Digital management platform and management method based on intelligent manufacturing

Abstract

The invention discloses a digital management platform and management method based on intelligent manufacturing, and relates to the technical field of intelligent manufacturing management, and the platform comprises the steps: obtaining a target manufacturing task queue of a target manufacturing production line, and obtaining a target manufacturing flow information queue; obtaining a task preset start time queue; obtaining a resource reliability coefficient queue; performing equipment fault probability analysis to obtain an equipment reliability coefficient queue; performing mapping task reliability maximum likelihood estimation to obtain a task reliability evaluation value queue; obtaining a task execution time tolerance threshold queue; obtaining a task adjustment start time queue; and performing production management on the target manufacturing production line based on the task adjustment start time queue. The technical problems that in the prior art, production task scheduling is not accurate, resource configuration is not optimized, and equipment failure prediction is insufficient are solved, and the technical effects of improving task scheduling efficiency, optimizing resource configuration and strengthening equipment reliability prediction and adjustment capacity are achieved.

Description

Technical Field

[0001] The present application relates to the technical field related to intelligent manufacturing management, and specifically to a digital management platform and management method based on intelligent manufacturing. Background Art

[0002] With the rapid development of intelligent manufacturing and industry, the traditional production management model can no longer fully meet the requirements of modern manufacturing for flexibility, efficiency and precision. In modern manufacturing, the complexity and variability of production task arrangement, resource scheduling, equipment management and other links have greatly increased the uncertainty in the production process. Traditional manual management or experience-based management models often cannot effectively cope with these challenges. How to achieve efficient production scheduling, resource management, equipment maintenance and fault prediction in a digital and networked environment has become an urgent problem to be solved in the current manufacturing industry. Especially in the process of production task scheduling, manufacturing enterprises need to face multiple problems such as reasonable scheduling of tasks, equipment maintenance and fault management, resource allocation and scheduling, etc. The current scheduling method cannot dynamically cope with the uncertainty in production tasks.

[0003] Therefore, in the current relevant technologies, there are technical problems such as inaccurate production task scheduling, unoptimized resource allocation, and insufficient equipment failure prediction. Summary of the invention

[0004] This application solves the technical problems of inaccurate production task scheduling, unoptimized resource allocation, and insufficient equipment failure prediction in the prior art by providing a digital management platform and management method based on intelligent manufacturing, thereby achieving the technical effects of improving task scheduling efficiency, optimizing resource allocation, and enhancing equipment reliability prediction and adjustment capabilities.

[0005] The present application provides a digital management platform based on intelligent manufacturing, the platform comprising: an information queue acquisition module, used to acquire a target manufacturing task queue of a target manufacturing production line, and perform a manufacturing process of each target manufacturing task according to the target manufacturing task queue to obtain a target manufacturing process information queue; a start time queue acquisition module, used to extract the preset start time of the task of the target manufacturing task queue respectively, to obtain a task preset start time queue; a resource reliability analysis module, used to traverse the target manufacturing task queue and the target manufacturing process information queue to perform resource reliability analysis, to obtain a resource reliability coefficient queue; an equipment failure probability analysis module, used to traverse the target manufacturing task queue and the target manufacturing process information queue to perform equipment failure probability analysis, and calculate the difference between the analysis result and 1 as the equipment reliability coefficient, obtaining an equipment reliability coefficient queue; a reliability estimation module, used to map the resource reliability coefficient queue and the equipment reliability coefficient queue to perform maximum likelihood estimation of task reliability, and obtain a task reliability evaluation value queue; a task duration analysis module, used to perform task duration analysis based on the target manufacturing process information queue, the task reliability evaluation value queue and the task preset start time queue, and obtain a task execution time tolerance threshold queue; a conflict identification module, used to perform conflict identification on the task execution time tolerance threshold queue, and perform associated adjustment of the task preset start time according to the conflict identification result, and obtain a task adjustment start time queue; a production management module, used to perform production management on the target manufacturing production line based on the task adjustment start time queue.

[0006] In a possible implementation, the digital management method based on intelligent manufacturing also performs the following processing: constructing a resource demand matrix according to the target manufacturing task queue and the target manufacturing process information queue, respectively, to obtain a resource demand matrix queue; obtaining a resource configuration list of the target manufacturing production line; performing a resource reliability analysis on the resource demand matrix queue based on the resource configuration list, to obtain the resource reliability coefficient queue.

