an embodiment of this application; Figure 6 A flowchart illustrating another intelligent management method for lean production in enterprises, provided as an embodiment of this application; Figure 7 A schematic diagram of the structure of an intelligent management device for lean production in an enterprise, provided in an embodiment of this application; Figure 8 This is a schematic diagram of the hardware structure of an intelligent management device for lean production in enterprises, provided as an embodiment of this application. Detailed Implementation

[0027] In the description of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. The "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Furthermore, "at least one" means one or more, and "multiple" means two or more. The terms "first," "second," etc., do not limit the quantity or order of execution, and "first," "second," etc., do not necessarily imply differences.

[0028] It should be noted that, in this application, the terms "exemplary" or "for example" are used to indicate that something is being described as an example, illustration, or illustration. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0029] In today's context of global economic integration, market competition is increasingly fierce, and enterprises face unprecedented challenges. To stand out and achieve sustainable development, companies must continuously improve production efficiency, optimize cost structures, and ensure product quality. Digital and intelligent transformation has become a core pathway for enterprises to achieve these goals. However, many companies have encountered significant obstacles in the process of transformation.

[0030] First, some enterprises severely lack digital tools. In modern production management systems, various digital tools play a crucial role in the real-time collection, analysis, and processing of production data. However, in reality, many enterprises still rely on traditional manual operations and experience-based management models. For example, in recording equipment operation data, manual, timed recording not only consumes significant manpower but also makes it difficult to guarantee the accuracy and timeliness of the data. This prevents enterprise management from dynamically adjusting and optimizing production processes based on timely and accurate data, thus severely limiting the improvement of production efficiency and effective cost control.

[0031] Secondly, some enterprises, due to their weak technological foundation, struggle to meet the technical requirements of transformation. Their production equipment is mostly outdated, lacking the hardware foundation needed for intelligent upgrades, such as built-in intelligent sensors for real-time monitoring of equipment operation and corresponding communication modules for wireless data transmission. For example, some traditional manufacturing enterprises' processing equipment cannot be directly connected to industrial internet platforms because its interfaces are incompatible with modern industrial network protocols, making it difficult for these enterprises to build a complete production data chain and severely limiting the depth and breadth of their digital transformation.

[0032] Furthermore, enterprises face significant risks when undergoing transformation. Digital transformation involves multiple aspects, including purchasing advanced digital equipment, building complex industrial internet platforms, and providing comprehensive technical training for employees, all of which require substantial financial investment. Moreover, the transformation process is fraught with uncertainties, such as the compatibility of new technologies with existing business processes and the market's acceptance of the transformed products. If the transformation fails to meet expectations, enterprises will suffer significant economic losses. These risks cause enterprises to hesitate in their transformation decisions, delaying their adaptation to market changes and the enhancement of their competitiveness.

[0033] From a production process perspective, the existing problems of extensive operation are quite prominent. Equipment idling is widespread. Idle operation not only wastes a significant amount of energy, such as electricity and fuel, increasing production costs, but also accelerates mechanical wear and tear, shortens equipment lifespan, and consequently increases maintenance and replacement costs. For example, on some automated production lines, due to unreasonable production planning or poor equipment coordination, some equipment remains idle while waiting for raw materials or the next process step. Simultaneously, the problem of raw material inventory backlog is serious. Lacking accurate market demand forecasting models and efficient inventory management systems, companies often overstock raw materials, tying up large amounts of capital in inventory and impacting cash flow and operational efficiency.

[0034] In the field of quality control, most companies adopt a passive, post-production inspection model. Quality inspection is only conducted after product production is complete; once defective products are discovered, the initial investment in raw materials, manpower, and time is irrecoverable. Furthermore, due to the unstable fluctuations in process parameters—for example, in the food processing industry, minute changes in environmental parameters such as temperature and humidity, as well as process parameters such as speed and pressure on the production line—if not monitored and adjusted in a timely manner, can lead to product quality problems. However, existing quality control methods are insufficient to monitor these parameter changes in real time, making it impossible to take corrective measures promptly during production. This increases production costs and reduces the company's economic benefits and market reputation.

[0035] In terms of energy management, there is also a significant lag. High-energy-consuming equipment, such as large furnaces and refrigeration units, has not been effectively linked to production load. When production load is low, these high-energy-consuming equipment continues to operate at high power, resulting in substantial energy waste; while during peak production periods, insufficient energy supply may affect production efficiency and product quality. Existing technologies have failed to establish a complete "data-decision-execution" lean closed-loop system. Data from various production stages is isolated, failing to provide comprehensive, accurate, and forward-looking support for enterprise production decisions, quality control, and energy management. When formulating relevant strategies, enterprises lack scientific data support and rely more on experience-based judgment, making it difficult to truly achieve the goals of lean production.

[0036] In view of this, the embodiments of this application provide an intelligent management method for lean production in enterprises. It can collect multi-dimensional production data, such as equipment operation data, process parameter data, and enterprise production data, through intelligent gateways deployed at the edge layer. Then, it stores the data and calculates core lean indicators such as equipment comprehensive efficiency, production efficiency, quality loss rate, and inventory turnover rate through a data platform. Finally, it generates intelligent management decisions based on the calculated core lean indicators, realizing the digitalization and intelligentization of production management and solving the problems of low management efficiency caused by incomplete data collection, inaccurate indicator calculation, and reliance on experience in decision-making in the prior art.

