A method and system for automatically generating production schedules based on injection molding industry production

Abstract

The present invention relates to the technical field of production management in the injection molding industry, and discloses a method and system for automatically generating a production scheduling plan based on the production of the injection molding industry. The method includes: acquiring production orders, equipment status, and material inventory data, generating original production parameter data, and deploying a distributed production scheduling network; collecting and analyzing multi-level equipment load data and equipment status data, and generating corresponding influencing factors; collecting order priority data, and dividing production task areas; optimizing and calculating the production scheduling plan based on the task area data, integrating multi-factor data to generate comprehensive production scheduling data, and then generating and outputting production scheduling instructions. The system includes production scheduling network deployment, equipment load analysis, equipment status analysis, production task area analysis, and production scheduling instruction generation modules. The present invention realizes intelligent and precise injection molding production scheduling through multi-data integration and scientific modeling, improves equipment management efficiency, order delivery capabilities, and overall production efficiency, and is suitable for automated production scheduling management in the injection molding industry.

Description

Technical Field

[0001] The present invention relates to the technical field of production management in the injection molding industry, and in particular to a method and system for automatically generating a production schedule based on production in the injection molding industry. Background Art

[0002] From a data processing perspective, manual scheduling makes it difficult to fully and real-timely acquire and process multi-dimensional data from the production process. Injection molding production involves a large amount of production order data, equipment status data, material inventory data, etc. Manual operations are not only inefficient but also prone to data omissions or misjudgments. For example, key priority data such as the urgency, complexity, and delivery period of production orders cannot be promptly and accurately incorporated into scheduling, resulting in an increased risk of delayed order delivery. At the same time, equipment status data such as operating parameters and fault warnings cannot be monitored and analyzed in real time, making it difficult to detect and adjust equipment load imbalances in advance, which can easily cause excessive wear or idleness of equipment, affecting production continuity and equipment life.

[0003] In terms of production scheduling strategy formulation, traditional methods lack scientific models and algorithmic support. Manual scheduling is often based on simple rules or historical experience, and it is impossible to build a complex scheduling network model, making it difficult to achieve global optimization of the production process. For example, in the division of production areas and the deployment of scheduling nodes, manual operations are difficult to make reasonable plans based on production structure data, resulting in unreasonable distribution of production tasks, low coordination efficiency between production areas, and increased material transportation costs and time costs. In addition, the analysis of equipment loads only stays on the surface, and it is impossible to conduct multi-level load data collection and in-depth analysis. It is difficult to accurately calculate the load gradient index, load density difference data, and load fluctuation trajectory data, making it impossible to scientifically evaluate the impact of equipment load changes on production scheduling plans.

[0004] From the perspective of dynamic adaptability in the production process, traditional scheduling methods are slow to respond to dynamic changes in the production process. During injection molding production, equipment status may suddenly become abnormal, material inventory may fluctuate, and order priorities may adjust due to changes in customer demand. Manual scheduling struggles to detect these changes in real time and adjust the production plan promptly, resulting in a disconnect between the production plan and actual production conditions, impacting production efficiency and product quality. For example, when equipment malfunctions, manual scheduling requires a long time to readjust production tasks, potentially disrupting the entire production process and causing significant financial losses.

[0005] Traditional methods lack automated and intelligent support for the generation and execution of production scheduling instructions. Scheduling instructions rely on manual compilation, which can be prone to ambiguity and inaccuracy. In addition, instruction delivery is inefficient, making it difficult to ensure timely and accurate execution of production plans at the production site. Furthermore, the inability to train and optimize scheduling models through optimization algorithms hinders the sustainable improvement of the scientific and rational nature of production scheduling. Summary of the Invention

[0006] The purpose of the present invention is to provide a method and system for automatically generating a production schedule based on the injection molding industry, so as to solve the problems raised in the above background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for automatically generating a production schedule based on injection molding industry production, the method comprising:

[0008] Step S1: Acquire production order data, equipment status data, and material inventory data for the injection molding industry; perform production parameter analysis on the production order data, equipment status data, and material inventory data to generate original production parameter data; and deploy a production scheduling network based on the original production parameter data to obtain a distributed production scheduling network.

[0009] Step S2: collecting and analyzing multi-level equipment load data on the distributed production scheduling network to generate equipment load impact factors;

[0010] Step S3: Sense the state of production equipment according to the distributed production scheduling network, obtain equipment state data, and generate equipment state influencing factors based on the equipment state data;

[0011] Step S4: collecting production order priority data according to the distributed production scheduling network to obtain order priority data; sampling and calibrating the order priority data to generate order calibration data; dividing production task areas based on the order calibration data to generate task area data;

[0012] Step S5: Based on the task area data, the order calibration data and the equipment status data are optimized and calculated for the production scheduling plan to generate a production scheduling optimization index; the production scheduling optimization index, the equipment load influencing factor and the equipment status influencing factor are integrated to generate comprehensive production scheduling data; a production scheduling instruction is generated for the comprehensive production scheduling data to generate production scheduling instruction data; an instruction is output for the production scheduling instruction data to execute automatic production scheduling operations.

[0013] Preferably, step S1 includes the following steps:

[0014] Step S11: Obtaining production order samples, equipment status samples, and material inventory samples for the injection molding industry;

[0015] Step S12: performing parameter analysis on the production order sample, equipment status sample, and material inventory sample to obtain production order data, equipment status data, and material inventory data;

[0016] Step S13: performing production parameter analysis on the production order data, equipment status data, and material inventory data to generate original production parameter data;

[0017] Step S14: Deploy a production scheduling network based on the original production parameter data to obtain a distributed production scheduling network.

[0018] Preferably, step S14 includes the following steps:

[0019] Step S141: performing production structure analysis on the original production parameter data to obtain production structure data;

[0020] Step S142: dividing the production area according to the production structure data to generate area boundary data;

[0021] Step S143: arranging production scheduling node positions based on the area boundary data to generate production scheduling node position data;

[0022] Step S144: deploying distributed production scheduling nodes using the multi-dimensional production scheduling node position data, and calibrating parameters of the deployed distributed production scheduling nodes to generate distributed calibrated production scheduling nodes;

[0023] Step S145: performing communication channel association on the distributed calibration scheduling nodes to generate a distributed scheduling network.

[0024] Preferably, step S2 includes the following steps:

[0025] Step S21: collecting multi-level equipment load data on the distributed production scheduling network to obtain multi-level load monitoring data, wherein the multi-level load monitoring data includes primary load monitoring data, intermediate load monitoring data, and advanced load monitoring data;

[0026] Step S22: performing multi-level load gradient calculation on the primary load monitoring data and the advanced load monitoring data to obtain a load gradient index; performing equipment load density difference impact analysis on the load gradient index based on the intermediate load monitoring data to generate load density difference data;

[0027] Step S23: performing a load fluctuation analysis on the load gradient index based on the load density difference data to generate load fluctuation data; performing a load fluctuation trajectory analysis on the load fluctuation data to generate load fluctuation trajectory data;

[0028] Step S24: analyzing the degree of equipment load variation on the load fluctuation data using the load fluctuation trajectory data to generate an equipment load impact factor.