[0007] In a possible implementation, the digital management method based on intelligent manufacturing also performs the following processing: pre-constructing a task reliability likelihood function; performing a one-to-one mapping between the resource reliability coefficient queue and the equipment reliability coefficient queue to obtain a mapped task reliability analysis result queue; using the maximum likelihood method to perform likelihood estimation on the mapped task reliability analysis result queue respectively through the task reliability likelihood function to obtain a task reliability maximum likelihood estimator queue; and using the task reliability maximum likelihood estimator queue as the task reliability evaluation value queue.

[0008] In a possible implementation, the digital management method based on intelligent manufacturing further performs the following processing: the task reliability likelihood function is: ; in, is the resource reliability coefficient, is the equipment reliability factor, It's a sample The estimated value of It's a sample The estimated value of is the sample value The corresponding probability value is is the number of subsamples, .

[0009] In a possible implementation, the digital management method based on intelligent manufacturing also performs the following processing: obtaining a sub-sample sequence queue based on the mapping task reliability analysis result queue; inputting the sub-sample sequence queues into the task reliability likelihood function for maximum likelihood estimation, and obtaining a task reliability maximum likelihood estimation queue.

[0010] In a possible implementation, the digital management method based on intelligent manufacturing also performs the following processing: pre-constructing a task execution time analyzer; using the task execution time analyzer to analyze the target manufacturing process information queue and the task reliability assessment value queue to obtain a task duration tolerance threshold queue; combining the task duration tolerance threshold queue and the task preset start time queue for time integration to obtain the task execution time tolerance threshold queue.

[0011] In a possible implementation, the digital management method based on intelligent manufacturing also performs the following processing: obtaining multiple sample target manufacturing process information, multiple sample task reliability assessment values ​​and multiple sample task duration tolerance thresholds as training samples; using the training samples to perform supervised learning on a framework constructed based on a feedforward neural network until the training converges, thereby obtaining the trained task execution time analyzer.

[0012] The present application also provides a digital management method based on intelligent manufacturing, the method comprising: obtaining a target manufacturing task queue of a target manufacturing production line, and performing a manufacturing process of each target manufacturing task according to the target manufacturing task queue to obtain a target manufacturing process information queue; respectively extracting the task preset start time of the target manufacturing task queue to obtain a task preset start time queue; traversing the target manufacturing task queue and the target manufacturing process information queue to perform resource reliability analysis to obtain a resource reliability coefficient queue; traversing the target manufacturing task queue and the target manufacturing process information queue to perform equipment failure probability analysis, and taking the difference between the analysis result and 1 as the equipment reliability coefficient to obtain an equipment reliability coefficient queue; mapping the resource reliability coefficient queue and the equipment reliability coefficient queue to perform task reliability maximum likelihood estimation to obtain a task reliability evaluation value queue; performing task duration analysis based on the target manufacturing process information queue, the task reliability evaluation value queue and the task preset start time queue to obtain a task execution time tolerance threshold queue; performing conflict identification on the task execution time tolerance threshold queue, and performing task preset start time association adjustment according to the conflict identification result to obtain a task adjustment start time queue; and performing production management on the target manufacturing production line based on the task adjustment start time queue.

[0013] Through the digital management platform and management method based on intelligent manufacturing proposed in this application, the target manufacturing task queue of the target manufacturing production line is obtained, and the target manufacturing process information queue is obtained; the task preset start time queue is obtained; the resource reliability coefficient queue is obtained; the equipment failure probability analysis is performed to obtain the equipment reliability coefficient queue; the maximum likelihood estimation of the mapping task reliability is performed to obtain the task reliability evaluation value queue; the task execution time tolerance threshold queue is obtained; the task adjustment start time queue is obtained; and production management of the target manufacturing production line is performed based on the task adjustment start time queue. The technical problems of inaccurate production task scheduling, unoptimized resource allocation, and insufficient equipment failure prediction existing in the prior art are solved, and the technical effects of improving task scheduling efficiency, optimizing resource allocation, and strengthening equipment reliability prediction and adjustment capabilities are achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure are briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the platform according to the embodiment of the present application. It should be understood that the previous or following operations are not necessarily performed precisely in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0015] Figure 1A schematic diagram of the structure of a digital management platform based on intelligent manufacturing provided in an embodiment of the present application; Figure 2 A flowchart of a digital management method based on intelligent manufacturing provided in an embodiment of the present application.

