Production and logistics management method and system based on digital twinning

Abstract

The invention discloses a production and logistics management method and system based on digital twinning, and the method comprises the steps: obtaining a real-time data flow from a production equipment sensor and a logistics node, carrying out the cleaning and formatting of the real-time data flow through a streaming processing technology, and obtaining a structured real-time data set; aiming at the structured real-time data set, detecting equipment abnormity and order fluctuation by applying a time sequence analysis algorithm; if equipment abnormity or order fluctuation is detected, carrying out real-time simulation on production and logistics links through a pre-established digital twinborn model to obtain a dynamic change trend of event influence; and according to the dynamic change trend, a reinforcement learning algorithm is adopted to optimize and adjust the production scheduling and logistics path, and an optimized production and transportation plan is determined. According to the invention, intelligent cooperation of production and logistics links is realized, abnormal conditions can be quickly responded, scheduling decisions are dynamically optimized, and the overall operation efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of digital management, and particularly discloses a production and logistics management method and system based on digital twinning. BACKGROUND

[0002] In the field of modern industry and supply chain management, building an efficient production and logistics system has become a key pillar to enhance the competitiveness of enterprises. The importance of this field is self-evident, as it directly relates to resource utilization efficiency, cost control, and market response speed.

[0003] However, the current mainstream methods often expose the defects of being difficult to adapt to changing needs and lacking deep data integration when dealing with complex dynamic environments. These methods usually rely on the traditional separation of planning and execution mode, making it difficult to adjust in time when facing unexpected events or multi-link coordination, resulting in resource waste and low efficiency. By analyzing the challenges faced by this field, it can be found that the core problem lies in how to achieve real-time dynamic response and precise optimization of the whole process.

[0004] The first problem is that in the production and logistics process, various events such as equipment abnormalities or order fluctuations have high uncertainty, and the existing system is difficult to simulate and intervene effectively before or at the time of the event. This uncertainty further exacerbates another key problem, which is the fragmented state of data between different links, lacking full-chain connection from design to execution, leading to decisions often based on local information rather than a holistic perspective, thereby affecting the overall optimization effect.

[0005] Therefore, how to build a real-time simulation and full-chain data integration mechanism in a complex and changing industrial environment to achieve dynamic adjustment and intelligent decision-making of production and logistics processes has become a key problem to be solved. SUMMARY

[0006] The present application provides a production and logistics management method and system based on digital twinning, aiming to solve at least one of the defects in the prior art.

[0007] One aspect of the present application relates to a production and logistics management method based on digital twinning, comprising the following steps: Obtain real-time data streams from production equipment sensors and logistics nodes, and use stream processing technology to clean and format the real-time data streams to obtain structured real-time data sets; Apply time series analysis algorithms to the structured real-time data sets to detect equipment abnormalities and order fluctuations; If equipment abnormalities or order fluctuations are detected, a pre-established digital twinning model is used to simulate the production and logistics process in real time to obtain the dynamic trend of event impact; According to the dynamic trend, the production scheduling and logistics path are optimized and adjusted by using a reinforcement learning algorithm to determine the optimized production and transportation plan; From the optimized production and transportation plan, key decision parameters are extracted, and combined with the whole chain data integration platform to generate cross-link collaborative scheduling instructions; For collaborative scheduling instructions, the instructions are decomposed into specific tasks through a distributed computing framework and distributed to production equipment and logistics nodes to obtain state feedback after execution; According to the state feedback, an online learning algorithm is applied to update the parameters of the digital twin model to optimize the accuracy of the subsequent simulation intervention mechanism.

[0008] Further, the steps of obtaining real-time data streams from production equipment sensors and logistics nodes, and using stream processing technology to clean and format the real-time data streams to obtain structured real-time data sets include: From the sensor and logistics node, obtain the real-time data stream, and use the message queue tool to receive the time series data of the sensor and the tracking data of the logistics node to obtain the original data stream; For the original data stream, use the stream processing tool to clean it. If there are missing values or outliers in the original data stream, judge and fill in the mean value or delete them through the preset threshold to obtain the cleaned data stream; According to the cleaned data stream, use the data conversion tool to perform formatting operations to unify the cleaned data stream into a pre-established structured data model to obtain structured data; Through the structured data, use the database tool for storage and integration. If the structured data meets the preset time window requirements, trigger the real-time data set generation to obtain the real-time data set.

[0009] Further, for the structured real-time data set, the steps of applying a time series analysis algorithm to detect equipment abnormalities and order fluctuations include: From the structured real-time data set, obtain the equipment state and operating parameter information, and use the pre-established comparison rules to compare the operating parameters one by one. If the operating parameters exceed the preset threshold range, mark them as abnormal signals to obtain data records with abnormal markers; For data records with abnormal markers, obtain abnormal signal and fault type information, compare them with historical trend data to determine whether the abnormal signal conforms to the known fault type pattern, and determine the specific fault classification result; According to the fault classification result, combined with the order quantity and fluctuation amplitude information, use the time window division method to segment the data with abnormal markers. If the fluctuation amplitude exceeds the preset threshold range, mark it as an order exception to obtain an abnormal fluctuation identifier; For abnormal fluctuation identification and real-time monitoring needs, obtain alarm threshold and business impact information, and comprehensively evaluate the data with abnormal marks through preset logical rules to determine whether to trigger the alarm mechanism, and obtain the final alarm trigger record.

[0010] Further, if equipment abnormalities or order fluctuations are detected, the step of obtaining the dynamic change trend of the event impact by real-time simulation of the production and logistics links through the pre-established digital twin model includes: Obtain equipment status and operating parameters from real-time data sets, and obtain the dynamic change trend of equipment operation by real-time simulation of operating parameters through the pre-established digital twin model; For the dynamic change trend of equipment operation, obtain abnormal signals and historical trend data, compare the abnormal signals with the historical trend through logical rules to determine whether it conforms to the preset fault type mode, and determine the fault classification result; According to the fault classification result and order quantity data, the order quantity data is segmented by time window division method, and if the fluctuation amplitude exceeds the preset threshold range, it is marked as abnormal fluctuation to obtain the abnormal fluctuation identification; For abnormal fluctuation identification, obtain alarm threshold and business impact information, and comprehensively evaluate the abnormal fluctuation identification through logical rules to determine whether to trigger the alarm mechanism, and obtain the alarm trigger record.

[0011] Further, according to the dynamic change trend, the step of optimizing and adjusting the production scheduling and logistics path by using reinforcement learning algorithm includes: According to the dynamic change trend, obtain real-time updated data from order demand and inventory status, and compare resource allocation through the pre-established database, if the resource allocation cannot meet the order demand, trigger the adjustment process to determine the preliminary production scheduling scheme; For the preliminary production scheduling scheme, obtain the related information of logistics path and path cost, and use path planning tool to evaluate the first distribution efficiency, if the first distribution efficiency is lower than the preset threshold, re-plan the logistics path to obtain the adjusted transportation plan; Through the adjusted transportation plan, combine the first time window and demand prediction data, and use scheduling management tool to allocate time for production and distribution links, if the distribution task cannot be completed within the first time window, adjust the resource allocation again to determine the final production scheduling result; According to the final production scheduling result, obtain the change data of inventory status and order demand, and use data comparison tool to continuously monitor the first distribution efficiency and path cost to determine whether the optimized transportation plan meets the expected goal.