[0037] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0038] Figure 1 This is an architecture diagram of an intelligent management system for lean production in enterprises, provided as an embodiment of this application. Figure 1 As shown, the intelligent management system for lean production in enterprises includes: an edge layer 101, a data platform 102, an application layer 103, and a terminal layer 104.

[0039] The edge layer 101 is used for data acquisition and transmission, and is equipped with a smart gateway supporting the OPC-UA / Modbus protocol. The smart gateway can connect to various production equipment, process systems, and energy monitoring devices to collect three types of data: equipment, process, and production data. This comprehensive production data lays a solid foundation for subsequent data processing and analysis.

[0040] Data Platform 102 is used for data processing, analysis, and decision support. Data Platform 102 includes an industrial data lake, a data governance module, and an algorithm engine.

[0041] The industrial data lake stores massive amounts of production data collected from edge layer 101. The data governance module cleans, integrates, and classifies this data, removing noise and duplicate information to ensure accuracy and consistency. For example, it standardizes similar data collected from different devices to ensure consistency in subsequent analysis. Based on the data governance, the algorithm engine uses various algorithm models to calculate core lean metrics.

[0042] The application layer 103 is used to transform the results analyzed by the data platform 102 into actual production management actions. The application layer 103 includes multiple modules such as equipment management, production scheduling, quality traceability, and supply chain collaboration.

[0043] In the equipment management module, when the data platform 102 reports abnormal situations such as low overall equipment effectiveness (OEE), the equipment management module can execute corresponding management operations according to intelligent management decisions.

[0044] For example, the equipment management module can activate a standby rolling mill, reducing the idle time of the main equipment and improving the overall operating efficiency of the equipment. For instance, in a copper processing company, when the main rolling mill malfunctions or becomes inefficient, the standby rolling mill can be put into use promptly, ensuring the continuity of production.

[0045] The production scheduling module is used to rationally allocate production tasks based on core lean metrics. For example, the module can prioritize urgent orders and reduce work-in-process inventory based on information such as order volume, equipment status, and inventory levels. For instance, when there is an urgent order demand, the module will adjust the production sequence to prioritize its production, while optimizing the connection between various production stages to avoid production delays and inventory backlogs.

[0046] The quality traceability module is used to pinpoint the stage and cause of product quality problems. For example, when a batch of products is found to have quality issues, the quality traceability module can quickly determine whether the problem is caused by abnormal melting temperature or improper raw material ratios, allowing for timely corrective measures to prevent similar problems from recurring.

[0047] The supply chain collaboration module is used to exchange information with suppliers and logistics partners to ensure the timeliness and stability of raw material supply, while optimizing logistics and reducing supply chain costs. For example, based on production plans and inventory levels, raw material purchase orders can be issued to suppliers in advance to ensure uninterrupted supply during production, while logistics and distribution can be rationally arranged to reduce transportation costs and time.

[0048] Terminal layer 104 provides a corresponding operating interface for enterprise managers. Terminal layer 104 includes a management dashboard and mobile terminals. The management dashboard can visually display various lean indicators, such as equipment OEE, quality loss rate, and inventory turnover rate. For example, the management dashboard can use charts and data displayed in different colors to quickly determine whether various indicators meet the standards. If abnormal indicators occur, they can be detected in a timely manner and corresponding measures can be taken. The mobile terminal is used to remotely acquire production data and management information, enabling remote monitoring and decision-making. For example, when enterprise leaders are traveling, they can use a mobile terminal application on their phones to view the real-time operating status of equipment in the enterprise's production site, order completion progress, and other information. When problems are discovered, instructions can be issued promptly for handling.

[0049] A mobile terminal can be a device with wireless or wired communication capabilities, deployed on land (indoors or outdoors, handheld or vehicle-mounted), on water (such as on ships), or in the air (e.g., on airplanes, balloons, and satellites). Also known as user equipment (UE), mobile station (MS), mobile terminal (MT), and terminal, it is a device that provides voice and / or data connectivity to users. Examples of mobile terminals include handheld devices with wireless connectivity and vehicle-mounted devices. Currently, mobile terminals can include: mobile phones, tablets, laptops, PDAs, mobile internet devices (MIDs), wearable devices (such as smartwatches, smart bracelets, pedometers, etc.), in-vehicle devices (such as cars, bicycles, electric vehicles, airplanes, ships, trains, high-speed trains, etc.), virtual reality (VR) devices, augmented reality (AR) devices, wireless terminal devices in industrial control, smart home devices (such as refrigerators, televisions, air conditioners, electricity meters, etc.), smart robots, workshop equipment, wireless terminal devices in self-driving, wireless terminal devices in remote medical surgery, wireless terminal devices in smart grids, wireless terminal devices in transportation safety, wireless terminal devices in smart cities, or wireless terminal devices in smart homes, and flying devices (such as smart robots, hot air balloons, drones, airplanes), etc. In one possible application scenario of this application, the terminal device is a terminal device that frequently operates on the ground, such as an in-vehicle device. In this application, for ease of description, the chip deployed in the above-mentioned device, such as a system-on-a-chip (SOC), a baseband chip, or other chips with communication functions, can also be referred to as a mobile terminal.

[0050] It should be noted that the various embodiments of this application can be referenced or learned from each other. For example, the same or similar steps, method embodiments, system embodiments and device embodiments can be referenced from each other without limitation.