[0016] Explanation of the accompanying drawings: information queue acquisition module 10, start time queue acquisition module 20, resource reliability analysis module 30, equipment failure probability analysis module 40, reliability estimation module 50, task duration analysis module 60, conflict identification module 70, production management module 80. DETAILED DESCRIPTION

[0017] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0018] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0019] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, platform, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0020] The present application embodiment provides a digital management platform based on intelligent manufacturing, such as Figure 1 As shown, the platform includes: The information queue obtaining module 10 is used to obtain the target manufacturing task queue of the target manufacturing production line, and perform the manufacturing process of each target manufacturing task according to the target manufacturing task queue to obtain the target manufacturing process information queue.

[0021] Preferably, a set of manufacturing tasks to be executed is extracted from the target manufacturing production line as a target manufacturing task queue, which may include different production steps, processes or production instructions, representing the specific operations that need to be completed in the entire production process. The manufacturing process of each target manufacturing task is then executed according to the target manufacturing task queue to generate specific manufacturing process information, which may include the processing time of the process, required equipment, resource consumption, process path, etc. When executing the manufacturing task, all process information is recorded and stored in the order of the tasks to form a target manufacturing process information queue, which can reflect the progress of each task on the production line, the required resource allocation, and potential bottlenecks or problems, thereby providing an accurate data basis for subsequent resource reliability analysis, equipment failure prediction and production optimization.

[0022] The start time queue obtaining module 20 is used to respectively extract the task preset start time of the target manufacturing task queue to obtain the task preset start time queue.

[0023] Preferably, extracting the preset start time of the tasks of the target manufacturing task queue separately means that for each manufacturing task, the preset start time of the task is extracted from the task queue, and then multiple task preset start time queues are obtained, wherein the preset start time refers to the scheduled start time set for the production task according to the production scheduling plan, that is, the preset start time of the task, which is usually determined in the production planning stage according to the urgency of the task, the availability of resources and the load of the production line. The preset start times of all the extracted manufacturing tasks are organized in the order of the task queue to form a task preset start time queue. Each element of this queue corresponds to the start time of the corresponding task in the task queue, so as to record and display the planned start time of all tasks, and the start time of the task can be dynamically adjusted according to this queue to optimize the production scheduling.

[0024] The resource reliability analysis module 30 is used to traverse the target manufacturing task queue and the target manufacturing process information queue to perform resource reliability analysis and obtain a resource reliability coefficient queue.

[0025] Preferably, the target manufacturing task queue (production requirements of multiple production links, such as material preparation, processing, assembly, etc.) and the target manufacturing process information queue (detailed manufacturing process information of each task, including the resources required for each task, the required equipment, production sequence, etc.), are traversed to obtain each task and the associated manufacturing process information, and perform resource reliability analysis, that is, to evaluate the reliability of various resources involved in the production process (such as raw materials, tools, etc.). Specifically, reliability evaluation refers to material reliability evaluation (such as whether the raw materials are in place within the scheduled time, whether the supply chain is smooth, etc.). The core goal of resource reliability analysis is to evaluate the problems that may exist in various resources in the actual production process. Problems such as material shortages can be identified, and the impact of these problems on production tasks can be predicted. By analyzing the resource reliability of each task, multiple resource reliability coefficients can be obtained to indicate the reliability of the resources required for each task. The resource reliability coefficient is usually a numerical value, indicating the reliability of a certain resource (such as personnel, materials, etc.) in the production process. The possible value range is from 0 (completely unavailable) to 1 (completely reliable). Finally, a resource reliability coefficient queue is formed, which corresponds to the resource reliability score of the corresponding task in the task queue. This helps the platform understand the resource risks that may be encountered during the execution of each task, and take measures (such as adjusting the task sequence, preparing materials in advance, etc.) to ensure the smooth progress of production tasks.

[0026] Furthermore, the specific configuration of the resource reliability analysis module 30 also includes constructing a resource demand matrix according to the target manufacturing task queue and the target manufacturing process information queue, respectively, to obtain a resource demand matrix queue; obtaining a resource configuration list of the target manufacturing production line; and performing a resource reliability analysis on the resource demand matrix queue based on the resource configuration list to obtain the resource reliability coefficient queue.