[0029] Preferably, step S24 includes the following steps:

[0030] Step S241: performing load fluctuation time series analysis on the equipment load fluctuation data using the equipment load fluctuation trajectory data to generate equipment load fluctuation time series data; performing dynamic graph conversion on the equipment load fluctuation time series data to generate an equipment load fluctuation dynamic graph;

[0031] Step S242: performing equipment load abnormality event analysis on the equipment load fluctuation dynamic graph to generate equipment load abnormality event data;

[0032] Step S243: performing a normal load change trend analysis on the equipment load fluctuation dynamic graph based on the equipment abnormal load event data to generate equipment normal load trend change data;

[0033] Step S244: Detect the degree of impact of abnormal events on the normal load trend change data of the equipment using the abnormal load event data of the equipment to generate abnormal load impact degree data; perform equipment load change analysis on the equipment load fluctuation data based on the abnormal load impact degree data to generate an equipment load impact factor.

[0034] Preferably, step S3 includes the following steps:

[0035] Step S31: Sense the state of production equipment according to the distributed production scheduling network and generate equipment sensing data;

[0036] Step S32: collecting device parameters from the device sensing data, and performing device status identification on the collected device parameters to obtain device status data;

[0037] Step S33: matching the device status data with a preset device database; if the device status data does not match the preset device database, marking the corresponding device status data as abnormal device status data; if the device status data does match the preset device database, marking the corresponding device status data as normal device status data;

[0038] Step S34: performing an equipment operation evaluation on the abnormal equipment status data to generate abnormal equipment operation evaluation data; performing an equipment abnormality impact analysis on the normal equipment status data based on the abnormal equipment operation evaluation data to generate an equipment status impact factor.

[0039] Preferably, step S31 includes the following steps:

[0040] Step S311: collecting production equipment sensor data from the distributed production scheduling network to obtain equipment sensor data;

[0041] Step S312: performing data filtering on the device sensor data to generate device filtered data; performing signal enhancement on the device filtered data to generate device enhanced data;

[0042] Step S313: performing signal feature analysis on the device enhancement data to generate device signal feature data; marking production area devices using the device signal feature data to generate area device marking data;

[0043] Step S314: performing production equipment type identification on the regional equipment tag data to generate equipment type identification data; performing production equipment behavior perception based on the equipment type identification data to generate equipment perception data.

[0044] Preferably, step S4 includes the following steps:

[0045] Step S41: collecting production order priority data according to the distributed production scheduling network to obtain order priority data, wherein the order priority data includes order urgency data, order complexity data, and order delivery period data;

[0046] Step S42: performing production order spatiotemporal distribution feature analysis on the order priority data to generate order spatiotemporal distribution feature data; performing order data sampling and calibration on the order spatiotemporal distribution feature data to generate order calibration data;

[0047] Step S43: performing a production order status evaluation on the order calibration data to generate order status evaluation data; performing a distribution feature evaluation on the order calibration data to generate order distribution feature evaluation data;

[0048] Step S44: Divide the production task areas according to the order status evaluation data and the order distribution characteristic evaluation data to generate task area data.

[0049] Preferably, step S5 includes the following steps:

[0050] Step S51: performing a production control chain analysis on the order calibration data and the equipment status data based on the task area data to generate control chain data;

[0051] Step S52: performing production scheduling optimization calculation on the control chain data using a production scheduling optimization calculation formula to generate a production scheduling optimization index; integrating the production scheduling optimization index, equipment load influencing factors, and equipment status influencing factors to generate comprehensive production scheduling data;

[0052] The production scheduling optimization calculation formula is as follows:

[0053]

[0054] in, Indicates the optimized value of the production schedule; Indicates the upper limit of the production scheduling cycle; It represents the production scheduling control coefficient; represents the system stability coefficient; Indicates the order priority factor; represents the deviation factor; Indicates the current production scheduling time point;

[0055] Step S53: dividing the comprehensive production scheduling data into data sets to generate a model training set and a model test set; performing model training on the model training set using an optimization algorithm to generate a production scheduling instruction training model; and performing model optimization iteration on the production scheduling instruction training model using the model test set to generate a production scheduling instruction model;

[0056] Step S54: importing the comprehensive production scheduling data into the production scheduling instruction model to generate production scheduling instructions and generate production scheduling instruction data; outputting instructions to the production scheduling instruction data to execute automatic production scheduling operations.

[0057] Preferably, the present invention further includes a system for automatically generating a production schedule based on the injection molding industry, which is used to execute the above-mentioned method for automatically generating a production schedule based on the injection molding industry. The system includes:

[0058] The production scheduling network deployment module is used to obtain production order data, equipment status data, and material inventory data for the injection molding industry; perform production parameter analysis on the production order data, equipment status data, and material inventory data to generate original production parameter data; and perform production scheduling network deployment based on the original production parameter data to obtain a distributed production scheduling network;

[0059] An equipment load analysis module is configured to collect multi-level equipment load data from the distributed production scheduling network to obtain load monitoring data; perform multi-level load gradient calculations on the load monitoring data to obtain a load gradient index; perform load fluctuation analysis on the load gradient index to generate load fluctuation data; perform load fluctuation trajectory analysis on the load fluctuation data to generate load fluctuation trajectory data; and perform equipment load change degree analysis on the load fluctuation data using the load fluctuation trajectory data to generate an equipment load impact factor.

[0060] An equipment status analysis module is configured to sense the status of production equipment based on the distributed production scheduling network and generate equipment perception data; collect equipment parameters from the equipment perception data and perform equipment status identification on the collected equipment parameters to obtain equipment status data; perform equipment abnormality impact analysis on the equipment status data to generate equipment status impact factors;

[0061] A production task area analysis module is configured to collect production order priority data based on the distributed production scheduling network to obtain order priority data; perform order data sampling and calibration on the order priority data to generate order calibration data; and divide production task areas based on the order calibration data to generate task area data;

[0062] A production scheduling instruction generation module is used to perform production scheduling optimization calculations on the order calibration data and equipment status data based on the task area data to generate a production scheduling optimization index; integrate the production scheduling optimization index, equipment load influencing factor and equipment status influencing factor to generate comprehensive production scheduling data; generate production scheduling instructions for the comprehensive production scheduling data to generate production scheduling instruction data; and output instructions for the production scheduling instruction data to execute automatic production scheduling operations.

[0063] Compared with the prior art, the present invention has the following beneficial effects:

[0064] In terms of data processing and scheduling network construction, the system automatically acquires production order data, equipment status data, and material inventory data. It then generates raw production parameter data through production parameter analysis, thereby deploying a distributed scheduling network. This process enables comprehensive integration and in-depth analysis of production-related data, breaking the data isolation and inefficient processing found in traditional manual scheduling. By analyzing the production structure, zoning, and deploying scheduling nodes based on raw production parameter data, a distributed scheduling network covering the entire production process is constructed. This allows production data to be circulated and shared across nodes in real time, providing a solid data foundation and efficient network support for subsequent production scheduling, significantly improving the scientific nature and rationality of production scheduling.

[0065] In terms of equipment load management, the system acquires primary, intermediate, and advanced load monitoring data through multi-level equipment load data collection. It then performs load gradient calculations, load density difference impact analysis, load fluctuation analysis, and load fluctuation trajectory analysis, ultimately generating equipment load impact factors. This multi-level, multi-dimensional analysis approach accurately captures subtle changes and fluctuation trends in equipment loads, allowing for early detection of abnormal equipment loads. For example, by analyzing load fluctuation trajectory data, it is possible to predict equipment load changes over a period of time, allowing for timely adjustments to production schedules, balance equipment loads, avoid overloading or idle equipment, improve equipment utilization and production efficiency, and reduce the risk of equipment failure and repair costs caused by abnormal loads.