[0051] Figure 2 A flowchart illustrating an intelligent management method for lean production in enterprises, provided as an embodiment of this application. Figure 2As shown, the method includes the following steps: Step 201: Collect all production element data through the smart gateway deployed at the edge layer.

[0052] The total production data includes equipment operation data, process parameter data, and enterprise production data.

[0053] For example, total production data refers to various key data involved in the production process of an enterprise. Equipment operation data includes equipment start-up and shutdown time, operating speed, load, etc.; process parameter data includes temperature, humidity, pressure, etc.; and enterprise production data includes order volume, inventory level, production progress, etc.

[0054] In some embodiments, the smart gateway supports the OPC-UA / Modbus protocol and can connect to different types of production equipment, process systems, and energy consumption monitoring devices, setting the sampling frequency according to the data type. For example, for rapidly changing equipment operating speeds, the sampling frequency can be set to 1 time / second; for relatively stable workshop temperature and humidity, the sampling frequency can be set to 1 time / minute.

[0055] It should be noted that the collected production data should undergo preliminary formatting to ensure data format uniformity and facilitate subsequent transmission and processing.

[0056] For example, in an automotive parts manufacturing company, the edge layer's smart gateway collects data on the stamping equipment's operating speed and mold temperature (equipment operation data), the welding workshop's humidity and welding current (process parameter data), as well as the day's order quantity and work-in-process inventory (company production data), providing basic data for subsequent analysis.

[0057] Step 202: Transmit the collected production total factor data to the industrial data lake of the data platform for storage, and calculate the core lean indicators based on the production total factor data.

[0058] The core lean metrics include Overall Equipment Effectiveness (OEE), Production Efficiency, Quality Loss Rate (QRS), and Inventory Turnover. An industrial data lake is a centralized data storage system for storing massive amounts of production data, which can be categorized and stored according to data source, time, and other dimensions. Core lean metrics are key indicators for evaluating a company's production and operational status: OEE reflects the effective operating capacity of equipment, Production Efficiency reflects output per unit time, Quality Loss Rate reflects losses caused by quality problems, and Inventory Turnover Rate reflects inventory turnover efficiency.

[0059] For example, each core lean indicator can be calculated using the following formulas: OEE = Effective Run Time / Planned Run Time; Inventory Turnover = Monthly Outbound Quantity / Average Inventory; Unit Energy Cost = Total Energy Cost / Total Output; Quality Loss Rate = Amount Lost on Defective Products / Total Output Value.

[0060] Step 203: Generate intelligent management decisions based on core lean indicators.

[0061] Intelligent management decisions include equipment management decisions, production scheduling decisions, and quality traceability decisions.

[0062] In some embodiments, equipment management decisions include generating equipment maintenance work orders and activating standby equipment; production scheduling decisions include prioritizing urgent orders and adjusting production sequence; and quality traceability decisions include problem location and alarm decisions and implementing corrective actions.

[0063] In one example, when the OEE value is below 80% or the operating data of key components shows an abnormal trend, such as the suction nozzle's suction count approaching its theoretical lifespan, this application can generate an equipment maintenance work order based on the decision to generate an equipment maintenance work order. This work order details possible causes of the failure, maintenance recommendations, and required spare parts information, and is then sent to the equipment maintenance personnel. Simultaneously, the standby equipment is activated according to the decision to activate standby equipment to ensure production continuity.

[0064] In another example, this application can make production scheduling decisions based on order volume, order priority, equipment status, and inventory. For urgent orders, production resources are prioritized and the production sequence is adjusted. For instance, when there are orders with short delivery times and high priority, production of some non-urgent orders is suspended, equipment resources are allocated to urgent orders, and subsequent production tasks are dynamically adjusted based on real-time production progress to ensure on-time order delivery.

[0065] In another example, this application can also monitor quality data and process parameters in real time during the production process. When the quality data exceeds the preset standard range, such as when a certain performance index of the product exceeds the tolerance range, a quality alarm is immediately issued. By tracing the production process data, the link where the quality problem occurred can be located, such as determining whether the quality problem is caused by abnormal welding temperature or raw material quality issues, so as to take corrective measures in a timely manner.

[0066] Based on the above technical solution, this application can collect multi-dimensional production data, including equipment operation data, process parameter data, and enterprise production data, through an intelligent gateway deployed at the edge layer. Then, the data is stored in a data platform and core lean indicators such as equipment comprehensive efficiency, production efficiency, quality loss rate, and inventory turnover rate are calculated. Finally, intelligent management decisions are generated based on the calculated core lean indicators, realizing the digitalization and intelligentization of production management and solving the problems of low management efficiency caused by incomplete data collection, inaccurate indicator calculation, and reliance on experience in decision-making in existing technologies.

[0067] Furthermore, this application can also perform in-depth data analysis based on artificial intelligence algorithms, thereby making the core lean metrics calculated more accurate.

[0068] As one possible embodiment of this application, combined with Figure 2 ,like Figure 3 As shown, the "calculation of core lean indicators based on total production factor data" in step 202 above can be achieved through the following steps.

[0069] Step 301: Input the total production factor data into the analysis model to generate the corresponding core lean indicators.

[0070] The analysis model consists of an input layer, a recurrent neural network (RNN) layer, a long short-term memory (LSTM) layer, a fully connected layer, and an output layer.

[0071] The input layer receives total production factor data. It can adapt to different types and formats of total production factor data, ensuring the data can be effectively processed by the model.