[0012] Further, the step of extracting key decision parameters from the optimized production and transportation plan, combining the full-chain data integration platform, and generating cross-linkage collaborative scheduling instructions includes: Obtaining real-time data of order demand and inventory status from the optimized production and transportation plan, comparing resource allocation through a pre-established database, if resource allocation cannot meet order demand, triggering data integration process to obtain preliminary collaborative scheduling parameters; According to the preliminary collaborative scheduling parameters, obtaining related data of the logistics path and the second time window, using path planning tools to evaluate the second distribution efficiency, if the second distribution efficiency is lower than the preset threshold, re-planning the logistics path to determine the adjusted collaborative scheduling instruction; Through the adjusted collaborative scheduling instruction, combining demand prediction and monitoring frequency data, using scheduling management tools to allocate time for production scheduling and distribution link, if the distribution task cannot be completed within the second time window, adjusting resource allocation again to determine the final collaborative scheduling scheme; According to the final collaborative scheduling scheme, obtaining dynamic data of inventory status and path cost, continuously monitoring the second distribution efficiency and collaborative scheduling effect through the data integration platform to obtain the optimized cross-linkage scheduling instruction.

[0013] Further, for the collaborative scheduling instruction, the step of decomposing the instruction into specific tasks through a distributed computing framework and assigning it to production equipment and logistics nodes to obtain state feedback after execution includes: According to the collaborative scheduling instruction, decompose the instruction through a distributed computing tool to generate multiple specific task units, preliminarily sort according to task priority and execution sequence, and obtain the decomposed task list; According to the task list, combining device load and node capacity data of production equipment and logistics nodes, using resource allocation tools for task allocation, if the device load exceeds the preset threshold, adjusting the task allocation order to determine the final task allocation scheme; According to the task allocation scheme, issuing specific tasks to production equipment and logistics nodes, collecting real-time state feedback data during execution, monitoring execution efficiency through feedback frequency, and obtaining dynamic updates of execution status; For dynamic updates of execution status, if the execution efficiency is lower than the preset threshold, optimizing task allocation through instruction adjustment tools, re-planning execution sequence combining device load and node capacity data, and determining the adjusted task execution arrangement.

[0014] Further, according to the state feedback, the step of updating the parameters of the digital twin model using online learning algorithm to optimize the accuracy of the subsequent simulation intervention mechanism includes: The execution data, which is the status feedback data, is obtained from the production node and the logistics node through data acquisition tools. The execution data is then structured and classified to obtain a list of classified execution data. A real-time update tool is used to dynamically monitor the execution data list. If the execution efficiency is lower than the preset threshold, the parameters of the simulation intervention mechanism are corrected by the parameter adjustment tool to determine the adjusted parameter configuration. By continuously tracking parameter configurations through feedback frequency monitoring tools, performance data under different business loads can be obtained. The data comparison tool is used to compare the performance data with the preset accuracy optimization target. If the preset threshold is not reached, the data backtracking tool is used to sort out the historical status feedback to obtain the updated data basis.

[0015] Furthermore, a real-time update tool is used to dynamically monitor the execution data list. If the execution efficiency is lower than a preset threshold, the parameters of the simulation intervention mechanism are corrected using a parameter adjustment tool. In the step of determining the adjusted parameter configuration, the system execution efficiency is calculated in real time using the following formula and compared with the preset threshold:

[0016] in, Indicates the system during a time period Internal execution efficiency This indicates the total number of sampling points within the monitoring time window. Indicates the first The output quantity of each sampling point Indicates the first Input resource quantity per sampling point Indicates the first The execution time interval for each sampling point; The adjusted parameter configuration is calculated using the following formula:

[0017] in, This indicates the adjusted new parameter configuration. This indicates the current parameter configuration. This represents the learning rate used to adjust the parameters. Describe the objective function For parameters gradient, This represents the preset efficiency threshold. This represents the actual efficiency value currently monitored.

[0018] Another aspect of the present application relates to a digital-twin-based production and logistics management system for implementing the above-mentioned digital-twin-based production and logistics management method, the digital-twin-based production and logistics management system comprising: a real-time data set acquisition module for acquiring real-time data streams from production equipment sensors and logistics nodes, cleaning and formatting the real-time data streams using stream processing technology to obtain structured real-time data sets; an anomaly detection module for detecting equipment anomalies and order fluctuations by applying time series analysis algorithms to the structured real-time data sets; a dynamic change trend acquisition module for, if an equipment anomaly or an order fluctuation is detected, simulating the production and logistics links in real time through a pre-established digital twin model to obtain a dynamic change trend of the event impact; a plan determination module for optimizing and adjusting the production scheduling and logistics path using a reinforcement learning algorithm according to the dynamic change trend to determine an optimized production and transportation plan; a collaborative scheduling instruction generation module for extracting key decision parameters from the optimized production and transportation plan and generating cross-link collaborative scheduling instructions in combination with a full-chain data integration platform; a state feedback acquisition module for, in response to the collaborative scheduling instructions, decomposing the instructions into specific tasks through a distributed computing framework and distributing them to production equipment and logistics nodes to obtain state feedback after execution; a precision optimization module for updating the parameters of the digital twin model using an online learning algorithm according to the state feedback to optimize the precision of the subsequent simulation intervention mechanism.

[0019] The present application has the following beneficial effects: The present application provides a digital-twin-based production and logistics management method and system, which detects anomalies through real-time data stream processing and time series analysis, simulates event impact using a digital twin model, optimizes production scheduling and logistics path using a reinforcement learning algorithm, generates cross-link collaborative scheduling instructions and decomposes them for execution, and finally updates model parameters using online learning. The present application realizes intelligent collaboration between production and logistics links, can quickly respond to abnormal situations, dynamically optimize scheduling decisions, and improve overall operational efficiency; through data-driven and artificial intelligence technology, a self-adaptive, closed-loop optimized production and logistics collaborative system is constructed, providing an effective solution for the intelligent transformation of manufacturing enterprises. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 Figure 1 is a flowchart of an embodiment of the digital-twin-based production and logistics management method of the present application; Figure 2 Figure 2 is a functional block diagram of an embodiment of the digital-twin-based production and logistics management system of the present application.

[0021] BRIEF DESCRIPTION OF DRAWINGS 10, real-time data set acquisition module; 20, anomaly detection module; 30, dynamic change trend acquisition module; 40, plan determination module; 50, collaborative scheduling instruction generation module; 60, state feedback acquisition module; 70, precision optimization module. DETAILED DESCRIPTION

[0022] In order to better understand the above technical solutions, the above technical solutions will be described in detail in combination with the drawings of the specification and specific embodiments.

[0023] As shown in Figure 1 The first embodiment of the present application proposes a production and logistics management method based on digital twinning, including the following steps: Step S100, real-time data stream is acquired from production equipment sensors and logistics nodes, and stream processing technology is used to clean and format the real-time data stream to obtain structured real-time data sets.

[0024] Through various sensors (such as temperature sensors, vibration sensors, pressure sensors, etc.) deployed on production equipment (such as machine tools, production lines, robots, etc.) and key nodes (such as warehouse entrances, transportation vehicle GPS, sorting equipment, etc.) in the logistics link, real-time data (such as device temperature changes, logistics position coordinates, material throughput, etc.) reflecting the running state of the equipment, the material flow situation, etc. are continuously collected. Subsequently, using stream processing technology (such as Apache Flink, Apache Kafka Streams, etc.), the continuously generated, unstructured or semi-structured real-time data stream is cleaned and formatted, and finally structured real-time data sets are formed.

[0025] Cleaning: removing noise (such as false positive abnormal values of sensors), repeated data (such as repeated reporting information of the same logistics node), missing values (such as data gaps caused by temporary sensor failure), etc. to ensure the accuracy and integrity of the data.

[0026] Formatting: converting the cleaned data into a unified structured format (such as JSON, CSV format conforming to a specific database table structure, or field form of a relational database, etc.), so that the data has a standard field definition, data type and logical association.

