Factory task management method based on cooperation of Internet of Things

Abstract

The invention discloses a factory task management method based on Internet of Things collaboration. The method comprises the following steps: constructing an Internet of Things collaborative sensing network, deploying multiple types of Internet of Things sensing equipment, and collecting equipment operation state data, material position and inventory data, personnel on-duty and skill data and production environment data in a factory; the method comprises the following steps of: preprocessing and fusing multi-source data, constructing an edge computing node cluster, preprocessing the collected multi-source sensing data, including data cleaning, duplicate removal and standardization processing, and generating a task execution basic data set in a uniform format through a data fusion algorithm. According to the factory task management method based on Internet of Things collaboration, real-time perception and data sharing of multiple elements such as equipment, materials, personnel and environment in a factory are achieved, the situation that data of all systems are isolated in traditional task management is broken through, and comprehensive data support is provided for accurate distribution and collaborative management of tasks.

Description

Technical Field

[0001] This invention relates to the field of intelligent factory management technology, specifically to a factory task management method based on Internet of Things (IoT) collaboration. Background Technology

[0002] As the manufacturing industry transforms towards intelligence and flexibility, factory production tasks are becoming increasingly complex, and multi-variety, small-batch production models are becoming the mainstream. This places higher demands on the efficiency and collaboration of factory task management. Traditional factory task management methods rely heavily on manual planning and paper records, which have problems such as inaccurate task allocation, lagging resource scheduling, and untimely feedback on execution status, resulting in low production efficiency, serious resource waste, and difficulty in adapting to the needs of intelligent production.

[0003] In existing technologies, some factories use information management systems for task management, but most of them suffer from isolated data and poor collaboration among the systems. For example, equipment management systems, material management systems, and personnel management systems operate independently, and data cannot be shared in real time. This makes it impossible to fully consider equipment operating status, material supply, and personnel skill matching when allocating tasks, which can easily lead to problems such as task execution interruption and resource conflicts. At the same time, existing task management methods mostly monitor the task execution process in stages, and cannot achieve real-time tracking of the entire process. When deviations occur in task execution, they cannot be adjusted in time, which in turn affects the overall production progress.

[0004] The development of IoT technology has made it possible to upgrade factory task management to be intelligent. IoT devices can realize real-time perception of multiple elements in the factory. However, how to build an efficient collaborative management mechanism based on IoT technology to achieve dynamic planning, accurate allocation and closed-loop control of tasks is still a technical problem that needs to be solved. Therefore, there is an urgent need for a factory task management method based on IoT collaboration to improve the collaborative efficiency and accuracy of task management, optimize resource allocation and ensure the efficient completion of production tasks. Summary of the Invention

[0005] The purpose of this invention is to provide a factory task management method based on Internet of Things collaboration, so as to solve the problems of poor collaboration, inaccurate task allocation, and untimely monitoring of execution status in existing factory task management technologies.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a factory task management method based on Internet of Things collaboration, comprising the following steps: Step 1: Construct an IoT collaborative sensing network Deploy various types of IoT sensing devices to collect data on equipment operation status, material location and inventory, personnel on-duty and skill data, and production environment data within the factory. Step 2: Multi-source data preprocessing and fusion An edge computing node cluster is built to preprocess the multi-source sensing data collected in step 1, including data cleaning, deduplication, and standardization. Then, a unified format task execution base dataset is generated through a data fusion algorithm. Step 3: Dynamic Programming and Initial Allocation Scheme Generation Based on the task execution dataset, a task requirement model is constructed to clarify the task's process requirements, time nodes, resource requirements, and quality standards. An improved genetic algorithm is used to dynamically plan the task and generate an initial task allocation scheme. Step 4: Optimize and validate the task allocation scheme The feasibility of the initial task allocation scheme is verified, and the optimal task allocation instruction is output by combining the real-time load, skill matching degree and material supply timeliness optimization scheme of each execution unit. Step 5: Real-time monitoring of task instruction issuance and execution status The optimal task allocation instruction is sent to the corresponding execution unit through the Internet of Things communication module, and the status data during the task execution process is collected in real time to build a task execution status monitoring model. Step 6: Task execution deviation adjustment and closed-loop management Based on the task execution status monitoring model, the task execution progress, quality and resource consumption are dynamically evaluated. If a deviation occurs, an adjustment mechanism is triggered to generate a task adjustment instruction and issue it for execution until the task is completed.