[0072] For example, production data can be input in a time-series format, with each time step including multiple features, such as the equipment's operating speed, temperature, pressure, and other parameters at a given moment, as well as corresponding process parameters and order-related information. The input layer can then pass this data to the next layer for processing.

[0073] RNN layers are used to extract features from time-series data in total production data.

[0074] For example, at each time step, the RNN layer receives the hidden state from the input layer and its own hidden state from the previous time step as input, and calculates the hidden state for the current time step using a specific weight matrix. The RNN layer can capture short-term dependencies in the data, such as the changing trends of device operating parameters over a short period.

[0075] The LSTM layer is used to perform weighted calculations based on the feature data from the previous time step and the feature data extracted by the RNN layer at the current time step, to obtain the adjusted feature data at the current time step.

[0076] The fully connected layer is used to map the feature data output by the LSTM layer to the dimension corresponding to the core lean metrics through fully connected operations.

[0077] For example, the fully connected layer performs a fully connected operation on the feature data output by the LSTM layer, mapping it to different dimensions to adapt to different task requirements. In this scenario, the fully connected layer will output feature representations used to compute various lean metrics.

[0078] The output layer is used to calculate and output the core lean metrics based on the mapped feature data output by the fully connected layer.

[0079] For example, the output layer calculates and outputs various lean metrics, such as OEE and productivity, based on the feature representations output by the fully connected layer. The calculation process of the output layer can be based on linear regression or other suitable algorithms to obtain numerical values ​​corresponding to the actual metrics.

[0080] Based on the above technical solution, this application can calculate core lean indicators through an analysis model consisting of an input layer, an RNN layer, an LSTM layer, a fully connected layer, and an output layer. It makes full use of the time series characteristics of production data, captures short-term and long-term dependencies, and further improves the accuracy of indicator calculation compared with traditional statistical methods, providing a more reliable basis for intelligent decision-making.

[0081] As one possible embodiment of this application, the LSTM layer includes a forget gate, an input gate, an output gate, and a memory unit.

[0082] The forget gate is used to calculate the forget weight based on the feature data output by the LSTM layer at the previous time step and the feature data input at the current time step. The forget weight is used to characterize the proportion of information retained from the long-term feature information stored in the memory unit at the previous time step.

[0083] For example, the forgetting weight ranges from 0 to 1, where 0 indicates complete forgetting of information from the previous moment and 1 indicates complete retention. For instance, when processing data related to equipment aging, the forgetting gate assigns a forgetting weight of 0.3 (retaining 30%) to equipment operating features from 3 months ago and a forgetting weight of 0.8 (retaining 80%) to features from 1 week ago.

[0084] The input gate is used to calculate the input weights based on the feature data output by the LSTM layer at the previous time step and the input feature data at the current time step. The input weights represent the proportion of long-term feature information added to the memory unit from the feature data input at the current time step.

[0085] For example, the input weights also take values ​​of 0-1. For instance, when a sudden device overload is detected, the input gate assigns an input weight of 0.9 to the current overload feature (90% of long-term information is added) to ensure that the model pays attention to the sudden situation.

[0086] The memory unit is used to update the stored long-term feature information based on the forgetting weight and the input weight.

[0087] For example, the memory unit merges the 30% of long-term information retained from the previous moment with the 90% of overload features of the current input to update the new long-term feature information, reflecting the latest changes in the device state.

[0088] The output gate is used to output the adjusted feature data of the current moment based on the long-term feature information stored in the memory unit.

[0089] For example, the output gate extracts features related to device load from the updated long-term feature information and uses them as the output features at the current moment for subsequent fully connected layer processing.

[0090] By using the gating mechanism of the LSTM layer, key long-term feature information can be effectively filtered and retained, solving the gradient vanishing problem when traditional RNNs process long sequence data, making the calculation of core lean indicators more accurate, especially suitable for enterprises with long production cycles and complex data relationships.

[0091] Furthermore, this application can also implement intelligent decision-making strategies based on classification algorithms to make decision-making more efficient.

[0092] As one possible embodiment of this application, combined with Figure 2 ,like Figure 4 As shown, step 203 above can be achieved through the following steps.

[0093] Step 401: Construct a decision model using a classification tree algorithm.

[0094] The decision model includes a root node, internal nodes, and leaf nodes, which are connected by decision branches.

[0095] The root node stores all production data and core lean metrics, such as OEE (Outcome Effectiveness), wear and tear of key components, order volume, order priority, and inventory levels. The root node is the starting point for decision-making; subsequent decision branches are based on the data in the root node.

[0096] Each internal node corresponds to a parameter in the core lean metrics and related production data. Internal nodes are used to divide decision branches based on the corresponding parameter. For example, using the equipment OEE value as a dividing attribute, when the OEE value is higher or lower than a certain threshold, the data will be divided into different branches. By continuously dividing the data, internal nodes gradually narrow down the decision-making scope, making the decisions more precise.

[0097] For example, an internal node with the parameter "OEE < 70%" will direct the branch with OEE = 65% (< 70%) in the root node data to the next level internal node; another internal node with the parameter "Order priority is high" will direct the branch with order priority = high to a specific leaf node.

[0098] Each leaf node corresponds to a decision in the intelligent management decision-making process. Each leaf node represents a specific decision action, such as generating an equipment maintenance work order, adjusting the production plan, or activating standby equipment.