[0027] Preferably, a resource demand matrix is ​​constructed according to the target manufacturing task queue and the target manufacturing process information queue, that is, for each manufacturing task, all resource requirements it depends on are extracted, including equipment resources (such as machines, production lines, etc.), personnel resources (such as operators, technicians, inspectors, etc.) and material resources (such as raw materials, parts, etc.), and then the resource requirements of each task in the task queue are constructed as a matrix, the rows of the matrix represent tasks, the columns represent resource types (such as equipment, personnel, raw materials, etc.), and each element of the matrix represents the demand for a certain resource for a specific task, that is, a resource demand matrix queue is obtained, which intuitively represents the resource demand relationship of all tasks; then the resource configuration list of the target manufacturing production line is obtained, that is, a list of all currently available resources of the production line is obtained, including information such as the type, quantity, status, and usage time of the resources; then the resource requirements are calculated based on the resource configuration list. The matrix queue performs resource reliability analysis to evaluate the reliability of production line resources in meeting the requirements of each task. Specifically, the requirements of each task in the resource requirement matrix are compared with the actual supply in the resource configuration list. If the resources in the configuration list can meet the requirements in the matrix, the resource reliability is high. If the resource configuration is insufficient (for example, insufficient equipment time or insufficient inventory materials), the resource reliability is low. The resource reliability coefficient is then calculated to quantify the ability of resources to meet requirements. For example, the ratio of actual available resources to task required resources is calculated as the resource reliability coefficient. The higher the resource reliability coefficient, the more reliable the resource support for the task; the lower the resource reliability coefficient, the more likely the resource may not be able to fully meet the task requirements. A resource reliability coefficient queue is composed of multiple resource reliability coefficients, where each coefficient corresponds to a manufacturing task, which helps to improve resource utilization efficiency.

[0028] The equipment failure probability analysis module 40 is used to traverse the target manufacturing task queue and the target manufacturing process information queue to perform equipment failure probability analysis, and use the difference between the analysis result and 1 as the equipment reliability coefficient to obtain the equipment reliability coefficient queue.

[0029] Preferably, the target manufacturing task queue and the target manufacturing process information queue are traversed, and the equipment failure probability analysis is performed on each task and its corresponding manufacturing process information one by one to determine the failure risk of the production equipment contained in each task. Specifically, the equipment failure probability analysis is a quantitative assessment of the risk of possible failure of the equipment required in the manufacturing process, which usually includes calculating the failure probability of the equipment based on the historical data of the equipment (such as failure records, maintenance frequency, service life, etc.) and real-time operating status (such as temperature, vibration, workload, etc.); using common failure prediction models (such as reliability mathematical models, machine learning prediction algorithms) to evaluate the possibility of each device failing in the current task; the analysis results include the equipment failure probability, which indicates the probability of failure of the equipment during the execution of the task, and then the analysis The difference between the result and 1 is taken as the equipment reliability coefficient, where equipment reliability is used to reflect the reliability of the equipment. Equipment reliability coefficient = 1-equipment failure probability. The higher the reliability coefficient of the equipment, the less likely the equipment is to fail and can work smoothly during the task execution process, thereby quantifying the equipment failure risk as a positive reliability indicator. Similarly, a failure probability analysis is performed on each task and corresponding equipment in the target manufacturing task queue to obtain the equipment reliability coefficients of multiple equipment in the target manufacturing task. These reliability coefficients are organized in task order to form an equipment reliability coefficient queue for risk warning, task optimization and resource allocation. This can reduce the impact of equipment failures on production plans during dynamic task execution, thereby improving the reliability and stability of the entire manufacturing process.

[0030] The reliability estimation module 50 is used to perform maximum likelihood estimation of task reliability on the resource reliability coefficient queue and the device reliability coefficient queue to obtain a task reliability evaluation value queue.