[0007] Preferably, the IoT sensing devices in step 1 include vibration sensors, temperature sensors, speed sensors, and operating condition identification sensors installed on production equipment; RFID readers and positioning tags deployed in material warehouses and production lines; and personnel positioning devices, skill identification terminals, temperature and humidity sensors, dust sensors, and light sensors deployed in production workshops. All IoT sensing devices are connected to the IoT gateway via Wi-Fi, Bluetooth, LoRa, and 5G modules.

[0008] Preferably, the specific process of data preprocessing in step 2 is as follows: Data cleaning: Outlier detection algorithms based on the 3σ principle are used to remove outliers, and missing data is supplemented by interpolation. Data deduplication: Based on the comparison of data feature values, duplicate data collection is deleted; Data standardization: The min-max standardization method is used to map different types of perceived data to the [0, 1] interval; Data fusion: A weighted average fusion algorithm is used to generate a task execution base dataset that includes equipment status dataset, material status dataset, personnel status dataset, and environmental status dataset.

[0009] Preferably, the improved genetic algorithm optimization task planning process in step 3 includes: Encoding: Integer encoding is used, with chromosome length equal to the number of tasks. Each gene position corresponds to an execution unit number for a task, and the execution unit is a combination of equipment and personnel. Initialize the population: Determine the population size based on the number of tasks. When the number of tasks is ≤50, the population size is 50-80. When the number of tasks is >50, the population size is 80-120. Fitness function: F=α·(1 / T)+β·U+γ·(1 / C), where T is the total time to complete the task, U is the resource utilization rate, C is the total cost of task execution, and α, β, γ are weight coefficients and α+β+γ=1; Selection operation: Using the roulette wheel selection method; Crossover operation: Uses a single-point crossover method; Mutation operation: Adaptive mutation probability is used, which is adjusted according to the population fitness variance; Iteration Termination: Set the maximum number of iterations to 100-200. Stop iterating when the number of iterations reaches the maximum value or the population fitness tends to stabilize.

[0010] Preferably, the feasibility verification in step 4 includes the equipment operating status of the execution unit, the adequacy of material supply, the matching degree of personnel skills, and the compliance of task execution time. The scheme optimization adopts the analytic hierarchy process, which uses real-time load, skill matching degree, and material supply timeliness as evaluation indicators and sets corresponding weights. The scheme is then adjusted after prioritizing the task allocation of the execution unit.

[0011] Preferably, in step 5, task instructions with high real-time requirements are transmitted via a 5G module, and long-distance task instructions in the workshop are transmitted via a LoRa module; the task execution status monitoring model adopts a random forest algorithm to evaluate the task execution progress, quality and resource consumption in real time and generate a status report.

[0012] Preferably, in step 6, the preset deviation thresholds are 5% for schedule deviation, 3% for quality deviation, and 8% for resource consumption deviation; the deviation types include deviations caused by equipment failure, material shortage, personnel operation, and environmental factors; the adjustment instructions include instructions for equipment maintenance, emergency material scheduling, personnel allocation, and production environment adjustment; and an execution summary report is generated after the task is completed to form a closed-loop management system.

[0013] Compared with the prior art, the beneficial effects of the present invention are: (1) This IoT-based collaborative factory task management method realizes real-time perception and data sharing of multiple elements such as equipment, materials, personnel and environment in the factory, breaks the situation of data isolation of various systems in traditional task management, and provides comprehensive data support for accurate task allocation and collaborative management. (2) The IoT-based collaborative factory task management method introduces adaptive mutation probability, which improves the global search capability and convergence speed of the algorithm. It can generate the optimal task allocation scheme in complex production scenarios. At the same time, it combines the real-time load of the execution unit, skill matching degree and material supply timeliness to optimize the scheme, thereby improving the accuracy and feasibility of task allocation. (3) This IoT-based collaborative factory task management method constructs a closed-loop control mechanism for the entire task execution process. By collecting task execution status data in real time and using machine learning algorithms for dynamic evaluation, it can promptly detect task execution deviations and generate targeted adjustment instructions, ensuring the stability and efficiency of task execution and reducing task execution risks. (4) This IoT-based collaborative factory task management method introduces edge computing nodes to process multi-source sensing data in real time, which reduces data transmission latency, improves data processing efficiency, meets the real-time requirements of factory task management, and reduces cloud computing pressure. (5) This IoT-based collaborative factory task management method can effectively improve the collaborative efficiency of factory task management, optimize resource allocation, reduce task execution costs, shorten task completion cycle, and provide reliable task management support for intelligent factory production. It has broad application prospects. Detailed Implementation