[0099] Step 402: Input the total production data and core lean indicators into the decision-making model to generate intelligent management decisions.

[0100] For example, in equipment management decisions, if the leaf node corresponds to "generate equipment maintenance work order", the decision condition can be "OEE < 85% and critical component wear > 80% of design life"; in production scheduling decisions, the condition for "prioritizing urgent orders" can be "order priority = high and delivery time ≤ 3 days and inventory is sufficient"; in quality traceability decisions, the condition for "problem location alarm" can be "parts size exceeds tolerance ±5%".

[0101] Based on the above technical solution, this application can construct a decision-making model through a classification tree algorithm, transforming complex management decisions into structured tree logic, making the decision-making process transparent, interpretable, and fast-responding, which is more efficient than manual decision-making and ensures that problems in the production process are dealt with in a timely manner.

[0102] In addition, this application can also perform data preprocessing on the collected production all-factor data to improve data quality.

[0103] As one possible embodiment of this application, combined with Figure 2 ,like Figure 5 As shown, the method may also include the following steps.

[0104] Step 501: Perform data preprocessing on all element data.

[0105] Data preprocessing operations include data standardization, data smoothing, and data missing information imputation.

[0106] Data standardization is used to adjust the range of data from different dimensions to a preset range according to a standardization function. For different types of collected data, due to differences in their numerical ranges and units, data standardization is performed first to ensure comparability. For example, equipment operating speed data may be in meters per second, while equipment temperature data is in degrees Celsius, resulting in significant differences in their numerical ranges. After standardization, the mean and standard deviation of all feature data are unified, thus eliminating the influence of different units and facilitating subsequent analysis.

[0107] In some embodiments, the normalization function is the following formula: ; in, The data has been standardized. For the data before processing, The mean of the data. This represents the standard deviation of the data.

[0108] The data smoothing process utilizes a moving average filtering algorithm to smooth the data. Since the collected data may contain noise interference, affecting the accuracy of data analysis, this application employs a moving average filtering method for data smoothing to remove noise. This effectively reduces data fluctuations, making the data smoother and highlighting its true trend.

[0109] In some embodiments, the moving average filtering algorithm is implemented using the following formula: ; in, For time points Data filtered by moving average at that time This represents the number of time points within the time window. For time points Data at that time, This indicates rounding down to the nearest integer.

[0110] Data imputation is used to fit data of the same dimension to a regression model. During data collection, some data may be missing. This application proposes a method based on regression model prediction to handle missing values.

[0111] Taking equipment operating parameter data as an example, if the equipment's operating speed data at a certain moment is missing, a regression model can be built using other relevant parameters of the equipment (such as motor current, load, etc.) and historical operating data to predict the missing speed value. In this way, missing data is filled in, ensuring data integrity and providing more comprehensive data support for subsequent data analysis.

[0112] After the above data preprocessing steps, the data is then transmitted to the data platform for subsequent storage, cleaning, integration, and analysis. This significantly improves the accuracy and reliability of data analysis, thereby enhancing the quality and effectiveness of lean production management in enterprises.

[0113] In addition, this application can also visualize the indicators through the management dashboard, making it easier for managers to intuitively view the company's operating status.

[0114] As one possible embodiment of this application, combined with Figure 2 ,like Figure 6 As shown, the method may also include the following steps.

[0115] Step 601: Generate visual charts based on core lean metrics.

[0116] The visualization charts include: bar charts to compare the overall efficiency of different devices, line charts to show the trend of quality loss rate changes, and dashboards to present real-time energy consumption data.

[0117] For example, bar charts can be used to compare the OEE of different devices, line charts can show the trend of quality loss rate, and dashboards can display energy consumption indicators. For instance, a company's bar chart shows the OEE of three lathes (80%, 85%, and 75%), the line chart shows that the quality loss rate has decreased from 3% to 1.5% in the past month, and the dashboard shows that the current energy consumption is 120 kW / h.

[0118] Step 602: Display visual charts through the management cockpit.

[0119] The colors of the blocks in the visualization charts are adjusted based on the values ​​of the core lean metrics.

[0120] In some embodiments, different indicators are color-coded according to set thresholds. For example, a red indicator is displayed when the device's OEE is below 80%, alerting management personnel. Management personnel can filter and drill down the displayed data via touchscreen or control console to gain a deeper understanding of the detailed information behind each indicator. Mobile terminals can synchronize data with the management dashboard in real time, allowing management personnel to remotely view the visualizations displayed in the management dashboard, such as the operating status of each device on the production line, real-time production progress, and quality alarm information. They can also perform approval and instruction operations on the mobile terminal, such as approving equipment maintenance requests and adjusting production plans, enabling remote monitoring and decision-making.

[0121] The foregoing mainly describes the solutions of the embodiments of this application from the perspective of device implementation. It is understood that each device, such as an intelligent management device for lean production in enterprises, includes at least one of the hardware structures and software modules corresponding to the execution of each function in order to achieve the above-mentioned functions. Those skilled in the art should readily recognize that, in conjunction with the units and algorithm steps of the various examples described in the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in a hardware or computer software-driven hardware manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0122] This application embodiment can divide the intelligent management device for lean production in enterprises into functional units according to the above method example. For example, each function can be divided into a separate functional unit, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or as a software functional unit. It should be noted that the unit division in this application embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0123] When using integrated units, Figure 7 A possible structural schematic diagram of the intelligent management device for lean production in enterprises (referred to as intelligent management device 70 for lean production in enterprises) involved in the above embodiments is shown. The intelligent management device 70 for lean production in enterprises includes a processing unit 701 and a communication unit 702, and may also include a storage unit 703. Figure 7 The structural diagram shown can be used to illustrate the structure of the intelligent management device for lean production in enterprises involved in the above embodiments.