[0027] The structured real-time data set is a dynamic data set that can reflect the running state of the production equipment and the flow situation of the logistics nodes in real time, has a unified data format and clear data logic.

[0028] Step S200, for the structured real-time data set, a time series analysis algorithm is applied to detect equipment anomalies and order fluctuations.

[0029] Based on the constructed structured real-time data set (including device operation indicators, order flow data, and other structured information recorded continuously over time), by applying time series analysis algorithms, the time correlation patterns and trend characteristics implied in the data are mined, so as to identify device operation abnormal states and order volume abnormal fluctuations in real time.

[0030] Step S300, if the device abnormality or order fluctuation is detected, the production and logistics links are simulated in real time through the pre-established digital twin model, and the dynamic change trend of the event influence is obtained.

[0031] When the production device abnormal state (such as sudden temperature rise, abnormal vibration frequency, etc.) or order business fluctuation (such as sudden increase / decrease in order volume, sudden mutation in regional order distribution, etc.) is detected in real time through the time series analysis algorithm, based on the pre-constructed production and logistics link digital twin model, the key parameters of the abnormal event (such as device abnormality occurrence time, fault type, order fluctuation amplitude, etc.) are input into the model, and the real-time state of the production process and logistics scheduling after the event is dynamically simulated and deduced, and finally the influence range, degree and dynamic trend over time of the event on production efficiency, material transfer, order fulfillment, etc. are output.

[0032] Step S400, according to the dynamic change trend, the reinforcement learning algorithm is used to optimize and adjust the production scheduling and logistics path, and the optimized production and transportation plan is determined.

[0033] After obtaining the dynamic influence trend of the device abnormality or order fluctuation event (such as the process delay time of the production link, the change of the transport capacity gap of the logistics link, etc. over time), by deploying the reinforcement learning algorithm, the production scheduling scheme and logistics path planning are optimized and adjusted in real time and dynamically, with the goal of minimizing the negative influence of the event (such as production downtime, order fulfillment delay), and finally the optimal production plan and transportation scheme adapted to the current abnormal situation are generated.

[0034] Step S500, extract the key decision parameters from the optimized production and transportation plan, and generate cross-link collaborative scheduling instructions in combination with the whole-chain data integration platform.

[0035] On the basis of obtaining the optimized production and transportation plan (including production scheduling adjustment details, logistics path optimization scheme, etc.), core decision parameters affecting the collaborative operation of multiple links such as production and logistics are extracted, and multi-dimensional verification and correlation analysis are performed relying on the full-chain data integration platform (a unified data hub that integrates full-process data such as production, warehousing, transportation, and orders), to finally generate standardized scheduling instructions that can coordinate the efficient linkage of production, logistics, warehousing, and other cross-link processes, realizing the collaborative configuration of resources and seamless connection of business processes. In this embodiment, the core decision parameters are collaborative scheduling parameters, which are core decision parameters directly affecting the collaborative operation efficiency of multiple links such as production, logistics, and warehousing extracted from the optimized production plan and transportation plan. The collaborative scheduling parameters build the "dynamic adaptation rules" between production and logistics by clearly defining the target requirements, resource constraints, time nodes, and connection logic of each link, and are the key data support for eliminating cross-link resource conflicts and realizing efficient linkage of the full chain, and are also the core basis for generating cross-link collaborative scheduling instructions.

[0036] Step S600, for the collaborative scheduling instruction, the instruction is decomposed into specific tasks through a distributed computing framework, and is allocated to production equipment and logistics nodes to obtain the state feedback after execution.

[0037] With the cross-link collaborative scheduling instruction (standardized instruction coordinating multiple links such as production and logistics) as input, relying on a distributed computing framework (such as Apache Spark, Hadoop YARN, etc. computing architecture supporting multi-node parallel processing), the instruction is decomposed into specific sub-tasks that can be directly executed by production equipment and logistics nodes, and is allocated to the corresponding execution units (such as machine tools, transportation vehicles, warehouse equipment, etc.) through a task scheduling mechanism, and real-time collection of the execution progress, completion status, and other state feedback of each sub-task is performed to form a closed-loop monitoring of instruction execution.

[0038] Step S700, according to the state feedback, an online learning algorithm is applied to update the parameters of the digital twin model, and the accuracy of the subsequent simulation intervention mechanism is optimized.

[0039] Based on the real-time state feedback data (such as actual operation parameters of production equipment, actual time efficiency of order fulfillment, actual effect of intervention measures, etc.) after the collaborative scheduling instructions are executed based on physical entities such as production equipment and logistics nodes, the core parameters in the digital twin model are continuously corrected and optimized by applying an online learning algorithm (a machine learning method that can receive new data in real time and dynamically adjust the model), so that the mapping accuracy and simulation prediction ability of the digital twin model for the physical system are continuously improved, thereby improving the accuracy of subsequent simulation deduction and intervention decision-making for abnormal events. The digital twin model is a digital mirror of the physical entity, and the core parameters of the digital twin model are the key data that support the "dynamic mapping, accurate interaction, and continuous optimization" of the model and the physical entity, directly determining the restoration accuracy of the model for the state of the physical entity, the prediction ability for future behavior, and the effectiveness of entity intervention. The updated parameters of the digital twin model include the parameters of the simulated intervention mechanism, which is the core function of the digital twin model to realize "virtual-physical interaction and active regulation". The parameters of the simulated intervention mechanism are used to define the intervention logic, execution strategy, and effect evaluation standard of the digital twin model for the physical entity, ensuring the rationality, executability, and dynamic adaptation of the intervention instructions to the entity state.

[0040] Further, the embodiment provides a production and logistics management method based on digital twinning, and the step S100 comprises: In step S110, real-time data streams are obtained from sensors and logistics nodes. Time series data from sensors and tracking data from logistics nodes are received using message queue tools to obtain raw data streams.

[0041] In the real-time data processing scenario in the logistics field, data stream acquisition from sensors and logistics nodes is the basis of the entire process. Sensors may be deployed on transport vehicles to collect environmental data such as temperature and humidity in real time, while logistics nodes record scanning times and location information of packages. These data are received through message queue tools such as Kafka to form raw data streams. Kafka processes large-scale data with high throughput, ensuring real-time and reliable data transmission, thereby laying a foundation for subsequent processing.

[0042] In step S120, the raw data stream is cleaned using a stream processing tool. If there are missing values or outliers in the raw data stream, the pre-set threshold is used to judge and fill in the mean value or delete them to obtain the cleaned data stream.

[0043] For the cleaning link of the original data stream, a stream processing tool such as Apache Flink is used. Assuming that the temperature value in the sensor data is missing or abnormal, for example, the temperature value of a certain collection is -50 degrees, which obviously exceeds the preset threshold of the reasonable range of -20 to 50 degrees, the system will automatically judge and mark it as an abnormal value, and then fill it with the average value of the past hour, such as 25 degrees, or directly delete the record. This cleaning method can effectively improve the data quality, avoid the interference of abnormal values on subsequent analysis, and ensure the accuracy of the data stream.

[0044] Step S130, according to the cleaned data stream, a data conversion tool is used to perform formatting operation, and the cleaned data stream is unified to a pre-established structured data model to obtain structured data.

[0045] The cleaned data stream needs to be formatted by a data conversion tool and unified to a structured data model. Assuming that the sensor data format in the original data stream is JSON and the logistics node data is CSV, the system will use tools such as Apache NiFi to convert it to a unified structured format, such as a table model containing timestamp, device ID, temperature value, location, etc. This conversion not only facilitates data storage and query, but also improves data consistency and provides convenience for subsequent analysis.

[0046] Step S140, through the structured data, a database tool is used for storage and integration, if the structured data meets the preset time window requirement, the real-time data set generation is triggered, and the real-time data set is obtained.