[0014] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] This embodiment takes the engine block machining task management of an automotive parts processing plant as an example, and adopts the factory task management method based on Internet of Things collaboration described in this invention. The specific implementation steps are as follows: Step 1: Construct an IoT collaborative sensing network Based on the factory's production layout and task management needs, deploy various types of IoT sensing devices to collect data on equipment operation status, material location and inventory, personnel on-duty and skill data, and production environment data within the factory. Furthermore, in step 1, the IoT sensing devices include vibration sensors, temperature sensors, speed sensors, and operating condition identification sensors installed on production equipment to collect equipment operating status data, including equipment vibration values, operating temperature, speed, and current operating conditions; RFID readers and positioning tags deployed in material warehouses and production lines to collect data on material location, inventory quantity, material type, and material flow status; and personnel positioning devices, skill recognition terminals, temperature and humidity sensors, dust sensors, and light sensors deployed in the production workshop to collect data on personnel on-duty status, location, skill level, current working status, and production environment data. All IoT sensing devices are connected to an IoT gateway via Wi-Fi, Bluetooth, LoRa, and 5G modules to build a fully covered IoT collaborative sensing network. Specifically, based on the production layout of this automotive parts processing plant, vibration sensors, temperature sensors, speed sensors, and operating condition identification sensors are installed on 10 processing machines (including lathes, milling machines, and grinding machines) on the engine block processing production line. The sensor models are as follows: vibration sensor uses PCB356A15, temperature sensor uses PT100, speed sensor uses Hall effect speed sensor, and operating condition identification sensor uses machine vision sensor; RFID readers (model ImpinjR420) and RFID positioning tags (model ImpinjM730) are deployed in the material warehouse and processing production line. It is used to collect the location, inventory quantity, and circulation status of materials such as cylinder blanks, cutting tools, and finished products; UWB personnel positioning equipment and skill recognition terminals are deployed in the production workshop to collect the on-duty status, location, skill level (such as primary, intermediate, and advanced) and current working status of 20 operators in the workshop; temperature and humidity sensors (model SHT30), dust sensors (model PMS5003), and light sensors (model BH1750) are deployed in the production workshop to collect production environment data. All IoT sensing devices are connected to the IoT gateway through Wi-Fi, LoRa, and 5G modules to build an IoT collaborative sensing network. Step 2: Multi-source data preprocessing and fusion An edge computing node cluster is built to preprocess the multi-source sensing data collected in step 1, including data cleaning, deduplication, and standardization. Then, a unified format task execution base dataset is generated through a data fusion algorithm. Furthermore, the specific process of data preprocessing in step 2 is as follows: Data cleaning: An outlier detection algorithm based on the 3σ principle is used to remove outliers from the perceived data. For example, if the vibration value of a lathe exceeds the mean ± 3 times the standard deviation, it is identified as an outlier and removed. The missing data is supplemented by interpolation. Data deduplication: Based on the comparison of data feature values, duplicate data collection is deleted to reduce data redundancy, such as data of the same RFID tag being read repeatedly within 1 second; Data standardization: The min-max standardization method is used to map data of different dimensions, such as equipment speed (range 500-3000 r / min), material inventory quantity (range 0-100 pieces), and personnel skill level (level 1-3), to the range [0, 1]. Data fusion: A weighted average fusion algorithm is used to generate a task execution base dataset that includes equipment status dataset, material status dataset, personnel status dataset, and environmental status dataset. The weights of equipment status data, material status data, personnel status data, and environmental status data are set to 0.4, 0.3, 0.2, and 0.1 respectively. Step 3: Dynamic Programming and Initial Allocation Scheme Generation Based on the task execution dataset, a task requirement model is constructed to clarify the task's process requirements, time nodes, resource requirements, and quality standards. An improved genetic algorithm is used to dynamically plan the task and generate an initial task allocation scheme. Furthermore, the improved genetic algorithm optimization task planning process in step 3 includes: Encoding: Integer encoding is used to assign each task to a corresponding execution unit. The chromosome length is the number of tasks, and each gene position corresponds to the execution unit number of a task. The execution unit is a combination of equipment and personnel. Initialize the population: Determine the population size based on the number of tasks. When the number of tasks is ≤50, the population size is 50-80. When the number of tasks is >50, the population size is 80-120. Fitness function: F=α·(1 / T)+β·U+γ·(1 / C), where T is the total time to complete the task, U is the resource utilization rate, C is the total cost of task execution, and α, β, γ are weight coefficients and α+β+γ=1. The weight coefficients are adjusted according to the factory production priority. Selection operation: The roulette wheel selection method is used to select superior individuals to enter the next generation of the population; Crossover operation: A single-point crossover method is used, randomly selecting a crossover point and exchanging gene segments after the crossover point of two parent chromosomes to generate offspring chromosomes; Mutation operation: An adaptive mutation probability is adopted, which is adjusted according to the population fitness variance. When the population fitness variance is small, the mutation probability is increased to avoid the algorithm getting trapped in local optima; when the population fitness variance is large, the mutation probability is decreased to ensure the stability of the algorithm. Iteration Termination: Set the maximum number of iterations to 100-200. Stop iterating when the number of iterations reaches the maximum value or the population fitness tends to stabilize, output the optimal chromosome, and generate the initial task allocation scheme based on the optimal chromosome. Specifically, based on the production order, the engine block machining task includes five processes (raw material machining, rough milling, finish milling, drilling, and grinding). Each process has specific requirements, timeframes (total completion time not exceeding 8 hours), resource needs (e.g., rough milling requires intermediate-level skilled personnel, milling machines, and