[0066] Equipment status perception and analysis is another highlight of this invention. The system achieves accurate perception of equipment status through a series of steps, including production equipment sensor data collection, data filtering, signal enhancement, and signal feature analysis. By matching equipment status data with a preset equipment database, it can quickly and accurately identify whether the equipment status is normal, conduct operational evaluations on abnormal equipment status data, and analyze its impact on normal equipment status to generate equipment status influencing factors. This enables the system to grasp the operating status of the equipment in real time, promptly discover potential equipment failure hazards, perform maintenance and care in advance, reduce production interruptions caused by equipment failures, ensure production continuity and stability, and at the same time extend the service life of the equipment and reduce equipment management costs.

[0067] In terms of order management and task zone division, the system analyzes the spatiotemporal distribution characteristics, samples, and calibrates order priority data (including order urgency, complexity, and delivery time), generates order calibration data, and then divides production task zones based on order status and distribution characteristics. This data-driven task zone division fully considers key order factors and can group orders with similar characteristics into the same task zone, achieving rational allocation and optimal configuration of production resources. For example, orders with high urgency and tight delivery deadlines can be assigned to a task zone, prioritizing production and ensuring on-time delivery. Highly complex orders can also be assigned to a task zone, concentrating superior equipment and human resources on production, improving production quality and efficiency. This rational and rational task zone division helps improve production coordination and flexibility, better meeting the market's diverse demand for order delivery.

[0068] Production scheduling optimization and instruction generation are one of the core advantages of the present invention. Based on task area data, the system generates a production scheduling optimization index through production control chain analysis and production scheduling optimization calculation formula, and integrates equipment load influencing factors and equipment status influencing factors to generate comprehensive production scheduling data. The model training set and test set are trained and optimized through the optimization algorithm to generate a production scheduling instruction model, and finally the production scheduling instruction data is generated and output to execute the automatic production scheduling operation. This process realizes the dynamic optimization and intelligent generation of the production scheduling plan, and can find the optimal production scheduling plan under the combined influence of various factors. The production scheduling optimization calculation formula fully considers various factors such as the production scheduling control coefficient, system stability coefficient, order priority factor, deviation factor, etc., and can adjust and optimize the production scheduling plan according to real-time data, thereby improving the adaptability and accuracy of the production scheduling plan. Through model training and optimization iteration, the production scheduling instruction model can continuously learn and accumulate experience, improve the generation accuracy and efficiency of production scheduling instructions, realize the automation and intelligence of injection molding production scheduling, greatly reduce manual intervention, reduce labor costs, and improve production efficiency and management level. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 This is a working principle diagram of the method for automatically generating a production schedule based on injection molding industry production according to the present invention;

[0070] Figure 2 Design drawings for production scheduling network deployment;

[0071] Figure 3 A design diagram for distributed scheduling node deployment. DETAILED DESCRIPTION

[0072] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0073] See also Figure 1-Figure 3 The present invention relates to a method for automatically generating a production schedule based on the injection molding industry, and the specific implementation steps are as follows:

[0074] Step S1: Obtain production order data, equipment status data, and material inventory data for the injection molding industry. By analyzing this data for production parameters, raw production parameter data is generated. This data is then used to deploy a production scheduling network and build a distributed production scheduling network. Production order data includes key information such as order quantity, product specifications, and delivery time; equipment status data includes equipment operating parameters, failure rates, and maintenance cycles; and material inventory data includes information such as raw material types, inventory levels, and procurement cycles.

[0075] Step S2: Implement multi-level equipment load data collection and analysis for the distributed production scheduling network. By collecting load data at different levels (e.g., primary, intermediate, and advanced load monitoring data), analyze the gradient changes, density differences, and fluctuation trajectories of equipment load, and generate equipment load impact factors to quantify the impact of equipment load on production scheduling.

[0076] Step S3: Real-time monitoring of production equipment status is performed through the distributed scheduling network, acquiring equipment status data (e.g., normal operation or abnormality). Based on the matching results of the equipment status data with the preset equipment database, an equipment status impact factor is generated to assess the potential impact of the equipment status on the production plan.

[0077] Step S4: Collect production order priority data (including order urgency, complexity, delivery period, etc.), generate order calibration data through data sampling and calibration, and divide production task areas based on this data to clarify the production goals and scope of each task area, providing a basis for subsequent production scheduling optimization.

[0078] Step S5: Based on the task area data, the order calibration data and equipment status data are optimized and calculated to generate a production scheduling optimization index. The production scheduling optimization index is integrated with the equipment load influencing factors and equipment status influencing factors to form comprehensive production scheduling data. Then, production scheduling instruction data is generated and output to realize the execution of automatic production scheduling operations.

[0079] The present invention will be further described below in conjunction with Examples 1 to 5:

[0080] Example 1:

[0081] This embodiment refines step S1, and the specific implementation method is as follows:

[0082] In the data acquisition phase, production order samples, equipment status samples, and material inventory samples from the injection molding industry are collected through the company's internal management information system or manual input interface. Production order samples contain complete customer order information, such as product name, specification model, required quantity, delivery date, quality standards, etc. This data directly reflects the customer's production needs; equipment status samples cover real-time data on injection molding machines, molds, auxiliary equipment, etc. during operation, including equipment number, operating time, temperature, pressure, energy consumption, fault alarm records, etc., which are used to evaluate the availability and performance status of the equipment; material inventory samples include the type of raw materials (such as polypropylene, polyethylene, etc.), inventory quantity, unit cost, safety stock threshold, procurement cycle, supplier information, etc., to ensure a stable supply of raw materials during the production process.

[0083] After acquiring the aforementioned sample data, parameter analysis is required to extract meaningful information. Specifically, for production order samples, data cleansing techniques are used to remove duplicate records, invalid orders, or anomalous data points. For example, orders with expired and unfulfilled delivery dates are deleted. Valid and executable orders are retained, and the remaining data is structured into production order data, including fields such as the order's unique identifier, product attributes, quantity, and delivery time. For equipment status samples, filtering algorithms are used to remove noise from sensor data, such as using a sliding average filter to process temperature fluctuations. Equipment operating parameters are also normalized to convert parameters from different devices to a unified standard, generating equipment status data including device identification, real-time parameters, operating status (normal / abnormal), and maintenance status. Material inventory samples require verification of inventory records against actual inventory, updating inbound and outbound records, and calculating the difference between current inventory and safety stock. This generates material inventory data, including material name, specification, on-hand inventory, minimum inventory, and procurement lead time.

[0084] After completing basic data processing, the production parameter analysis phase begins. Integrating the production process characteristics of the injection molding industry, in-depth analysis is conducted on production order data, equipment status data, and material inventory data. For example, for production order data, the injection molding process parameters required for each order, such as mold type, injection temperature, pressure setting, cooling time, and production cycle, are analyzed. Combined with product specifications and quality standards, a production process parameter table is generated. For equipment status data, the capacity utilization rate, mean time between failures (MTBF), and maintenance cycle of each device are analyzed to assess the equipment's production capacity and reliability. Material inventory data is used to analyze the consumption rate and inventory turnover rate of raw materials. Combined with the material requirements of production orders, the material requirement plan (MRP) for each order is calculated, generating material consumption parameters and inventory warning thresholds. Through this analysis, raw production parameter data containing key information such as production processes, equipment requirements, and material requirements is integrated to form the foundation for subsequent production scheduling network deployment.