[0047] Structured data storage and integration can be achieved through a database tool such as PostgreSQL. The data is stored by timestamp partitioning, and if it meets the preset time window requirement, such as generating a data set every 5 minutes, the real-time data set generation is triggered. This mechanism can ensure efficient management and timely update of data to meet real-time monitoring requirements. The generated real-time data set is used to monitor abnormal situations in the logistics process, such as temperature exceeding the standard or package delay, thereby improving logistics efficiency and customer satisfaction.

[0048] Further, the production and logistics management method based on digital twinning provided by the embodiment comprises the following steps: Step S210, obtaining device state and running parameter information from the structured real-time data set, and comparing the running parameters item by item using pre-established comparison rules, if the running parameters exceed the preset threshold range, marking it as an abnormal signal, and obtaining data records with abnormal marks.

[0049] In real-time data processing scenarios in the logistics field, obtaining device status and operating parameter information from structured real-time data sets is a core step in anomaly detection. Structured data sets typically contain sensor data for transportation vehicles, such as engine speed, oil temperature, tire pressure, and other operating parameters, as well as device status, such as whether it is operating normally. Assuming that the real-time data set for a certain transportation vehicle shows that the engine speed has reached 5000 revolutions per minute, exceeding the preset threshold of 4000 revolutions per minute, the system will mark this data as an abnormal signal based on pre-established comparison rules.

[0050] Step S220, for data records with abnormal labels, obtain abnormal signal and fault type information, compare with historical trend data to determine whether the abnormal signal conforms to the known fault type pattern, and determine the specific fault classification result.

[0051] By using a streaming processing tool such as Apache Spark Streaming to compare data item by item, and combining threshold ranges (such as 3000-4000 revolutions per minute for normal speed), the abnormality can be determined. This method can quickly identify potential problems and ensure the accuracy of subsequent analysis.

[0052] Step S230, based on the fault classification result, combined with order quantity and fluctuation amplitude information, use time window division to segment the data with abnormal labels, if the fluctuation amplitude exceeds the preset threshold range, mark it as an order anomaly, and obtain an abnormal fluctuation identifier.

[0053] For data records with abnormal labels, the system will further analyze the matching of abnormal signals and historical trend data to determine the fault type. For example, historical trend data shows that when the engine speed continuously exceeds 4000 revolutions per minute and the oil temperature is higher than 90 degrees Celsius, it corresponds to the "engine overheating" fault pattern. Assuming that the current abnormal signal shows a speed of 4500 revolutions per minute and an oil temperature of 95 degrees Celsius, the system confirms that it conforms to the "engine overheating" pattern by comparing historical data, and generates a fault classification result. This comparison based on historical trends can effectively distinguish between occasional anomalies and systematic faults, improving the accuracy of fault diagnosis.

[0054] Step S240, for abnormal fluctuation identifiers and real-time monitoring requirements, obtain alarm threshold and business impact information, and comprehensively evaluate data with abnormal labels based on preset logical rules to determine whether to trigger the alarm mechanism, and obtain the final alarm trigger record.

[0055] In combination with order quantity and fluctuation range information, the system will segment the abnormal data. Assuming that a logistics node records an order quantity of 100 orders in a 5-minute time window, which suddenly increases to 200 orders, the fluctuation range exceeds the preset threshold of 50%. Through time window division, the system marks this segment of data as an order anomaly. In specific implementation, data is stored by time window using a database such as ClickHouse, and the fluctuation range is calculated through SQL query. This segmentation processing method can accurately capture the time period of order anomaly, facilitating subsequent analysis and management.

[0056] For abnormal fluctuation identification and real-time monitoring needs, the system will comprehensively evaluate whether to trigger the alarm mechanism. For example, order anomaly combined with business impact information (such as abnormal orders involving high-value goods), the system determines whether to immediately trigger an alarm through preset logic rules.

[0057] Assuming that an abnormal fluctuation identification shows an order surge involving cold chain goods, the system generates an alarm trigger record in combination with an alarm threshold (such as cold chain goods anomalies requiring notification within 2 minutes), notifying dispatch personnel to adjust transportation plans. This comprehensive evaluation mechanism can quickly respond to business risks, improving the timeliness and reliability of logistics management.

[0058] The above steps form a complete abnormal detection and alarm system through stream processing, historical trend comparison, segmentation processing, and comprehensive evaluation. Each step closely connects from device state monitoring to fault classification, to order anomaly detection and alarm triggering, collectively ensuring the stability of the logistics process. This approach not only quickly identifies problems, but also optimizes resource scheduling through precise fault classification and alarm mechanisms, significantly improving logistics efficiency.

[0059] Further, the production and logistics management method based on digital twinning provided by the embodiment comprises the following steps: Step S310, obtaining device status and running parameters from real-time data sets, and simulating running parameters in real time through a pre-established digital twinning model to obtain dynamic change trends of device running.

[0060] In the real-time data processing scenario in the logistics field, potential problems are effectively identified by monitoring the device status and running parameters of the transportation vehicle. For real-time data set processing, assuming that the sensor data of a transportation vehicle includes speed, fuel consumption, and load, etc., through a pre-constructed digital twinning model, the running state of the vehicle under different working conditions is simulated. For example, the model simulation shows that the fuel consumption of the vehicle should be 15 liters per 100 kilometers when the load is 5 tons and the speed is 80 kilometers per hour, and if the real-time data deviates from this value significantly, it may indicate an anomaly. This simulation method can dynamically reflect the running trend of the device, providing a basis for subsequent analysis.

[0061] Step S320, for the dynamic change trend of the device operation, the abnormal signal and the historical trend data are obtained, and the abnormal signal is compared with the historical trend data through logical rules to determine whether it conforms to the preset fault type mode, and the fault classification result is determined.

[0062] For the dynamic change trend of the device operation, the system will extract the abnormal signal and compare it with the historical trend data. Assuming that the speed of a vehicle suddenly drops from 60 kilometers per hour to 20 kilometers per hour in a short time, and historical data shows that similar situations are often accompanied by brake system problems, the system determines that it may conform to the "brake abnormality" fault mode through logical rules, and then determines the fault classification result. This comparison method can quickly locate the problem type and provide accurate guidance for subsequent processing.

[0063] Step S330, according to the fault classification result and the order quantity data, the order quantity data is segmented and processed by time window division method, if the fluctuation amplitude exceeds the preset threshold range, it is marked as abnormal fluctuation, and the abnormal fluctuation identifier is obtained.

[0064] In the processing of order quantity data, the system uses time window division method to segment and analyze the data. Assuming that the order quantity of a logistics node increases from 50 to 120 in 10 minutes, the fluctuation amplitude far exceeds the preset threshold of 30%, the system will mark this period as abnormal fluctuation. This segmented processing can clearly capture the key time points of data change, which is convenient for further analysis of abnormal reasons. The size of the time window can be adjusted according to business needs, such as shortening to 5 minutes during peak hours to improve sensitivity.

[0065] Step S340, for the abnormal fluctuation identifier, the alarm threshold and business impact information are obtained, the abnormal fluctuation identifier is comprehensively evaluated through logical rules, it is judged whether the alarm mechanism is triggered, and the alarm trigger record is obtained.

[0066] For the abnormal fluctuation identifier, the system will comprehensively evaluate the alarm threshold and business impact information. Assuming that an abnormal fluctuation involves the transportation of special goods, and the alarm threshold stipulates that such situations need to be responded within 3 minutes, the system determines that the alarm mechanism needs to be triggered through logical rules, generates the corresponding alarm trigger record, and notifies the relevant personnel to take measures. This evaluation method can ensure that important abnormalities are handled in a timely manner and reduce business risks. The weight of business impact information can be dynamically adjusted according to the value of goods or the urgency of transportation to optimize the priority of alarm.