[0031] Preferably, the resource reliability coefficient queue (the reliability coefficient of each resource associated with the production task, indicating the reliability of the resource when executing the task) and the equipment reliability coefficient queue (the reliability coefficient of each equipment related to the production task, indicating the probability of failure of the equipment during the task execution) are mapped to the maximum likelihood estimation of the task reliability, wherein the task reliability maximum likelihood estimation (MLE) is a statistical method that estimates the most likely task reliability by modeling given data (resource reliability coefficient and equipment reliability coefficient). Specifically, the mapping task reliability maximum likelihood estimation refers to inferring the overall reliability of the task based on resource reliability and equipment reliability. The resource reliability and equipment reliability are mapped to the task respectively, and the data of the two queues are matched according to the task order, and they are combined Together, the overall reliability of the task is formed. The overall reliability of each task is calculated by combining the reliability coefficients of all resources and equipment related to the task. The task reliability is jointly affected by the failure probability of all resources and equipment. The maximum likelihood estimation calculates the most likely task reliability through the known resource and equipment reliability, that is, the estimation error of the overall reliability of each task is minimized; the reliability coefficients of the resources and equipment of each task are comprehensively calculated through the maximum likelihood estimation method, and finally the reliability evaluation values ​​of multiple tasks are obtained, which indicates the probability of the task being successfully completed as planned. It is usually a value between 0 and 1. The larger the value, the higher the possibility of the task being completed on time. It also forms a task reliability evaluation value queue, which can improve production efficiency, reduce production risks caused by equipment failures, and optimize the overall performance of the production process.

[0032] Furthermore, the specific configuration of the reliability estimation module 50 also includes pre-building a task reliability likelihood function; performing a one-to-one mapping between the resource reliability coefficient queue and the equipment reliability coefficient queue to obtain a mapping task reliability analysis result queue; utilizing the maximum likelihood method to perform likelihood estimation on the mapping task reliability analysis result queue respectively through the task reliability likelihood function to obtain a task reliability maximum likelihood estimator queue; and using the task reliability maximum likelihood estimator queue as the task reliability evaluation value queue.

[0033] The specific configuration of the reliability estimation module 50 further includes that the task reliability likelihood function is: ; in, is the resource reliability coefficient, is the equipment reliability factor, It's a sample The estimated value of It's a sample The estimated value of is the sample value The corresponding probability value is is the number of subsamples, .

[0034] Preferably, the task reliability likelihood function is constructed using the resource reliability coefficient and the equipment reliability coefficient. , used to quantify the joint impact of resource and equipment reliability on task reliability, where is the resource reliability coefficient (such as the reliability of materials and manpower), is the equipment reliability factor (such as the operational reliability of the equipment), It's a sample The estimated value of It's a sample The estimated value of is the sample value The corresponding probability value is is the number of subsamples, ; One-to-one mapping is performed on the resource reliability coefficient queue and the equipment reliability coefficient queue, that is, the resource reliability coefficient and the equipment reliability coefficient are combined in accordance with the task order to generate a mapping task reliability analysis result queue, which represents the joint state of resource and equipment reliability of each task; then the maximum likelihood method is used to estimate the likelihood of the mapping task reliability analysis result queue, that is, based on the task reliability likelihood function, the maximum likelihood function is optimized. and The value of makes the sample and The joint probability value of Maximization, which involves calculating the likelihood of each task Optimize the parameters based on the likelihood values ​​of all tasks and Then the maximum likelihood estimator is output, which is the optimal value describing the task reliability, and multiple task reliability maximum likelihood estimators (i.e. the reliability index of each task) are generated to form a task reliability maximum likelihood estimator queue. Finally, the maximum likelihood estimator queue is used as the task reliability evaluation value queue, which directly reflects the overall reliability of each task under the conditions of resource and equipment reliability, thereby ensuring the smooth completion of the task under the conditions of resource and equipment.

[0035] Furthermore, the specific configuration of the reliability estimation module 50 also includes obtaining a sub-sample sequence queue according to the mapping task reliability analysis result queue; inputting the sub-sample sequence queues into the task reliability likelihood function for maximum likelihood estimation, and obtaining a task reliability maximum likelihood estimation queue.