[0085] During the production scheduling network deployment phase, the original production parameter data is first analyzed for production structure. By analyzing each link in the production process, the critical paths and dependencies of order production are identified, such as the sequence of injection molding, cooling and setting, deburring, quality inspection, and packaging, as well as the equipment and material requirements for each process. Furthermore, the possibility of parallel production between different orders is analyzed, such as whether the same mold or equipment can handle multiple orders simultaneously. This generates production structure data reflecting production logic and resource dependencies, and presents the connection between each link in the form of charts or flow charts.

[0086] Production areas are divided based on production structure data. According to the physical layout and process characteristics of injection molding production, the entire production system is divided into multiple production areas with clear functions, such as the injection molding area (including multiple injection molding machines), the cooling area (for product cooling and shaping), the quality inspection area (product quality inspection), the packaging area (finished product packaging), the raw material warehouse, and the finished product warehouse. Each area has clear boundaries and functional positioning. For example, the injection molding area is responsible for product molding production, the cooling area is responsible for controlling the temperature and time of the product cooling process, and the quality inspection area is responsible for checking whether the product size, appearance, and performance meet the standards. After the division is completed, regional boundary data containing the name, location, functional description, and equipment list of each area is generated to provide a spatial reference for the subsequent production scheduling node layout.

[0087] The production scheduling nodes are arranged based on the regional boundary data. In each production area, the specific location and function of the production scheduling nodes are determined according to the equipment distribution, process route and logistics path. For example, in the injection molding area, each injection molding machine serves as an independent production scheduling node, and the node parameters include equipment number, model, maximum production capacity, available time window, etc.; in the cooling area, the production scheduling nodes are set according to the cooling equipment or cooling line, and the node parameters include cooling capacity, processing time, temperature control range, etc.; in the quality inspection area, the quality inspection station is used as the node, and the parameters include quality inspection efficiency, inspection standards, production capacity limit, etc. By rationally arranging the production scheduling nodes, it is ensured that each production link has a corresponding node for data collection and task scheduling, and the production scheduling node location data containing node identification, location, type, and parameters is generated to form a spatial layout of distributed production scheduling nodes.

[0088] After completing the node layout, distributed production scheduling node deployment and calibration are carried out. Through Internet of Things (IoT) technology or industrial networks, each production scheduling node is connected to the distributed system to achieve data communication and collaborative work between nodes. For each node, its parameter settings are initialized, such as the default values ​​of the production process parameters of the injection molding machine node and the temperature setting range of the cooling node, and the accuracy of the parameters is verified through real-time data collection. At the same time, communication tests are carried out between nodes to ensure the stability and real-time performance of data transmission. For example, the production progress data transmission between the injection molding machine node and the cooling node is tested for delays or losses. Through parameter calibration and communication testing, distributed calibrated production scheduling nodes are generated to ensure that each node can accurately receive and execute production scheduling instructions and provide real-time feedback of status data.

[0089] Associate the communication channels of the distributed calibration scheduling nodes. Establish logical connections between nodes, and define the direction and rules of data flow based on the sequence and dependencies of the production process. For example, after the injection molding machine node completes production, it transmits product information to the cooling node, and after the cooling node completes processing, it passes the data to the quality inspection node; the raw material warehouse node sends a material preparation notification to the injection molding area node based on the material requirements of the production order. By setting different communication protocols (such as OPCUA, MQTT, etc.) and data interfaces, information interaction between heterogeneous devices and systems is achieved, and a distributed scheduling network covering all nodes is generated. The network has the functions of real-time data collection, task allocation, status monitoring and collaborative scheduling, and can dynamically respond to changes in the production process, providing an efficient data transmission and processing platform for subsequent equipment load analysis, equipment status perception and order priority processing.

[0090] Example 2:

[0091] This embodiment refines step S2, and the specific implementation method is as follows:

[0092] In the multi-level equipment load data collection phase, sensors, data collectors, and other devices deployed in the distributed scheduling network periodically monitor the load of various types of production equipment in the injection molding production system. Load data collection covers the entire lifecycle of equipment operation, including the load status of the equipment during different production stages (such as standby, injection molding, cooling, and maintenance). Primary load monitoring data collects equipment load information in non-production or basic operating states, such as the energy consumption of injection molding machines in standby mode and the basic computing power utilization of the control system. Intermediate load monitoring data collects the staged load of equipment during routine production tasks, such as the pressure and temperature fluctuation data of injection molding machines during normal injection cycles, as well as the average operating load of the equipment during continuous production. Advanced load monitoring data collects equipment load data during peak production tasks or abnormal operating conditions, such as the parameters of equipment operating at full load when processing highly complex orders, and load mutation data before sudden failures. By setting different collection frequencies (such as primary load data is collected once a minute and advanced load data is collected once a second), it is ensured that the multi-level load monitoring data can fully reflect the load characteristics of the equipment in different scenarios, forming a multi-level load monitoring data set containing timestamps, equipment identification, load type, and parameter values.

[0093] After acquiring multi-level load monitoring data, multi-level load gradient calculation and density difference analysis are performed. For both primary and advanced load monitoring data, load gradient values ​​are calculated for different time periods. This is the ratio of load change to time interval. This quantifies the rate of change in equipment load from baseline to peak. For example, when an injection molding machine switches from standby mode to full-load production, its motor current increases from 5A to 40A over a period of 2 minutes. The load gradient value is (40-5) / 2 = 17.5A / minute. By statistically analyzing the load gradient values ​​of multiple devices, a load gradient index is generated to reflect the sensitivity of the device to load changes. Furthermore, based on intermediate load monitoring data, load density differences between different devices and within the same device over different time periods are analyzed. Load density is calculated as the ratio of the effective production task volume per unit time (e.g., the number of qualified products completed per hour) to the device's maximum production capacity. For example, a load density of 70% for a device during a certain time period indicates that its actual capacity utilization has reached 70%. By comparing the load density of different devices, we can identify devices with excessively high or low loads and generate load density difference data. For example, we found that the load density of device No. 3 in the injection molding area reached 90%, while that of device No. 5 was only 40%, indicating an uneven distribution of equipment resources.

[0094] Load fluctuation analysis and trajectory construction are performed in conjunction with load density difference data. Load fluctuation analysis assesses equipment load stability by calculating statistical quantities such as the standard deviation and coefficient of variation of load data. For example, if the standard deviation of pressure values ​​for a piece of equipment over 10 consecutive production cycles is 5 bar, this indicates minimal load fluctuation and high operational stability. However, if the standard deviation reaches 15 bar, this indicates significant load fluctuation, potentially indicating unstable process parameters or equipment failure. Based on the fluctuation statistics, load fluctuation data is generated, recording the amplitude, frequency, and occurrence time of load fluctuations for each piece of equipment. Trajectory analysis is then performed on the load fluctuation data, plotting a load fluctuation trajectory curve for each piece of equipment with time as the horizontal axis and load value as the vertical axis. The trajectory curve visually illustrates load trends over time, such as whether there are periodic fluctuations, sudden increases, or decreases. By comparing the trajectory curves of different pieces of equipment, common fluctuation patterns or abnormal fluctuation points can be identified, generating load fluctuation trajectory data that provides a visual basis for subsequent analysis of the extent of equipment load variation.

[0095] When analyzing the degree of equipment load variation, we first perform a time series analysis of the load fluctuation trajectory data to identify the time series characteristics of the load fluctuations. For example, some equipment may experience a load peak around 10:00 AM each day, corresponding to the concentrated order placement period in the production plan. Alternatively, the load fluctuations of other equipment may exhibit an hourly periodicity, coinciding with the cooling cycle of the injection molding process. By extracting time series characteristics (such as cycle length, peak time, and valley time), we generate time series data for equipment load fluctuations and convert this data into a dynamic chart, presenting an animated representation of the load evolution over time, making it easier to visually observe the load variation patterns.