[0124] when Figure 7 The structural diagram shown is used to illustrate the structure of the intelligent management device for lean production in the above embodiments. The processing unit 701 is used to control and manage the operation of the intelligent management device for lean production, the communication unit 702 is used for the intelligent management device for lean production to communicate with other devices, and the storage unit 703 is used to store the program code and data of the intelligent management device for lean production.

[0125] For example, the communication unit 702 is used to collect all production element data through a smart gateway deployed at the edge layer; the all production element data includes equipment operation data, process parameter data, and enterprise production data.

[0126] The communication unit 702 is used to transmit the collected production data to the industrial data lake of the data platform for storage.

[0127] The processing unit 701 is used to calculate core lean indicators based on the total production data; the core lean indicators include overall equipment efficiency, production efficiency, quality loss rate and inventory turnover rate.

[0128] The processing unit 701 is used to generate intelligent management decisions based on core lean indicators; the intelligent management decisions include equipment management decisions, production scheduling decisions, and quality traceability decisions.

[0129] In one possible implementation, the processing unit 701 is used to: input production total factor data into the analysis model to generate corresponding core lean indicators; the analysis model consists of an input layer, a recurrent neural network (RNN) layer, a long short-term memory (LSTM) layer, a fully connected layer, and an output layer; wherein, the input layer is used to receive production total factor data; the RNN layer is used to extract features from the time-series data in the production total factor data; the LSTM layer is used to perform weighted calculations based on the feature data from the previous time step and the feature data extracted by the RNN layer at the current time step to obtain the adjusted feature data at the current time step; the fully connected layer is used to map the feature data output by the LSTM layer to the dimension corresponding to the core lean indicators through a fully connected operation; the output layer is used to calculate and output the core lean indicators based on the mapped feature data output by the fully connected layer.

[0130] In one possible implementation, the LSTM layer includes a forget gate, an input gate, an output gate, and a memory unit. The forget gate is used to calculate the forgetting weight based on the feature data output by the LSTM layer at the previous time step and the feature data input at the current time step. The forgetting weight is used to characterize the proportion of information retained from the long-term feature information stored in the memory unit at the previous time step. The input gate is used to calculate the input weight based on the feature data output by the LSTM layer at the previous time step and the input feature data at the current time step. The input weight is used to characterize the proportion of information added from the feature data input at the current time step to the long-term feature information stored in the memory unit. The memory unit is used to update the stored long-term feature information based on the forgetting weight and the input weight. The output gate is used to output the adjusted feature data at the current time step based on the long-term feature information stored in the memory unit.

[0131] In one possible implementation, the processing unit 701 is used to: construct a decision model using a classification tree algorithm; wherein the decision model includes a root node, internal nodes, and leaf nodes; the root node, internal nodes, and leaf nodes are connected by decision branches; the root node is used to store total production factor data and core lean indicators; each internal node corresponds to one parameter in the core lean indicators and related production data; the internal nodes are used to divide decision branches according to the corresponding parameters; each leaf node corresponds to one decision in intelligent management decision-making; and the total production factor data and core lean indicators are input into the decision model to generate intelligent management decisions.

[0132] In one possible implementation, equipment management decisions include generating equipment maintenance work orders and activating standby equipment; production scheduling decisions include prioritizing urgent orders and adjusting production sequences; and quality traceability decisions include locating and alerting problems and implementing corrective actions.

[0133] In one possible implementation, the processing unit 701 is used to: perform data preprocessing operations on all-factor data; the data preprocessing operations include data standardization, data smoothing, and data missing information imputation; wherein, the data standardization operation is used to adjust the data range of different dimensions to a preset range according to the standardization function; the data smoothing operation is used to smooth the data based on the moving average filtering algorithm; and the data missing information imputation operation is used to fit the data of the same dimension through a regression model.

[0134] In one possible implementation, the normalization function is given by the following formula: ; in, The data has been standardized. For the data before processing, The mean of the data. The standard deviation of the data; The moving average filtering algorithm is implemented using the following formula: ; in, For time points Data filtered by moving average at that time This represents the number of time points within the time window. For time points Data at that time, This indicates rounding down to the nearest integer.

[0135] In one possible implementation, the processing unit 701 is used to: generate visualization charts based on core lean indicators; the visualization charts include: a bar chart to show the overall efficiency comparison of different devices, a line chart to show the trend of quality loss rate changes, and a dashboard to present real-time energy consumption data; display the visualization charts through a management dashboard; the color of the color blocks in the visualization charts is adjusted based on the values ​​of the core lean indicators.

[0136] The processing unit 701 can be a processor or a controller, and the communication unit 702 can be a communication interface, transceiver, transceiver circuit, transceiver device, etc. The term "communication interface" is a general term and may include one or more interfaces. The storage unit 703 can be a memory. When the intelligent management device 70 used for lean production in enterprises is a chip, the processing unit 701 can be a processor or a controller, and the communication unit 702 can be an input interface and / or an output interface, pins, or circuits, etc. The storage unit 703 can be a storage unit within the chip (e.g., a register, cache, etc.) or a storage unit located outside the chip (e.g., read-only memory (ROM), random access memory (RAM, etc.)).