specific cutting tools), and quality standards (e.g., dimensional tolerance ±0.02mm). These requirements are quantified into task parameters using integer encoding. Each gene locus corresponds to an execution unit number for a process task. An execution unit is a combination of equipment and personnel, with a total of 10 machines × 20 personnel = 200 execution units. The chromosome length is 5. For example, chromosome [35, 89, 123, 56, 178] represents the five process tasks assigned to numbers 35, 89, and 123 respectively. The system uses execution units of 56 and 178; the number of tasks is 5; the population size is set to 60; α=0.4, β=0.3, γ=0.3; and the fitness function is F=0.4·(1 / T)+0.3·U+0.3·(1 / C). A roulette wheel selection method is used to select 30 superior individuals to enter the next generation population. A random crossover point of 2 is selected, and gene segments after the crossover points of the two parent chromosomes are exchanged. The adaptive mutation probability range is set to 0.01-0.05, adjusted according to the population fitness variance. When the variance < 0.05, the mutation probability is set to 0.05; when the variance ≥ 0.05, the mutation probability is set to 0.01. The maximum number of iterations is set to 120. When the iteration reaches 80 times, the population fitness tends to stabilize, the iteration stops, the optimal chromosome is output, and the initial task allocation scheme is generated. Step 4: Optimize and validate the task allocation scheme The feasibility of the initial task allocation scheme is verified, and the optimal task allocation instruction is output by combining the real-time load, skill matching degree and material supply timeliness optimization scheme of each execution unit. Furthermore, the feasibility verification in step 4 includes the equipment operating status of the execution unit, the adequacy of material supply, the matching degree of personnel skills, and the compliance of task execution time. If there are infeasible items, the process returns to step 3 to regenerate the initial task allocation plan. The plan optimization adopts the analytic hierarchy process, which uses real-time load (load rate = number of currently executed tasks / maximum number of executable tasks), skill matching degree (calculated by the matching coefficient between personnel skill level and task process requirements), and material supply timeliness (the difference between the time when materials arrive at the execution unit and the time when the task starts) as evaluation indicators and sets corresponding weights. The plan is then adjusted after prioritizing the task allocation of the execution unit. Specifically, the initial task allocation plan was checked to ensure that the equipment in each execution unit was operating normally (e.g., the vibration and temperature of the milling machine assigned to the rough milling task were within normal ranges), that the material supply was sufficient (the number of tools required for the rough milling task was ≥5), that the personnel skills were matched (the personnel were of intermediate skill level), and that the task execution time met the 8-hour requirement. No infeasibility was verified. The Analytic Hierarchy Process (AHP) was used, with real-time load, skill matching degree, and material supply timeliness as evaluation indicators, weighted at 0.3, 0.4, and 0.3 respectively. The task allocation priority of each execution unit was ranked, and the execution unit of one process task in the initial plan was adjusted, moving the grinding task from an execution unit with an 80% load rate to an execution unit with a 50% load rate. The optimal task allocation instruction was then output. Step 5: Real-time monitoring of task instruction issuance and execution status The optimal task allocation instruction is sent to the corresponding execution unit through the Internet of Things communication module, and the status data during the task execution process is collected in real time to build a task execution status monitoring model. Furthermore, in step 5, task instructions with high real-time requirements are transmitted through the 5G module, and long-distance task instructions in the workshop are transmitted through the LoRa module; the task execution status monitoring model adopts the random forest algorithm to evaluate the task execution progress, quality and resource consumption in real time and generate status reports. Specifically, the optimal task allocation instructions are sent to the equipment control terminals, personnel mobile terminals, and material scheduling terminals of each execution unit via the 5G module. For example, a rough milling task instruction is sent to execution unit numbered 89 (milling machine + intermediate operator), and a tool scheduling instruction is sent to the material scheduling terminal. Data during task execution is collected in real time through the Internet of Things collaborative sensing network, such as the milling machine's rotation speed (2000 r / min), the number of tools consumed (2), the operator's work progress (60% completed), and the workshop temperature and humidity (25℃, 50%RH). A task execution status monitoring model is constructed using the random forest algorithm. The collected execution status data is input, and the task execution progress, quality, and resource consumption evaluation results are output to generate a real-time status report. Step 6: Task execution deviation adjustment and closed-loop management Based on the task execution status monitoring model, the task execution progress, quality and resource consumption are dynamically evaluated. If a deviation occurs, an adjustment mechanism is triggered to generate a task adjustment instruction and issue it for execution until the task is completed. Furthermore, in step 6, the preset deviation thresholds are 5% for schedule deviation, 3% for quality deviation, and 8% for resource consumption deviation; the types of deviations include deviations caused by equipment failure, material shortage, personnel operation, and environmental factors; the adjustment instructions include instructions for equipment maintenance, emergency material scheduling, personnel allocation, and production environment adjustment; after the task is completed, an execution summary report is generated to form a closed-loop management system. Specifically, during the rough milling task, the monitoring model detected a 6% progress deviation, exceeding the preset 5% threshold, and thus determined it to be a progress deviation. Data analysis revealed that the progress deviation was due to decreased equipment processing efficiency caused by tool wear. A tool replacement instruction was generated and sent to the material scheduling terminal and the operator's mobile terminal. After the operator completed the tool replacement, processing efficiency data was collected in real time, and the progress deviation was found to have decreased to 2%, thus eliminating the deviation. After all tasks were completed, the task execution data was summarized, and a task execution summary report was generated, including the completion time of each task, resource consumption, and quality achievement status, providing data support for subsequent cylinder block machining task management.