[0096] Abnormal equipment load events are detected based on load fluctuation dynamic graphs. By setting abnormal thresholds (such as load values ​​exceeding 110% or falling below 30% of the equipment's rated capacity), load events outside the normal range are automatically identified. For example, if the pressure parameter of an injection molding machine suddenly exceeds 120% of the rated pressure and persists for 5 minutes, the system automatically marks it as an abnormal load event, records the time of occurrence, duration, equipment involved, and load parameter values, and generates abnormal equipment load event data. Abnormal events may be caused by urgent order scheduling, equipment failure, incorrect process parameter settings, and other reasons. Further investigation of the cause is required to incorporate it into the basis for production scheduling adjustments.

[0097] After identifying abnormal load events, the load fluctuation dynamic graph is analyzed for normal load change trends. By removing abnormal event data points and performing curve fitting (such as polynomial fitting or exponential smoothing) on ​​the remaining data, normal load trend data for the equipment is generated, reflecting the load variation patterns of the equipment under normal conditions. For example, the normal load trend of a piece of equipment may show a gradual increase with extended production time, associated with mold wear or changes in material properties, or it may exhibit stable cyclical fluctuations consistent with a fixed production cycle. Normal load trend data provides a benchmark for determining the equipment's operating status and helps distinguish normal from abnormal fluctuations.

[0098] Abnormal load event data is used to assess the impact of abnormal events on normal load trends. The deviation between the load value during the abnormal event and the normal trend value, as well as the cumulative impact of the deviation duration on production efficiency, are calculated to generate abnormal load impact data. For example, an abnormal load event causes equipment to shut down for maintenance for 2 hours, and the backlog of production tasks during this period needs to be completed overtime. The deviation is 2 hours of production capacity loss. Based on the abnormal load impact data and combined with the normal load trend change data, the overall degree of change in the equipment load is comprehensively analyzed to generate the equipment load impact factor. This factor uses a quantitative value (such as 0-10 points) to represent the level of impact of load changes on the production schedule. The higher the value, the greater the impact, providing key input parameters for production schedule optimization.

[0099] Example 3:

[0100] This embodiment refines step S3, and the specific implementation method is as follows:

[0101] In the production equipment status sensing phase, a sensor cluster deployed within the distributed scheduling network collects real-time operational data from injection molding equipment. Sensor types include temperature, pressure, vibration, and current sensors, installed in key locations such as the injection molding machine body, mold, motor, and hydraulic system. In step S311, raw equipment sensor data is transmitted in real time to the data acquisition module in the form of analog or digital signals. For example, the injection molding machine barrel temperature sensor outputs a 4-20mA analog signal, while the vibration sensor outputs a MEMS-based digital acceleration signal. This generates a raw data set containing noise and interference. The data format includes fields such as timestamp, sensor ID, and raw signal value.

[0102] Step S312 preprocesses the raw sensor data. First, a digital filtering algorithm is used to remove noise. For example, a Butterworth low-pass filter is used to filter high-frequency noise, retaining low-frequency signal components that reflect the actual operating status of the device, and generating device filtered data. Taking vibration signals as an example, filtering can eliminate environmental vibration interference and highlight the characteristic frequency components generated by equipment bearing wear or gear meshing. Signal enhancement processing is then performed. A normalization algorithm is used to unify the output value ranges of different sensors to the [0,1] range. For example, the 0-200°C range of a temperature sensor is mapped to a value of 0-1. At the same time, adaptive gain amplification technology is used to improve the signal-to-noise ratio of weak signals, generating device enhanced data to facilitate subsequent feature analysis.

[0103] Step S313 performs feature extraction and device labeling on the enhanced equipment data. Signal feature analysis uses a combination of time domain analysis, frequency domain analysis, and time-frequency domain analysis to extract characteristic parameters such as effective value, peak value, frequency component, kurtosis, and variance. For example, a Fourier transform is performed on the current signal of the injection molding machine motor to extract the fundamental frequency and harmonic components to determine whether the motor has a rotor fault; time domain statistics are performed on the hydraulic system pressure signal to calculate the mean and standard deviation to evaluate pressure stability. Based on the extracted characteristic parameters and combined with preset equipment feature templates (such as the normal vibration frequency range of different models of injection molding machines), the equipment in the production area is uniquely labeled, and regional equipment labeling data containing the equipment number, installation location, and characteristic parameter vector is generated to achieve precise positioning and identification of the equipment.

[0104] Step S314 identifies the device type based on the regional device tag data. By comparing the device's characteristic parameters with a built-in device type database (which stores characteristic templates for different injection molding machines, molds, and auxiliary equipment models), a pattern matching algorithm (such as template matching or neural network classification) is used to identify the device type and generate device type identification data. For example, this identifies a device as a "Haitian MA2800 injection molding machine" or a "Tairui TR180 servo injection molding machine." Based on this device type identification, the system integrates the injection molding production process knowledge base to detect the device's real-time behavior status, such as determining whether the device is in the "injection stage," "cooling stage," "standby state," or "fault shutdown state." This generates device perception data containing the device ID, behavior type, and timestamp, providing real-time information for device status assessment.

[0105] Step S32 extracts key equipment parameters from the device sensing data, such as the injection molding machine's clamping force, injection speed, melt temperature, cooling time, mold usage and wear, and the equipment's cumulative operating time and downtime. These parameters are analyzed in real time using pre-set status assessment rules, including threshold determination (e.g., whether the melt temperature exceeds a set range of ±5°C), trend analysis (e.g., whether the equipment vibration value has increased for three consecutive cycles), and correlation analysis (e.g., the correlation between injection pressure and product defect rate). If the parameters meet the rules, the equipment status is determined to be "normal"; if the parameters are abnormal, they are determined to be "abnormal." Ultimately, device status data is generated, including the device ID, status type (normal / abnormal), and the name of the abnormal parameter.

[0106] Step S33 matches the device status data with a pre-set device database. This database contains information on normal operating parameter ranges, historical failure cases, and maintenance records for various types of equipment. For example, the normal clamping force range for a certain injection molding machine is 1500-1800 kN, and the permissible peak vibration acceleration is below 10 m / s². During the matching process, if all parameter values ​​in the device status data are within the corresponding normal range in the database, the data is marked as "normal device status data." If any parameter is out of range or matches a historical failure signature (e.g., the vibration frequency matches the characteristic frequency of bearing wear), the data is marked as "abnormal device status data," triggering an early warning mechanism.

[0107] Step S34 conducts an in-depth evaluation of the abnormal device status data. Using a fault diagnosis algorithm (such as a rule-based expert system or machine learning model), the correlation between abnormal parameters is analyzed to trace the cause of the abnormality. For example, an abnormal melt temperature may be caused by a heater coil failure, thermostat failure, or excessive raw material moisture. This generates abnormal device operation evaluation data containing the device ID, the cause of the abnormality, and the impact level. Based on this evaluation data, the potential impact of the abnormal device on surrounding equipment or the production process is analyzed. For example, a failure in an injection molding machine may cause the subsequent cooling line to stop waiting for material, or require adjustments to the production tasks of adjacent equipment to balance production capacity. This generates an equipment status impact factor. This factor is calculated using the following formula:

[0108]

[0109] in, Indicates the device status influencing factor, Indicates the number of exception parameters, Indicates the The weight of each abnormal parameter (set according to the impact of the parameter on production, ranging from 0.1 to 1.0), Indicates the The severity value of each abnormal parameter (range 1-10, the larger the value, the more serious the impact). For example, a device has two abnormal parameters, the melt temperature has an abnormal weight of 0.8 and a severity of 8, and the vibration value has an abnormal weight of 0.6 and a severity of 6. The device status impact factor is , indicating that the exception has a significant impact on the production schedule and needs to be handled as a priority.