[0067] The above steps form a complete monitoring system through real-time simulation, historical comparison, segmented analysis and comprehensive evaluation. Each step supports each other, from the dynamic change of the device state to the fault classification, to the labeling and alarm triggering of order abnormalities, to jointly ensure the stability of logistics transportation. This method can quickly respond to potential problems and improve management efficiency.

[0068] Further, the embodiment provides a production and logistics management method based on digital twinning, step S400 includes: Step S410, according to the dynamic change trend, real-time updated data is obtained from order demand and inventory state, and the resource allocation situation is compared through the pre-established database, if the resource allocation cannot meet the order demand, the adjustment process is triggered, and a preliminary production scheduling scheme is determined.

[0069] For the monitoring of dynamic change trend, starting from the real-time data of order demand and inventory state, whether the resources can meet the business demand is analyzed. Assuming that a logistics center processes 2000 orders per day, and the current inventory shows that only 1500 orders can be supported by the available distribution vehicles, through the pre-established database comparison, it is found that there is a gap in resource allocation, the system will automatically trigger the adjustment process, and a preliminary production scheduling scheme of increasing temporary vehicles is generated. This way can timely find the problem of resource shortage and provide a basis for subsequent optimization.

[0070] Step S420, for the preliminary production scheduling scheme, related information of logistics path and path cost is obtained, the first distribution efficiency is evaluated by using path planning tool, if the first distribution efficiency is lower than the preset threshold, the logistics path is re-planned, and the adjusted transportation plan is obtained.

[0071] For the preliminary production scheduling scheme, the system will evaluate in combination with the logistics path and path cost information. Assuming that a distribution path is 100 kilometers long, and the cost is 2 yuan per kilometer, if the distribution efficiency is found to be lower than the preset threshold after evaluation by using the path planning tool, for example, only 20 kilometers of distribution task can be completed per hour on average, the system will re-plan the path to 80 kilometers, and the adjusted transportation plan will reduce the cost to 1.8 yuan per kilometer. This re-planning can effectively improve the distribution efficiency and reduce the operating cost.

[0072] Step S430, through the adjusted transportation plan, in combination with the first time window and demand prediction data, the production and distribution links are time allocated by using the scheduling management tool, if the distribution task cannot be completed within the first time window, the resource allocation is adjusted again, and the final production scheduling result is determined.

[0073] In the adjusted transportation plan, the system will allocate time in combination with the time window and demand prediction data. Assuming that a batch of goods needs to be delivered within 2 hours, and the prediction shows that the road congestion during peak period may cause delay, the scheduling management tool will arrange the delivery time 30 minutes in advance, if it still cannot meet the time window requirement, the resources are adjusted again, for example, a standby vehicle is added, and the final production scheduling result is determined. This time allocation method can guarantee the punctuality of the distribution task and improve the customer satisfaction.

[0074] Step S440, according to the final production scheduling result, obtain the change data of inventory status and order demand, continuously monitor the first distribution efficiency and path cost through the data comparison tool, and determine whether the optimized transportation plan meets the expected target.

[0075] For the final production scheduling result, the system will continuously monitor the inventory status and order demand changes. Suppose the order volume suddenly increases to 2500 orders one day, and the inventory vehicles are still allocated according to the original plan. Through the data comparison tool, it is found that the distribution efficiency decreases by 10% and the path cost increases by 15%. The system will timely adjust the transportation plan to ensure that the optimization result meets the expected target. This continuous monitoring mechanism can dynamically adapt to business fluctuations and ensure the stability of the logistics system.

[0076] Further, the production and logistics management method based on digital twinning provided by the embodiment comprises the following steps: Step S510, obtaining real-time data of order demand and inventory status from the optimized production and transportation plan, comparing resource allocation through the pre-established database, if the resource allocation cannot meet the order demand, triggering the data integration process to obtain the preliminary collaborative scheduling parameters.

[0077] In the resource allocation and scheduling optimization scenario in the logistics field, real-time data acquisition of order demand and inventory status is the basis for optimizing production and transportation plan. Suppose the daily order volume of a logistics center is 1800 orders, and the inventory shows that only 1400 orders can be supported by the available vehicles. Through the pre-established database comparison, it is found that the resource gap is obvious. At this time, the system will trigger the data integration process to generate preliminary collaborative scheduling parameters, which involve vehicle allocation and personnel arrangement information, so as to quickly identify the problem of resource shortage and provide data support for subsequent adjustment. The collaborative scheduling parameters refer to the core decision parameters extracted from the optimized production plan and transportation plan, which have a decisive influence on the cross-link collaborative operation of production and logistics. The collaborative scheduling parameters can clearly define the target requirements, resource constraints, time nodes and connection logic of production, transportation, warehousing and other links, which are the key data support for realizing efficient linkage of each link, avoiding resource conflicts and ensuring smooth operation of the whole chain. It is also the core basis for generating cross-link collaborative scheduling instructions.

[0078] Step S520, according to the preliminary collaborative scheduling parameters, obtaining related data of logistics path and second time window, evaluating the second distribution efficiency by using path planning tool, if the second distribution efficiency is lower than the preset threshold, re-planning the logistics path to determine the adjusted collaborative scheduling instruction.

[0079] For the preliminary collaborative scheduling parameters, the system will combine the logistics path and time window data to evaluate the distribution efficiency using path planning tools. Assuming that the total length of a path is 120 kilometers and the preset distribution efficiency threshold is 25 kilometers per hour, and the actual evaluation finds that it can only reach 18 kilometers per hour, the system will re-plan the path to shorten the total length to 90 kilometers and determine the adjusted collaborative scheduling instructions. This re-planning can effectively improve the distribution speed and reduce time waste.

[0080] Step S530, through the adjusted collaborative scheduling instructions, combined with demand prediction and monitoring frequency data, using scheduling management tools to allocate time for production scheduling and distribution links, if the distribution task cannot be completed within the second time window, the resource allocation is adjusted again, and the final collaborative scheduling scheme is determined.

[0081] Based on the adjusted collaborative scheduling instructions, the system will combine demand prediction and monitoring frequency data to allocate time using scheduling management tools. Assuming that a batch of goods needs to be delivered within 3 hours, and the forecast shows that the peak period may be delayed by 1 hour, the system will arrange the delivery task 1 hour in advance. If it still cannot meet the time window requirement, secondary adjustment of resources will be made, such as temporarily increasing 2 delivery vehicles, and finally forming a collaborative scheduling scheme. This time allocation method can ensure the timely completion of tasks and improve service reliability.

[0082] Step S540, according to the final collaborative scheduling scheme, obtain the dynamic data of inventory status and path cost, through the data integration platform to continuously monitor the second distribution efficiency and collaborative scheduling effect, get the optimized cross-link scheduling instructions.

[0083] For the final collaborative scheduling scheme, the system continuously monitors the dynamic changes of inventory status and path cost through the data integration platform. Assuming that the order volume suddenly increases to 2200 orders, the path cost increases by 12%, and the distribution efficiency decreases by 8%, the system will adjust the plan in time to optimize the cross-link scheduling instructions to maximize resource utilization. This continuous monitoring mechanism can dynamically adapt to business fluctuations and maintain the stability of the logistics system.

[0084] Further, the production and logistics management method based on digital twinning provided by the embodiment, step S600 includes: Step S610, according to the collaborative scheduling instructions, decompose the instructions through the distributed computing tool to generate multiple specific task units, preliminarily sort the task priority and execution sequence, and obtain the decomposed task list.