[0016] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0017] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A factory task management method based on Internet of Things (IoT) collaboration, characterized in that, Includes the following steps: Step 1: Construct an IoT collaborative sensing network Deploy various types of IoT sensing devices to collect data on equipment operation status, material location and inventory, personnel on-duty and skill data, and production environment data within the factory. Step 2: Multi-source data preprocessing and fusion An edge computing node cluster is built to preprocess the multi-source sensing data collected in step 1, including data cleaning, deduplication, and standardization. Then, a unified format task execution base dataset is generated through a data fusion algorithm. Step 3: Dynamic Programming and Initial Allocation Scheme Generation Based on the task execution dataset, a task requirement model is constructed to clarify the task's process requirements, time nodes, resource requirements, and quality standards. An improved genetic algorithm is used to dynamically plan the task and generate an initial task allocation scheme. Step 4: Optimize and validate the task allocation scheme The feasibility of the initial task allocation scheme is verified, and the optimal task allocation instruction is output by combining the real-time load, skill matching degree and material supply timeliness optimization scheme of each execution unit. Step 5: Real-time monitoring of task instruction issuance and execution status The optimal task allocation instruction is sent to the corresponding execution unit through the Internet of Things communication module, and the status data during the task execution process is collected in real time to build a task execution status monitoring model. Step 6: Task execution deviation adjustment and closed-loop management Based on the task execution status monitoring model, the task execution progress, quality and resource consumption are dynamically evaluated. If a deviation occurs, an adjustment mechanism is triggered to generate a task adjustment instruction and issue it for execution until the task is completed.