[0110] Example 4:

[0111] This embodiment refines step S4 and describes it in detail with reference to specific examples as follows:

[0112] In the production order priority data collection phase, the distributed scheduling network is connected to the enterprise order management system to obtain production order information in real time. For example, for an injection molding company, the orders received on a certain day include:

[0113] Order A: The customer is an automotive parts manufacturer and needs 10,000 instrument panel covers with a delivery time of 3 days. The product structure includes 5 inserts and requires the use of high-precision molds;

[0114] Order B: The customer is a home appliance company that needs 50,000 air conditioner panels with a 7-day delivery time. The product is a standard part using a universal mold.

[0115] Order C: The customer is a medical equipment manufacturer and needs 2,000 medical test tube racks with a delivery time of 5 days. The product must comply with the ISO13485 standard and the process must include sterilization. Based on the order attributes, extract the order priority data:

[0116] Order urgency data: Order A has a delivery period of 3 days, with 72 hours remaining until the current delivery date, indicating high urgency; Order B has 168 hours remaining, indicating medium urgency; Order C has 120 hours remaining, indicating medium-high urgency.

[0117] Order complexity data: Order A includes insert injection molding, which requires multi-station molds and precise temperature control, and is highly complex; Order B is a standard part with low complexity; Order C requires sterilization and is medium-to-high complexity.

[0118] Order delivery period data: The delivery time nodes of orders A, B, and C are 17:00 on June 13, 2025, 12:00 on June 17, and 10:00 on June 15, 2025, respectively.

[0119] Step S42 analyzes the temporal and spatial distribution characteristics of the order priority data. For example, the production area is divided into injection molding area 1 (equipped with high-precision mold injection molding machines), injection molding area 2 (equipped with general mold injection molding machines), and post-processing area (including sterilization, packaging, etc.):

[0120] Time distribution characteristics: Orders A and C need to be completed within 3-5 days, concentrated in the short production window; Order B has a longer cycle and can be arranged flexibly.

[0121] Spatial distribution characteristics: Order A requires high-precision equipment in Injection Molding Area 1, Order B can be produced in Injection Molding Area 2, and Order C requires equipment in Injection Molding Area 1 and sterilization equipment in the post-processing area. Analysis generates temporal and spatial distribution characteristic data for orders. For example, Order A's equipment demand period in Injection Molding Area 1 is June 10th-13th, while Order C's demand period in Injection Molding Area 1 is June 10th-14th, suggesting potential equipment competition between the two. The estimated capacity utilization rate for Order B in Injection Molding Area 2 is 60%, indicating capacity redundancy.

[0122] Sampling and calibration of spatiotemporal distribution feature data:

[0123] Data sampling: Stratified sampling is performed based on order urgency, with complete data for orders with high urgency (such as Order A) being prioritized, followed by orders with medium to high urgency (Order C), and finally orders with medium urgency (Order B).

[0124] Data calibration: Mark the task type for each order, such as order A is marked as "high-precision injection molding + insert assembly", order B is marked as "standard injection molding", and order C is marked as "medical-grade injection molding + sterilization". Order calibration data containing order ID, spatiotemporal characteristics, and task type is generated.

[0125] Step S43 performs multi-dimensional evaluation on the order calibration data:

[0126] Order status evaluation:

[0127] Order A: Production has not started yet, the remaining time is tight, and production needs to be scheduled first;

[0128] Order B: Raw materials are in stock and can be started as planned;

[0129] Order C: The sterilization process resources (such as a sterilizer) are currently 80% occupied, requiring equipment scheduling. Order status evaluation data is generated, marking Order A as "High Risk Overdue," Order C as "Requires Resource Coordination," and Order B as "Normal."

[0130] Distribution characteristics evaluation:

[0131] In Injection Molding Area 1, the equipment demand times for Orders A and C overlapped between June 10th and 13th, resulting in an estimated equipment load rate of 120% and a resource conflict.

[0132] During the same period, the load rate of Injection Molding Zone 2 was only 50%, resulting in wasted production capacity. This generated order distribution feature evaluation data, identifying Injection Molding Zone 1 as a "high-load area" and Injection Molding Zone 2 as a "low-load area."

[0133] Step S44 divides the production task areas based on the evaluation results:

[0134] Classification by order urgency and process type:

[0135] Urgent high-precision task area: This includes orders A and C, and prioritizes the high-precision equipment in Injection Molding Area 1. Order A, due to its more urgent delivery date, will be scheduled for production on Injection Molding Area 1's Machine 1 from 8:00 AM on June 10th to 8:00 PM on June 12th. Order C will be rescheduled to Injection Molding Area 1's Machine 2 from 2:00 PM on June 11th to 10:00 AM on June 14th to avoid equipment conflicts with Order A.

[0136] Standard task area: includes order B, assigned to injection molding area 2 No. 3 and No. 4 machines, and flexibly arrange production by utilizing production capacity redundancy. The time window is from 14:00 on June 10 to 24:00 on June 16, and production is carried out in batches to balance daily production capacity.

[0137] By resource requirements:

[0138] Sterilization task area: In response to the sterilization needs of Order C, sterilization cabinet resources will be reserved from 8:00 to 10:00 on June 14 to ensure connection with the injection molding production process.

[0139] Mission area boundary definition:

[0140] The production target for the urgent high-precision task area is to "complete Order A within 3 days and Order C within 5 days." Equipment is limited to injection molding machines No. 1 and No. 2 in Area 1, and materials are delivered by dedicated personnel on a priority basis.

[0141] The production target of the standard task area is "complete order B within 7 days". The equipment (machines 3-6 in injection molding area 2) can be adjusted dynamically, and materials can be collected in stages.

[0142] By dividing the task area, the task area data including the task area ID, order members, equipment list, time window, and resource requirements are generated. For example:

[0143] Mission Area 1 (Urgent High Precision): Orders A and C; Equipment: Injection Molding Area 1, Machines 1 and 2; Time: June 10-14; Resources: High-precision molds, insert assembly tooling, sterilization cabinet (order C only).

[0144] Mission Area 2 (Standard): Order B; Equipment: Injection Molding Area 2, Machines 3-6; Time: June 10-16; Resources: General molds, conventional packaging materials.

[0145] After the task areas are divided, each becomes an independent production unit, allowing the production scheduling system to develop differentiated strategies for different task types. For example, the urgent, high-precision task area uses a strategy of real-time monitoring of equipment status and prioritizing abnormal alarms to ensure on-time order delivery; the standard task area adopts a strategy of batch production and dynamic adjustment of production rhythm to improve equipment utilization.

[0146] Example 5:

[0147] This embodiment refines step S5 and describes it in detail with reference to a specific example as follows:

[0148] Take the production task of an injection molding company on a certain day as an example, which includes 3 orders:

[0149] Order X (high-precision parts): Requires 5,000 pieces, with a 4-day delivery time, and requires injection molding machine A (high-precision equipment with a production capacity of 1,000 pieces / day).

[0150] Order Y (standard parts): 12,000 pieces required, 7-day delivery, using injection molding machine B (general-purpose equipment with a production capacity of 2,000 pieces per day).