[0137] The communication unit can also be called a transceiver unit. The antenna and control circuit with transceiver functions in the intelligent management device 70 for lean production can be considered as the communication unit 702 of the intelligent management device 70 for lean production, and the processor with processing functions can be considered as the processing unit 701 of the intelligent management device 70 for lean production. Optionally, the device in the communication unit 702 that implements the receiving function can be considered as the communication unit, which is used to execute the receiving steps in the embodiments of this application. The communication unit can be a receiver, a receiver circuit, etc. The device in the communication unit 702 that implements the transmitting function can be considered as the transmitting unit, which is used to execute the transmitting steps in the embodiments of this application. The transmitting unit can be a transmitter, a transmitter, a transmitting circuit, etc.

[0138] Figure 7 If the integrated units in the process are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, in essence, or the parts that contribute to the prior art, or all or part of the technical solutions, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. Storage media for storing computer software products include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory, random access memory, magnetic disks, or optical disks.

[0139] Figure 7 The units in the process can also be called modules; for example, a processing unit can be called a processing module.

[0140] This application also provides a hardware structure diagram of an intelligent management device for lean production in enterprises (denoted as intelligent management device 80 for lean production in enterprises), see [link to diagram]. Figure 8 The intelligent management device 80 for lean manufacturing includes a processor 801, and optionally, a memory 802 connected to the processor 801.

[0141] In the first possible implementation, see Figure 8 The intelligent management device 80 for lean manufacturing in enterprises also includes a transceiver 803. The processor 801, memory 802, and transceiver 803 are connected via a bus. The transceiver 803 is used to communicate with other devices or communication networks. Optionally, the transceiver 803 may include a transmitter and a receiver. The device in the transceiver 803 that implements the receiving function can be considered as a receiver, which is used to perform the receiving steps in the embodiments of this application. The device in the transceiver 803 that implements the transmitting function can be considered as a transmitter, which is used to perform the transmitting steps in the embodiments of this application.

[0142] Based on the first possible implementation method Figure 8 The structural diagram shown can be used to illustrate the structure of the intelligent management device for lean production in enterprises involved in the above embodiments.

[0143] in, Figure 8 This can also be illustrated by a system chip in an intelligent management device used for lean manufacturing in enterprises. In this case, the actions performed by the aforementioned intelligent management device for lean manufacturing can be implemented by this system chip. The specific actions performed can be found above and will not be repeated here.

[0144] In implementation, each step of the method provided in this embodiment can be completed by integrated logic circuits in the processor or by instructions in software form. The steps of the method disclosed in the embodiments of this application can be directly manifested as being executed by a hardware processor, or being executed by a combination of hardware and software modules in the processor.

[0145] The processor in this application may include, but is not limited to, at least one of the following: a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a microcontroller unit (MCU), or an artificial intelligence processor, etc., which are various computing devices that run software. Each computing device may include one or more cores for executing software instructions to perform calculations or processing. The processor may be a separate semiconductor chip or integrated with other circuits into a single semiconductor chip. For example, it may be integrated with other circuits (such as encoding / decoding circuits, hardware acceleration circuits, or various bus and interface circuits) to form a SoC (System-on-a-Chip), or it may be integrated as a built-in processor within an ASIC. The ASIC with the integrated processor may be packaged separately or together with other circuits. In addition to the cores for executing software instructions to perform calculations or processing, the processor may further include necessary hardware accelerators, such as field-programmable gate arrays (FPGAs), PLDs (programmable logic devices), or logic circuits that implement dedicated logic operations.

[0146] The memory in the embodiments of this application may include at least one of the following types: read-only memory (ROM) or other types of static storage devices capable of storing static information and instructions; random access memory (RAM) or other types of dynamic storage devices capable of storing information and instructions; or electrically erasable programmable-only memory (EEPROM). In some scenarios, the memory may also be a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media, or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures that can be accessed by a computer, but is not limited thereto.

[0147] This application also provides a computer-readable storage medium including instructions that, when run on a computer, cause the computer to perform any of the methods described above.

[0148] This application also provides a computer program product containing instructions that, when run on a computer, cause the computer to perform any of the methods described above.

[0149] This application also provides a chip including a processor and an interface circuit. The interface circuit is coupled to the processor. The processor is used to run computer programs or instructions to implement the above-described method. The interface circuit is used to communicate with other modules outside the chip.

[0150] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software programs, implementation can be, in whole or in part, in the form of a computer program product. This computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device containing one or more servers, data centers, etc., that can be integrated with the medium. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state disks (SSDs)).

[0151] Although this application has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings, disclosure, and appended claims, will understand and implement other variations of the disclosed embodiments in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple instances. A single processor or other unit can implement several functions listed in the claims. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce good results.

[0152] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of this application as defined by the appended claims, and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from the spirit and scope of this application. Thus, if such modifications and modifications of this application fall within the scope of the claims of this application and their equivalents, this application is also intended to include such modifications and modifications.