[0085] In the collaborative scheduling and task allocation scenarios of the logistics field, the application of distributed computing tools can break down complex instructions into multiple executable task units. Suppose a logistics center receives a delivery instruction covering the entire region, involving the delivery of 2000 orders. The system first uses distributed computing tools to decompose the instruction into 10 task units divided by region, with each unit responsible for approximately 200 delivery tasks. During initial sorting, the system generates a task list based on task priority (e.g., prioritizing urgent orders) and execution sequence (e.g., prioritizing deliveries to closer areas). This decomposition method refines large tasks into actionable units, facilitating subsequent allocation and management.

[0086] Step S620: Based on the task list, and combined with the equipment load and node capacity data of production equipment and logistics nodes, tasks are allocated using a resource allocation tool. If the equipment load exceeds the preset threshold, the task allocation order is adjusted to determine the final task assignment scheme.

[0087] For resource allocation of the task list, the system combines the load and capacity data of production equipment and logistics nodes to assign tasks. Suppose a logistics node has 5 delivery vehicles, each with a maximum daily load of 400 orders, while the task list allocates 2200 delivery orders to this node, significantly exceeding the load threshold. The resource allocation tool will adjust the task allocation order, transferring the excess 200 orders to neighboring nodes, while considering that neighboring nodes have a vehicle capacity of 800 orders, ensuring task balance. This adjustment avoids equipment overload and ensures a smooth delivery process.

[0088] Step S630: According to the task assignment plan, specific tasks are issued to production equipment and logistics nodes, and status feedback data during the execution process is collected in real time. The execution efficiency is monitored by the feedback frequency to obtain dynamic updates of the execution status.

[0089] The following formula is used to assign specific tasks to production equipment and logistics nodes: (1) In formula (1), Indicates to the first The time for assigning tasks to each production equipment or logistics node. This indicates the total number of tasks. Indicates the first The workload of each task Indicates the first The priority weight of each task. Indicates the first The processing capacity of each node Indicates the first The availability coefficient of each node.

[0090] The monitoring of the execution efficiency by the feedback frequency is realized by the following formula: (2) In formula (2), represents the execution efficiency monitoring index, represents the actual feedback frequency, represents the expected feedback frequency, represents the variance of the feedback frequency, represents the square of the mean of the feedback frequency.

[0091] After the task allocation scheme is issued, the system will collect the state feedback data in the execution process in real time, and monitor the execution efficiency by the feedback frequency. Assuming that the feedback frequency of a certain regional distribution task is set to update once every 30 minutes, the system finds that the distribution efficiency of a certain vehicle is lower than the preset 20 orders per hour, and only reaches 15 orders per hour. Dynamically update the display that the vehicle may be delayed due to road conditions, and the system will record this state and provide a basis for subsequent adjustment. This real-time monitoring mechanism helps to discover problems in time and ensures the controllability of task execution.

[0092] Step S640, for the dynamic update of the execution state, if the execution efficiency is lower than the preset threshold, the task allocation is optimized by the instruction adjustment tool, the execution time sequence is re-planned combined with the device load and node capacity data, and the adjusted task execution arrangement is judged.

[0093] The adjusted task execution arrangement is obtained by the following formula: (3) In formula (3), represents the optimized task execution time sequence arrangement, represents the set of all possible time sequence schemes, represents the total number of tasks, represents the priority weight of task , represents the execution time of task , represents the delay penalty coefficient, represents the actual completion time of task , represents the deadline of task .

[0094] When execution efficiency falls below a preset threshold, the instruction adjustment tool intervenes to optimize task allocation. For example, if a node experiences reduced delivery efficiency due to peak-hour traffic congestion, the system, combining equipment load and node capacity data, reschedules the execution timeline, shifting some tasks to nighttime deliveries and adding a backup vehicle. The adjusted schedule shows that the completion rate of nighttime delivery tasks has improved to the preset standard. This optimization method can flexibly respond to unforeseen circumstances, ensuring the smooth progress of the overall delivery mission.

[0095] Furthermore, in the production and logistics management method based on digital twins provided in this embodiment, step S700 includes: Step S710: Obtain execution data of status feedback from production nodes and logistics nodes through data acquisition tools, perform structured classification on the execution data, and obtain a list of classified execution data.

[0096] The application of data acquisition tools obtains real-time execution data from production and logistics nodes, providing a foundation for subsequent analysis. This execution data is then structured and categorized into groups such as delivery progress, vehicle load, and node capacity, forming a clear list of execution data. This classification method facilitates rapid identification of problem areas, laying the foundation for dynamic monitoring.

[0097] Step S720: Use a real-time update tool to dynamically monitor the execution data list. If the execution efficiency is lower than the preset threshold, use a parameter adjustment tool to correct the parameters of the simulation intervention mechanism and determine the adjusted parameter configuration.

[0098] The system execution efficiency is calculated in real time using the following formula and compared with a preset threshold: (4) In formula (4), Indicates the system during a time period Internal execution efficiency This indicates the total number of sampling points within the monitoring time window. Indicates the first The output quantity of each sampling point Indicates the first Input resource quantity per sampling point Indicates the first The execution time interval for each sampling point.

[0099] The adjusted parameter configuration is calculated using the following formula: (5) In formula (5), This indicates the adjusted new parameter configuration. This indicates the current parameter configuration. a learning rate representing a parameter adjustment, a target function a gradient of a parameter a preset efficiency threshold value, a preset efficiency threshold value, a preset efficiency threshold value,

[0100] For dynamic monitoring of the execution data list, the real-time updating tool refreshes the data every certain time to ensure that the manager is aware of the latest situation. Assuming that a logistics center is responsible for the delivery of 1000 goods, the system monitoring finds that the delivery efficiency of a certain area is 10 per hour, which is lower than the preset threshold of 15 per hour. At this time, the parameter adjustment tool intervenes to correct the parameters of the simulated intervention mechanism, such as adjusting the delivery route priority to avoid peak hours or increasing temporary vehicle support. The adjusted parameter configuration shows that the efficiency of the area has increased to 14 per hour, close to the expected target. The corrected parameters of the simulated intervention mechanism are dynamic adjustment parameters set to solve problems such as "intervention effect not up to expectations" and "model and entity adaptation deviation". The core role is to enable the intervention mechanism to have the ability of "self-calibration" by continuously correcting the intervention logic, strategy or threshold to narrow the gap between simulated intervention and physical entity actual response. The corrected parameters of the simulated intervention mechanism need to be designed specifically in combination with the reasons for deviation (such as model prediction error, actuator lag, environmental disturbance), including deviation diagnosis parameters, adjustment amplitude parameters, iterative correction parameters, constraint adaptation parameters and effect verification parameters. When the intervention effect is not good, first locate the problem through the deviation diagnosis parameter, then quantify the correction strength with the adjustment amplitude parameter, set the adjustment rhythm according to the iterative correction parameter, execute the correction under the restriction of the constraint adaptation parameter, and finally judge whether the correction is effective through the effect verification parameter. This process enables the intervention mechanism of the digital twin model to dynamically adapt to the aging, environmental changes or sudden disturbances of the physical entity, continuously improving the accuracy of "virtual-physical interaction".

[0101] Step S730, continuously track the parameter configuration through the feedback frequency monitoring tool to obtain performance data under different business loads.

[0102] With the help of special monitoring tools (such as real-time data acquisition systems, load pressure sensors, parameter log analysis platforms, etc.), the core parameter configurations in the digital twin model, production scheduling system or other complex collaborative systems are continuously tracked and recorded, and then the actual running performance data (such as response speed, error rate, stability, resource utilization rate, etc.) of these parameters under different business load conditions (such as low load, medium load, high load, sudden peak load, etc.) are extracted.