[0151] Order Z (complex structural parts): 3,000 pieces required, 5-day delivery, injection molding machine C (multi-functional equipment, 600 pieces / day production capacity) required.

[0152] Step S51: Based on the task area division results (assuming that orders X and Z are divided into the "high-precision task area" and order Y is divided into the "standard task area"), the production control chain analysis is performed on the order calibration data and equipment status data. Taking order X as an example, the control chain includes:

[0153] Material preparation: Raw material A must be collected one day in advance (inventory quantity 8,000 kg, safety stock 5,000 kg);

[0154] Injection molding process: Injection molding machine A needs to be preheated for 2 hours, with a production cycle of 15 minutes per piece and a daily operation time of 20 hours (reserving 4 hours for maintenance);

[0155] Cooling process: The processing capacity of cooling equipment D is 800 pieces / day and needs to be synchronized with the injection molding progress;

[0156] Quality inspection: Quality inspection station E processes a maximum of 1,200 pieces per day, and the inspection must be completed within 24 hours after the injection molding is completed;

[0157] Packaging: Packaging Line F has a daily production capacity of 3,000 pieces and must start immediately after passing quality inspection. By analyzing the chronological order and dependencies of each step, control chain data is generated, and a flowchart is created to indicate each step's prerequisite tasks, duration, resource requirements, and time constraints.

[0158] Step S52: Calculate the control chain data using the production scheduling optimization calculation formula. The formula is:

[0159]

[0160] in, Indicates the optimized value of the production schedule; Indicates the upper limit of the production scheduling cycle; It represents the production scheduling control coefficient; represents the system stability coefficient; Indicates the order priority factor; represents the deviation factor; Indicates the current production scheduling time point;

[0161] Calculation process:

[0162] This indicates that the production schedule for order X needs to be adjusted first to reduce the optimization value.

[0163] The calculated production scheduling optimization index (e.g., order X is 0.113, order Y is 0.35, and order Z is 0.28) is integrated with the equipment load influencing factors (injection molding machine A load rate is 90%, factor 0.8; injection molding machine B load rate is 60%, factor 0.4; injection molding machine C load rate is 85%, factor 0.7) and the equipment status influencing factors (injection molding machine A has no abnormality, factor 0; injection molding machine C vibration value has a slight abnormality, factor 0.2) to form comprehensive production scheduling data, as shown in the following table:

[0164]

[0165] Step S53: Divide the comprehensive production scheduling data into a model training set (70% data, such as historical production scheduling data for orders X and Y) and a model test set (30% data, such as data for the current order Z). A genetic algorithm is used to train the model on the training set, with a population size of 50, 100 iterations, a crossover probability of 0.8, and a mutation probability of 0.05. During training, the algorithm automatically adjusts the production scheduling sequence, equipment allocation, and time window to minimize the optimal value in the comprehensive production scheduling data. For example, for the historical data for orders X and Y, the algorithm may generate two solutions:

[0166] Solution 1: Order X is produced first (4 days), followed by Order Y (6 days), for a total cycle of 10 days. The utilization rates of equipment A and B are 100% and 85%, respectively.

[0167] Solution 2: Parallel production of order X (first 3 days) and order Y (7 days). Machine A processes order X first and then supports order Y. The total cycle time is 7 days. The utilization rate of machine A is 90%, and the utilization rate of machine B is 100%.

[0168] By comparing the combined optimization values ​​of the two solutions, the superior solution was selected as the parameter baseline for the training model. The model was then validated using the test set (order Z), and the parameters were adjusted to reduce the prediction error to less than 5%, generating the final scheduling instruction model.

[0169] Step S54: Import the comprehensive production scheduling data into the production scheduling instruction model to generate specific production scheduling instructions. Taking order X as an example, the model output is:

[0170] Equipment allocation: Injection molding machine A starts preheating at 14:00 on day 0 and officially starts production at 0:00 on day 1, running for 20 hours per day, and is completed in 5 days (due to the actual production capacity being affected by maintenance, the original plan of 4 days was adjusted to 5 days);

[0171] Material distribution: Raw material A is delivered to injection molding machine station A at 10:00 on day 0, supplied in batches (1000 kg per day);

[0172] Process connection: Cooling equipment D starts at 8:00 on the first day and receives injection molded parts simultaneously; quality inspection station E starts testing the first batch of products at 16:00 on the first day; packaging line F starts packaging qualified products at 00:00 on the second day.

[0173] The command data is pushed to each device terminal via the Manufacturing Execution System (MES). Upon receiving it, the control system of injection molding machine A automatically loads the process parameters (temperature 220°C, pressure 120 bar), and cooling device D simultaneously adjusts the temperature control curve. During production, the system monitors the progress of order X in real time. On the third day, the hydraulic oil temperature of injection molding machine A rises abnormally (85°C, exceeding the warning value of 80°C). This triggers the equipment status impact factor (updated to 0.3), and the model automatically adjusts the subsequent production plan: production is suspended at 6:00 PM on the third day, and a maintenance team is dispatched to address the situation within two hours. The remaining production capacity is increased by 200 units per day starting on the fourth day (by extending the operating time to 22 hours) to ensure delivery on time.

[0174] Through the above process, this embodiment achieves full automation, from control chain analysis and optimization calculations to instruction generation. By dynamically adjusting the production scheduling strategy based on the process requirements and equipment status of specific orders through model training, this not only resolves resource conflicts in the parallel production of multiple orders (such as the time allocation of injection molding machine A between orders X and Z), but also provides a real-time data feedback mechanism to address equipment anomalies, ensuring the flexibility and reliability of production scheduling.

[0175] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0176] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for automatically generating a production schedule based on the injection molding industry, characterized in that: The following steps are involved: Step S1: Obtain production order data, equipment status data, and material inventory data for the injection molding industry; Performing production parameter analysis on the production order data, equipment status data, and material inventory data to generate original production parameter data; deploying a production scheduling network based on the original production parameter data to obtain a distributed production scheduling network; Step S2: collecting and analyzing multi-level equipment load data on the distributed production scheduling network to generate equipment load impact factors; Step S3: Sense the state of production equipment according to the distributed production scheduling network, obtain equipment state data, and generate equipment state influencing factors based on the equipment state data; Step S4: collecting production order priority data according to the distributed production scheduling network to obtain order priority data; performing order data sampling and calibration on the order priority data to generate order calibration data; Divide the production task area based on the order calibration data to generate task area data; Step S5: performing production scheduling optimization calculation on the order calibration data and the equipment status data based on the task area data to generate a production scheduling optimization index; integrating the production scheduling optimization index, the equipment load influencing factor, and the equipment status influencing factor to generate comprehensive production scheduling data; Generating production scheduling instructions based on the comprehensive production scheduling data to generate production scheduling instruction data; Outputting instructions for the production scheduling instruction data to execute automatic production scheduling operations; Step S5 includes the following steps: Step S51: performing a production control chain analysis on the order calibration data and the equipment status data based on the task area data to generate control chain data; Step S52: performing production scheduling optimization calculation on the control chain data using a production scheduling optimization calculation formula to generate a production scheduling optimization index; integrating the production scheduling optimization index, equipment load influencing factors, and equipment status influencing factors to generate comprehensive production scheduling data; The production scheduling optimization calculation formula is as follows: in, Indicates the optimized value of the production schedule; Indicates the upper limit of the production scheduling cycle; It represents the production scheduling control coefficient; represents the system stability coefficient; Indicates the order priority factor; represents the deviation factor; Indicates the current production scheduling time point; Step S53: dividing the comprehensive production scheduling data into data sets to generate a model training set and a model test set; performing model training on the model training set using an optimization algorithm to generate a production scheduling instruction training model; and performing model optimization iteration on the production scheduling instruction training model using the model test set to generate a production scheduling instruction model; Step S54: importing the comprehensive production scheduling data into the production scheduling instruction model to generate production scheduling instructions and generate production scheduling instruction data; outputting instructions to the production scheduling instruction data to execute automatic production scheduling operations.