Claims

1. A method for intelligent management of lean production in an enterprise, characterized in that, The application relates to a production full-factor data collection method and system. An intelligent gateway deployed through an edge layer collects production full-factor data; The production full-factor data comprises equipment operation data, process parameter data and enterprise production data; The collected production full-factor data is transmitted to an industrial data lake of a data center for storage, and core lean indexes are calculated based on the production full-factor data; the core lean indexes comprise equipment comprehensive efficiency, production efficiency, quality loss rate and inventory turnover rate; Intelligent management decisions are generated based on the core lean indexes; the intelligent management decisions comprise equipment management decisions, production scheduling decisions and quality traceability decisions.

2. The method of claim 1, wherein, The core lean indexes are calculated based on the production full-factor data, comprising: The production full-factor data is input into an analysis model to generate corresponding core lean indexes; the analysis model comprises an input layer, an RNN (recurrent neural network) layer, an LSTM (long short-term memory) layer, a full connection layer and an output layer; The input layer is used for receiving the production full-factor data; the RNN layer is used for feature extraction of data with time sequence characteristics in the production full-factor data; the LSTM layer is used for weighted calculation based on feature data of a previous moment and feature data extracted by the RNN layer at a current moment to obtain adjusted feature data of the current moment; the full connection layer is used for mapping the feature data output by the LSTM layer to a dimension corresponding to the core lean indexes through full connection operation; and the output layer is used for calculating and outputting the core lean indexes according to the mapped feature data output by the full connection layer.

3. The method of claim 2, wherein, The LSTM layer comprises a forgetting gate, an input gate, an output gate and a memory unit; The forgetting gate is used for calculating a forgetting weight according to feature data output by the LSTM layer at a previous moment and feature data input at a current moment; the forgetting weight is used for representing a proportion of information retained from long-term feature information stored in the memory unit at the previous moment; The input gate is used for calculating an input weight according to feature data output by the LSTM layer at a previous moment and input feature data at a current moment; the input weight is used for representing a proportion of information of the input feature data at the current moment added to the long-term feature information stored in the memory unit; The memory unit is used for updating the stored long-term feature information based on the forgetting weight and the input weight; The output gate is used for outputting adjusted feature data of the current moment according to the long-term feature information stored in the memory unit.

4. The method of claim 1, wherein, The intelligent management decisions are generated based on the core lean indexes, comprising: A decision model is constructed by using a classification tree algorithm; the decision model comprises a root node, internal nodes and leaf nodes; the root node, the internal nodes and the leaf nodes are connected through decision branches; the root node is used for storing the production full-factor data and the core lean indexes; each internal node corresponds to a parameter in the core lean indexes and related production data; the internal node is used for dividing decision branches according to the corresponding parameter; and each leaf node corresponds to a decision in the intelligent management decisions. The production total factor data and the core lean indicators are input into the decision model to generate intelligent management decisions.

5. The method of claim 1, wherein, The device management decisions include device maintenance work order decisions and backup device activation decisions; the production scheduling decisions include emergency order processing priority decisions and production sequence adjustment decisions; and the quality traceability decisions include problem positioning alarm decisions and corrective action execution decisions.

6. The method of claim 1, wherein, Before the collected production total factor data is transmitted to the industrial data lake of the data center for storage and the core lean indicators are calculated based on the production total factor data, the method further includes: Data preprocessing operations are performed on the total factor data; the data preprocessing operations include data standardization operations, data smoothing processing operations, and data missing filling operations; the data standardization operations are used to adjust the data ranges of different dimensions to a preset range according to a standardization function; the data smoothing processing operations are based on a sliding average filtering algorithm to perform smoothing processing on the data; and the data missing filling operations are used to fit the data of the same dimension through a regression model.

7. The method of claim 6, wherein, The standardization function is as follows: ； wherein, is the data after standardization, is the data before processing, is the data mean, is the data standard deviation; The sliding average filtering algorithm is implemented by the following formula: ； wherein is the time point the data filtered with an exponential moving average, is the number of time points within the time window, is the time point the data, denotes the floor function.

8. The method of claim 1, wherein, The method further includes: Visual charts are generated based on the core lean indicators; the visual charts include a column chart for showing the comparison of the equipment comprehensive efficiency of different devices, a line chart for showing the change trend of the quality loss rate, and a dashboard for presenting real-time energy consumption data; The visual charts are displayed through a management cockpit; the color of the color blocks in the visual charts is adjusted based on the numerical value of the core lean indicators.

9. An intelligent management device for lean production of enterprises, characterized in that, The device includes a communication unit and a processing unit; The communication unit is configured to collect production total factor data through an intelligent gateway deployed in an edge layer; the production total factor data includes device operation data, process parameter data, and enterprise production data; The communication unit is configured to transmit the collected production total factor data to an industrial data lake of a data center for storage; the processing unit is configured to calculate core lean indicators based on the production total factor data; the core lean indicators include equipment comprehensive efficiency, production efficiency, quality loss rate, and inventory turnover rate; The processing unit is configured to generate intelligent management decisions based on the core lean indicators; the intelligent management decisions include device management decisions, production scheduling decisions, and quality traceability decisions.

10. An intelligent management device for lean production of enterprises, characterized in that, The device includes a processor and a communication interface; the communication interface and the processor are coupled, and the processor is configured to run computer programs or instructions to implement the intelligent management method for enterprise lean production as claimed in any one of claims 1-8. The device includes a processor and a communication interface; the communication interface and the processor are coupled, and the processor is configured to run computer programs or instructions to implement the intelligent management method for enterprise lean production as claimed in any one of claims 1-8.

Images