[0103] For the continuous tracking of the feedback frequency monitoring tool, the system sets the performance data of the parameter configuration to be updated every 20 minutes to observe its adaptability under different business loads. During the peak of business, the load of a certain logistics node reaches 2000 orders per day, far exceeding the processing capacity of 1500 orders. Through the feedback frequency monitoring tool, it is found that the adjusted parameter configuration can still maintain the delivery efficiency under high load, indicating that the parameter optimization has strong adaptability. This continuous tracking helps to verify the adjustment effect and provides data support for subsequent optimization.

[0104] Step S740, the performance data is compared with the preset precision optimization target by using the data comparison tool. If the preset threshold is not reached, the historical state feedback is arranged by executing the data backtracking tool to obtain the updated data basis.

[0105] The data basis is updated by integrating the historical state feedback through the data backtracking tool. The updated data basis is as follows: (6) In formula (6), represents the updated data basis state, represents the current state retention weight, represents the current data state, represents the total number of historical states, represents the weight of the th historical state, represents the th historical state feedback data.

[0106] For the comparison of performance data and preset precision optimization target, the data comparison tool analyzes whether the adjusted efficiency meets the expectation. If the delivery completion rate target of a certain node is 95%, but the actual value is only 88%, the preset threshold is not reached. At this time, the data backtracking tool is executed to arrange the historical state feedback of the past 7 days to analyze the reasons for low efficiency, for example, it is found that a certain vehicle is frequently affected by traffic delays. Based on this, the system updates the data basis to provide reference for further optimization.

[0107] See Figure 2The embodiment provides a production and logistics management system based on digital twinning, which is used for executing the production and logistics management method based on digital twinning, and comprises a real-time data set acquisition module 10, an anomaly detection module 20, a dynamic change trend acquisition module 30, a plan determination module 40, a collaborative scheduling instruction generation module 50, a state feedback acquisition module 60 and a precision optimization module 70. The real-time data set acquisition module 10 is used for acquiring real-time data streams from production equipment sensors and logistics nodes, and adopts a stream processing technology to clean and format the real-time data streams, so as to obtain a structured real-time data set. The anomaly detection module 20 is used for detecting equipment anomalies and order fluctuations by applying a time series analysis algorithm to the structured real-time data set. The dynamic change trend acquisition module 30 is used for, if the equipment anomalies or the order fluctuations are detected, simulating the production and logistics links in real time by using a pre-established digital twinning model, so as to obtain a dynamic change trend of event influence. The plan determination module 40 is used for, according to the dynamic change trend, optimizing and adjusting the production scheduling and the logistics path by using a reinforcement learning algorithm, and determining an optimized production and transportation plan. The collaborative scheduling instruction generation module 50 is used for extracting key decision parameters from the optimized production and transportation plan, combining a whole-chain data integration platform, and generating a cross-link collaborative scheduling instruction. The state feedback acquisition module 60 is used for, according to the collaborative scheduling instruction, decomposing the instruction into specific tasks by using a distributed computing framework, and distributing the specific tasks to the production equipment and the logistics nodes, so as to obtain state feedback after execution. The precision optimization module 70 is used for, according to the state feedback, updating parameters of the digital twinning model by using an online learning algorithm, and optimizing the precision of a subsequent simulation intervention mechanism.

[0108] Compared with the prior art, the production and logistics management method and system based on digital twinning provided by the embodiment detect anomalies through real-time data stream processing and time series analysis, simulate event influence by using a digital twinning model, optimize production scheduling and logistics paths by using a reinforcement learning algorithm, generate cross-link collaborative scheduling instructions and decompose execution, and finally update model parameters by using online learning. The embodiment realizes intelligent collaboration of production and logistics links, can quickly respond to abnormal situations, dynamically optimizes scheduling decisions, and improves overall operation efficiency. Through data driving and artificial intelligence technology, the embodiment constructs a self-adaptive and closed-loop optimized production and logistics collaborative system, and provides an effective solution for intelligent transformation of manufacturing enterprises.

[0109] While the preferred embodiments of the application have been described, additional variations and modifications can be made to these embodiments by those skilled in the art once they have the benefit of the present disclosure without departing from the spirit and scope of the application. Accordingly, it is intended that the appended claims include all such variations and modifications as fall within the scope of the present application. It is apparent that those skilled in the art can modify and adapt the present application in various ways without departing from the spirit and scope of the present application. It is therefore intended that the present application encompass all such modifications and variations as fall within the scope of the claims and their equivalents.

Claims

1. A production and logistics management method based on digital twins, characterized in that, Includes the following steps: Real-time data streams are acquired from production equipment sensors and logistics nodes, and the real-time data streams are cleaned and formatted using streaming processing technology to obtain a structured real-time dataset. For structured real-time datasets, time series analysis algorithms are applied to detect equipment anomalies and order fluctuations; If an abnormality in the equipment or a fluctuation in orders is detected, the production and logistics processes are simulated in real time using a pre-established digital twin model to obtain the dynamic trend of the event's impact. Based on the dynamic change trend, reinforcement learning algorithms are used to optimize and adjust production scheduling and logistics routes to determine the optimized production and transportation plan; Key decision parameters are extracted from the optimized production and transportation plans, and combined with the full-chain data integration platform to generate cross-link collaborative scheduling instructions; For the aforementioned collaborative scheduling instruction, the instruction is decomposed into specific tasks through a distributed computing framework, allocated to production equipment and logistics nodes, and status feedback is obtained after execution. Based on the status feedback, the parameters of the digital twin model are updated using an online learning algorithm to optimize the accuracy of subsequent simulation intervention mechanisms.

2. The production and logistics management method based on digital twins as described in claim 1, characterized in that, The steps of acquiring real-time data streams from production equipment sensors and logistics nodes, and cleaning and formatting the real-time data streams using streaming processing technology to obtain structured real-time datasets include: Real-time data streams are acquired from sensors and logistics nodes. A message queue tool is used to receive the time-series data from the sensors and the tracking data from the logistics nodes to obtain the raw data stream. For the original data stream, a streaming processing tool is used to clean it. If there are missing or outlier values ​​in the original data stream, a preset threshold is used to determine whether to fill in the mean or delete them, so as to obtain a cleaned data stream. Based on the cleaned data stream, a data transformation tool is used to perform a formatting operation, unifying the cleaned data stream into a pre-established structured data model to obtain structured data; The structured data is stored and integrated using database tools. If the structured data meets the preset time window requirements, a real-time dataset is generated, thus obtaining a real-time dataset.

3. The production and logistics management method based on digital twins as described in claim 1, characterized in that, The steps for detecting equipment anomalies and order fluctuations using time series analysis algorithms on structured real-time datasets include: Device status and operating parameter information are obtained from structured real-time datasets. The operating parameters are compared item by item using pre-established comparison rules. If the operating parameter exceeds the preset threshold range, it is marked as an abnormal signal, and data records with abnormal markings are obtained. For data records marked with anomalies, acquire the anomaly signal and fault type information, compare it with the historical trend data, determine whether the anomaly signal conforms to the known fault type pattern, and determine the specific fault classification result. Based on the fault classification results, combined with the order quantity and fluctuation range information, the data with anomaly markers are segmented using a time window division method. If the fluctuation range exceeds the preset threshold range, it is marked as an order anomaly, thus obtaining an abnormal fluctuation identifier. In response to the abnormal fluctuation indicators and real-time monitoring requirements, alarm thresholds and business impact information are obtained. Data with abnormal markers is comprehensively evaluated through preset logical rules to determine whether an alarm mechanism is triggered, and the final alarm trigger record is obtained.