2. The factory task management method based on IoT collaboration according to claim 1, characterized in that: The IoT sensing devices mentioned in step 1 include vibration sensors, temperature sensors, speed sensors, and operating condition identification sensors installed on production equipment; RFID readers and positioning tags deployed in material warehouses and production lines; and personnel positioning devices, skill identification terminals, temperature and humidity sensors, dust sensors, and light sensors deployed in production workshops. All IoT sensing devices are connected to the IoT gateway via Wi-Fi, Bluetooth, LoRa, and 5G modules.

3. The factory task management method based on IoT collaboration according to claim 1, characterized in that: The specific process of data preprocessing in step 2 is as follows: Data cleaning: Outlier detection algorithms based on the 3σ principle are used to remove outliers, and missing data is supplemented by interpolation. Data deduplication: Based on the comparison of data feature values, duplicate data collection is deleted; Data standardization: The min-max standardization method is used to map different types of perceived data to the [0, 1] interval; Data fusion: A weighted average fusion algorithm is used to generate a task execution base dataset that includes equipment status dataset, material status dataset, personnel status dataset, and environmental status dataset.

4. The factory task management method based on IoT collaboration according to claim 1, characterized in that: The improved genetic algorithm optimization task planning process in step 3 includes: Encoding: Integer encoding is used, with chromosome length equal to the number of tasks. Each gene position corresponds to an execution unit number for a task, and the execution unit is a combination of equipment and personnel. Initialize the population: Determine the population size based on the number of tasks. When the number of tasks is ≤50, the population size is 50-80. When the number of tasks is >50, the population size is 80-120. Fitness function: F=α·(1 / T)+β·U+γ·(1 / C), where T is the total time to complete the task, U is the resource utilization rate, C is the total cost of task execution, and α, β, γ are weight coefficients and α+β+γ=1; Selection operation: Using the roulette wheel selection method; Crossover operation: Uses a single-point crossover method; Mutation operation: Adaptive mutation probability is used, which is adjusted according to the population fitness variance; Iteration Termination: Set the maximum number of iterations to 100-200. Stop iterating when the number of iterations reaches the maximum value or the population fitness tends to stabilize.

5. The factory task management method based on IoT collaboration according to claim 1, characterized in that: The feasibility verification in step 4 includes the equipment operating status of the execution unit, the adequacy of material supply, the matching degree of personnel skills, and the compliance of task execution time. The scheme optimization adopts the analytic hierarchy process, which uses real-time load, skill matching degree, and material supply timeliness as evaluation indicators and sets corresponding weights. The scheme is adjusted after prioritizing the task allocation of the execution unit.

6. The factory task management method based on IoT collaboration according to claim 1, characterized in that: In step 5, task instructions with high real-time requirements are transmitted via the 5G module, and long-distance task instructions in the workshop are transmitted via the LoRa module. The task execution status monitoring model adopts the random forest algorithm to evaluate the task execution progress, quality and resource consumption in real time and generate status reports.

7. The factory task management method based on IoT collaboration according to claim 1, characterized in that: In step 6, the preset deviation thresholds are 5% for schedule deviation, 3% for quality deviation, and 8% for resource consumption deviation. Deviation types include deviations caused by equipment failure, material shortage, personnel operation, and environmental factors. Adjustment instructions include instructions for equipment maintenance, emergency material scheduling, personnel allocation, and production environment adjustment. After the task is completed, an execution summary report is generated to form a closed-loop management system.