2. The method for automatically generating a production schedule based on injection molding industry production according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: Obtaining production order samples, equipment status samples, and material inventory samples for the injection molding industry; Step S12: performing parameter analysis on the production order sample, equipment status sample, and material inventory sample to obtain production order data, equipment status data, and material inventory data; Step S13: performing production parameter analysis on the production order data, equipment status data, and material inventory data to generate original production parameter data; Step S14: Deploy a production scheduling network based on the original production parameter data to obtain a distributed production scheduling network.

3. The method for automatically generating a production schedule based on injection molding industry production according to claim 2 is characterized in that: Step S14 includes the following steps: Step S141: performing production structure analysis on the original production parameter data to obtain production structure data; Step S142: dividing the production area according to the production structure data to generate area boundary data; Step S143: arranging production scheduling node positions based on the area boundary data to generate production scheduling node position data; Step S144: deploying distributed production scheduling nodes using the multi-dimensional production scheduling node position data, and calibrating parameters of the deployed distributed production scheduling nodes to generate distributed calibrated production scheduling nodes; Step S145: performing communication channel association on the distributed calibration scheduling nodes to generate a distributed scheduling network.

4. The method for automatically generating a production schedule based on injection molding industry production according to claim 1 is characterized in that: Step S2 includes the following steps: Step S21: collecting multi-level equipment load data on the distributed production scheduling network to obtain multi-level load monitoring data, wherein the multi-level load monitoring data includes primary load monitoring data, intermediate load monitoring data, and advanced load monitoring data; Step S22: performing multi-level load gradient calculation on the primary load monitoring data and the advanced load monitoring data to obtain a load gradient index; performing equipment load density difference impact analysis on the load gradient index based on the intermediate load monitoring data to generate load density difference data; Step S23: performing a load fluctuation analysis on the load gradient index based on the load density difference data to generate load fluctuation data; performing a load fluctuation trajectory analysis on the load fluctuation data to generate load fluctuation trajectory data; Step S24: analyzing the degree of equipment load variation on the load fluctuation data using the load fluctuation trajectory data to generate an equipment load impact factor.

5. The method for automatically generating a production schedule based on injection molding industry production according to claim 4 is characterized in that: Step S24 includes the following steps: Step S241: performing load fluctuation time series analysis on the equipment load fluctuation data using the equipment load fluctuation trajectory data to generate equipment load fluctuation time series data; performing dynamic graph conversion on the equipment load fluctuation time series data to generate an equipment load fluctuation dynamic graph; Step S242: performing equipment load abnormality event analysis on the equipment load fluctuation dynamic graph to generate equipment load abnormality event data; Step S243: performing a normal load change trend analysis on the equipment load fluctuation dynamic graph based on the equipment abnormal load event data to generate equipment normal load trend change data; Step S244: Detect the degree of impact of abnormal events on the normal load trend change data of the equipment using the abnormal load event data of the equipment to generate abnormal load impact degree data; perform equipment load change analysis on the equipment load fluctuation data based on the abnormal load impact degree data to generate an equipment load impact factor.

6. The method for automatically generating a production schedule based on injection molding industry production according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: Sense the state of production equipment according to the distributed production scheduling network and generate equipment sensing data; Step S32: collecting device parameters from the device sensing data, and performing device status identification on the collected device parameters to obtain device status data; Step S33: matching the device status data with a preset device database; if the device status data does not match the preset device database, marking the corresponding device status data as abnormal device status data; if the device status data does match the preset device database, marking the corresponding device status data as normal device status data; Step S34: performing an equipment operation evaluation on the abnormal equipment status data to generate abnormal equipment operation evaluation data; performing an equipment abnormality impact analysis on the normal equipment status data based on the abnormal equipment operation evaluation data to generate an equipment status impact factor.

7. The method for automatically generating a production schedule based on injection molding industry production according to claim 6 is characterized in that: Step S31 includes the following steps: Step S311: collecting production equipment sensor data from the distributed production scheduling network to obtain equipment sensor data; Step S312: performing data filtering on the device sensor data to generate device filtered data; performing signal enhancement on the device filtered data to generate device enhanced data; Step S313: performing signal feature analysis on the device enhancement data to generate device signal feature data; marking production area devices using the device signal feature data to generate area device marking data; Step S314: performing production equipment type identification on the regional equipment tag data to generate equipment type identification data; performing production equipment behavior perception based on the equipment type identification data to generate equipment perception data.

8. The method for automatically generating a production schedule based on injection molding industry production according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41: collecting production order priority data according to the distributed production scheduling network to obtain order priority data, wherein the order priority data includes order urgency data, order complexity data, and order delivery period data; Step S42: performing production order spatiotemporal distribution feature analysis on the order priority data to generate order spatiotemporal distribution feature data; performing order data sampling and calibration on the order spatiotemporal distribution feature data to generate order calibration data; Step S43: performing a production order status evaluation on the order calibration data to generate order status evaluation data; performing a distribution feature evaluation on the order calibration data to generate order distribution feature evaluation data; Step S44: Divide the production task areas according to the order status evaluation data and the order distribution characteristic evaluation data to generate task area data.

9. A system for automatically generating production schedules based on the injection molding industry, characterized in that: A method for automatically generating a production schedule based on injection molding industry production according to any one of claims 1 to 8, the system comprising: The production scheduling network deployment module is used to obtain production order data, equipment status data, and material inventory data for the injection molding industry; perform production parameter analysis on the production order data, equipment status data, and material inventory data to generate original production parameter data; and perform production scheduling network deployment based on the original production parameter data to obtain a distributed production scheduling network; An equipment load analysis module is configured to collect multi-level equipment load data from the distributed production scheduling network to obtain load monitoring data; perform multi-level load gradient calculations on the load monitoring data to obtain a load gradient index; perform load fluctuation analysis on the load gradient index to generate load fluctuation data; perform load fluctuation trajectory analysis on the load fluctuation data to generate load fluctuation trajectory data; and perform equipment load change degree analysis on the load fluctuation data using the load fluctuation trajectory data to generate an equipment load impact factor. An equipment status analysis module is configured to sense the status of production equipment based on the distributed production scheduling network and generate equipment perception data; collect equipment parameters from the equipment perception data and perform equipment status identification on the collected equipment parameters to obtain equipment status data; perform equipment abnormality impact analysis on the equipment status data to generate equipment status impact factors; A production task area analysis module is configured to collect production order priority data based on the distributed production scheduling network to obtain order priority data; perform order data sampling and calibration on the order priority data to generate order calibration data; and divide production task areas based on the order calibration data to generate task area data; A production scheduling instruction generation module is used to perform production scheduling optimization calculations on the order calibration data and equipment status data based on the task area data to generate a production scheduling optimization index; integrate the production scheduling optimization index, equipment load influencing factor and equipment status influencing factor to generate comprehensive production scheduling data; generate production scheduling instructions for the comprehensive production scheduling data to generate production scheduling instruction data; and output instructions for the production scheduling instruction data to execute automatic production scheduling operations.

Images