4. The production and logistics management method based on digital twins as described in claim 1, characterized in that, If an equipment malfunction or order fluctuation is detected, the steps to simulate the production and logistics processes in real time using a pre-established digital twin model to obtain the dynamic trend of the event's impact include: The device status and operating parameters are obtained from real-time datasets, and the operating parameters are simulated in real time using a pre-established digital twin model to obtain the dynamic trend of device operation. Based on the dynamic changes in equipment operation, abnormal signals and historical trend data are acquired. The abnormal signals and historical trends are compared using logical rules to determine whether they conform to a preset fault type pattern and to determine the fault classification result. Based on the fault classification results and order quantity data, the order quantity data is segmented using a time window division method. If the fluctuation range exceeds the preset threshold range, it is marked as abnormal fluctuation, and an abnormal fluctuation identifier is obtained. For the abnormal fluctuation identifier, obtain the alarm threshold and business impact information, comprehensively evaluate the abnormal fluctuation identifier through logical rules, determine whether to trigger the alarm mechanism, and obtain the alarm trigger record.

5. The production and logistics management method based on digital twins as described in claim 1, characterized in that, Based on the aforementioned dynamic trends, the steps for optimizing and adjusting production scheduling and logistics routes using reinforcement learning algorithms to determine the optimized production and transportation plan include: Based on the dynamic trend, real-time updated data is obtained from order demand and inventory status. The resource allocation is compared with the pre-established database. If the resource allocation cannot meet the order demand, the adjustment process is triggered to determine the preliminary production scheduling plan. For the preliminary production scheduling plan, relevant information on logistics routes and route costs is obtained. The first delivery efficiency is evaluated using a route planning tool. If the first delivery efficiency is lower than a preset threshold, the logistics route is replanned to obtain an adjusted transportation plan. By adjusting the transportation plan, combining the first time window and demand forecast data, and using scheduling management tools to allocate time for production and delivery, if the delivery task cannot be completed within the first time window, the resource allocation will be adjusted a second time to determine the final production scheduling result. Based on the final production scheduling results, data on changes in inventory status and order demand are obtained. Data comparison tools are used to continuously monitor the first delivery efficiency and route cost to determine whether the optimized transportation plan meets the expected goals.

6. The production and logistics management method based on digital twins as described in claim 1, characterized in that, The steps of extracting key decision parameters from the optimized production and transportation plan, and combining them with the full-chain data integration platform to generate cross-link collaborative scheduling instructions include: Real-time data on order demand and inventory status are obtained from the optimized production and transportation plan. The resource allocation is compared with the pre-established database. If the resource allocation cannot meet the order demand, the data integration process is triggered to obtain the preliminary collaborative scheduling parameters. Based on the preliminary collaborative scheduling parameters, relevant data of the logistics path and the second time window are obtained. The second delivery efficiency is evaluated using a path planning tool. If the second delivery efficiency is lower than the preset threshold, the logistics path is replanned, and the adjusted collaborative scheduling instructions are determined. By adjusting the collaborative scheduling instructions, combining demand forecasting and monitoring frequency data, and using scheduling management tools to allocate time for production scheduling and delivery, if the delivery task cannot be completed within the second time window, the resource allocation will be adjusted a second time to determine the final collaborative scheduling scheme. Based on the final collaborative scheduling scheme, dynamic data on inventory status and route costs are obtained. The second delivery efficiency and collaborative scheduling effect are continuously monitored through the data integration platform to obtain optimized cross-link scheduling instructions.

7. The production and logistics management method based on digital twins as described in claim 1, characterized in that, For the aforementioned collaborative scheduling instruction, the steps of decomposing the instruction into specific tasks using a distributed computing framework, allocating them to production equipment and logistics nodes, and obtaining post-execution status feedback include: According to the collaborative scheduling instruction, the instruction is decomposed by a distributed computing tool to generate multiple specific task units. The tasks are initially sorted according to their priority and execution sequence to obtain a decomposed task list. Based on the task list, and combined with the equipment load and node capacity data of production equipment and logistics nodes, tasks are allocated using a resource allocation tool. If the equipment load exceeds a preset threshold, the task allocation order is adjusted to determine the final task assignment scheme. According to the task assignment scheme, specific tasks are issued to production equipment and logistics nodes, and status feedback data during the execution process is collected in real time. The execution efficiency is monitored by the feedback frequency to obtain dynamic updates of the execution status. In response to the dynamic updates of the execution status, if the execution efficiency is lower than the preset threshold, the task allocation is optimized through the instruction adjustment tool. The execution sequence is re-planned based on the device load and node capacity data, and the adjusted task execution arrangement is determined.

8. The production and logistics management method based on digital twins as described in claim 1, characterized in that, Based on the aforementioned status feedback, the steps for updating the parameters of the digital twin model using an online learning algorithm and optimizing the accuracy of subsequent simulation intervention mechanisms include: The execution data of the status feedback is obtained from the production node and the logistics node through data acquisition tools, and the execution data is structured and classified to obtain a list of classified execution data. The execution data list is dynamically monitored using a real-time update tool. If the execution efficiency is lower than a preset threshold, the parameters of the simulation intervention mechanism are corrected using a parameter adjustment tool to determine the adjusted parameter configuration. The parameter configuration is continuously tracked using a feedback frequency monitoring tool to obtain performance data under different business loads. The performance data is compared with the preset accuracy optimization target using a data comparison tool. If the preset threshold is not reached, the historical status feedback is sorted out by executing a data backtracking tool to obtain the updated data basis.

9. The production and logistics management method based on digital twins as described in claim 8, characterized in that, In the step of using a real-time update tool to dynamically monitor the execution data list, and if the execution efficiency is lower than a preset threshold, adjusting the parameters of the simulation intervention mechanism using a parameter adjustment tool to determine the adjusted parameter configuration, the system execution efficiency is calculated in real time using the following formula and compared with the preset threshold: in, Indicates the system during a time period Internal execution efficiency This indicates the total number of sampling points within the monitoring time window. Indicates the first The output quantity of each sampling point Indicates the first Input resource quantity per sampling point Indicates the first The execution time interval for each sampling point; The adjusted parameter configuration is calculated using the following formula: in, This indicates the adjusted new parameter configuration. This indicates the current parameter configuration. This represents the learning rate used to adjust the parameters. Describe the objective function For parameters gradient, This represents the preset efficiency threshold. This represents the actual efficiency value currently monitored.

10. A production and logistics management system based on digital twins, used to execute the production and logistics management method based on digital twins as described in any one of claims 1 to 9, characterized in that, The digital twin-based production and logistics management system includes: The real-time dataset acquisition module (10) is used to acquire real-time data streams from production equipment sensors and logistics nodes, and to clean and format the real-time data streams using streaming processing technology to obtain structured real-time datasets. The anomaly detection module (20) is used to detect equipment anomalies and order fluctuations by applying time series analysis algorithms to structured real-time datasets; The dynamic change trend acquisition module (30) is used to simulate the production and logistics links in real time through a pre-established digital twin model if the equipment abnormality or order fluctuation is detected, so as to obtain the dynamic change trend of the event impact. The planning determination module (40) is used to optimize and adjust the production scheduling and logistics path according to the dynamic change trend, and determine the optimized production and transportation plan; The collaborative scheduling instruction generation module (50) is used to extract key decision parameters from the optimized production and transportation plan, and generate cross-link collaborative scheduling instructions in conjunction with the full-chain data integration platform. The status feedback acquisition module (60) is used to decompose the collaborative scheduling instruction into specific tasks through a distributed computing framework, allocate them to production equipment and logistics nodes, and acquire status feedback after execution. The accuracy optimization module (70) is used to update the parameters of the digital twin model by applying an online learning algorithm based on the state feedback, thereby optimizing the accuracy of the subsequent simulation intervention mechanism.

Images