[0036] Preferably, the mapping task reliability analysis result queue is decomposed into multiple sub-sequences, and the sub-sequence queue is extracted from the mapping task reliability analysis result queue (including the resource and equipment reliability analysis results corresponding to each task) for further reliability analysis, wherein the sub-sequence usually represents the refined results of reliability evaluation under different resource and equipment conditions for a certain task, and then the sub-sequence queue is input into the task reliability likelihood function for maximum likelihood estimation, that is, through the task reliability likelihood function, the parameter value that makes the data observation most likely is found, specifically including taking the sub-sequence of each task (that is, the resource and equipment reliability of the task) as input, inputting the task reliability likelihood function, calculating the reliability of each task by maximizing the task reliability likelihood function, that is, calculating the probability of task execution for each sub-sequence, and obtaining the reliability index of each task by performing maximum likelihood estimation on each sub-sequence, and then generating a task reliability maximum likelihood estimator queue, which reflects the true reliability of the task under specific resource and equipment conditions, thereby providing strong support for production scheduling and task management.

[0037] The task duration analysis module 60 is used to perform task duration analysis based on the target manufacturing process information queue, the task reliability evaluation value queue and the task preset start time queue to obtain a task execution time tolerance threshold queue.

[0038] Preferably, the target manufacturing process information queue, the task reliability evaluation value queue and the task preset start time queue are used as key input data to perform task duration analysis, that is, taking into account the planned duration of the task itself and possible delays or advances, predicting the possible time range of each task in actual execution, specifically including extracting the expected duration of the task from the target manufacturing process information (for example, a task is expected to last 4 hours); combining the task reliability evaluation value, analyzing how the actual completion time of the task may be affected by equipment or resource problems, for example, if the task reliability is low (such as 0.6), it means that the task may require additional time to complete; combining the task preset start time, analyzing the possible start and end time range of the task, as well as the time relationship with other tasks; through task duration analysis, predicting the time range of the task under theoretical conditions and under various influences, and then Multiple task execution time tolerance thresholds are obtained, indicating that each task corresponds to a time period, wherein the task execution time tolerance threshold is the tolerable execution time range of each task, which is used to measure the time elasticity of task completion, that is, the maximum deviation range of the actual completion time of the task relative to the planned time without affecting the overall production plan. Specifically, the theoretical execution time of the task is obtained according to the target manufacturing process information, and the possible delay time or early completion time of the task is deduced according to the task reliability evaluation value. Combined with the preset start time of the task and the scheduling of other tasks, the actual time limit of each task is analyzed, and finally the tolerance threshold of each task is obtained, that is, the range within which the actual execution time of the task can be adjusted, and the tolerance threshold of each task is combined to form a task execution time tolerance threshold queue, thereby realizing dynamic scheduling and risk control in the intelligent manufacturing process.

[0039] Furthermore, the specific configuration of the task duration analysis module 60 also includes pre-building a task execution time analyzer; using the task execution time analyzer to analyze the target manufacturing process information queue and the task reliability assessment value queue to obtain a task duration tolerance threshold queue; combining the task duration tolerance threshold queue and the task preset start time queue for time integration to obtain the task execution time tolerance threshold queue.

[0040] Preferably, a task execution time analyzer is pre-built based on a neural network model to predict the actual execution time of each task and evaluate its time elasticity (i.e., tolerance threshold). Specifically, the neural network model is trained in combination with historical task data, resource and equipment status, and manufacturing process information. That is, historical production data is used as a training data set, and the model learns the execution time pattern of the task and the influencing factors (such as equipment failure, resource shortage, etc.) to obtain an analyzer that can predict the task execution time. The input features include task type, equipment reliability, resource configuration, historical task execution time, process information, etc. The output result predicts the execution time of the task and its time range (shortest time, longest time). The task execution time analyzer is then used to analyze the target manufacturing process information queue and the task reliability evaluation value queue, including converting the target manufacturing process information (such as task The task execution time analyzer inputs the task execution time into the task execution time analyzer, predicts the actual execution time of the task, considers the equipment reliability, resource status and manufacturing process complexity, generates the time range (shortest execution time, longest execution time), and calculates the time elasticity (tolerance threshold); finally, the task duration tolerance threshold queue and the task preset start time queue are combined for time integration, that is, the preset start time of each task is combined to calculate the possible end time of the task (preset start time + tolerance threshold). If the end time of the task overlaps with the preset start time of the subsequent task, the start time or tolerance threshold of the task is adjusted to generate the task execution time tolerance threshold queue, which represents the available time range of each task in the production plan and avoids time conflicts between tasks, thereby improving the executability and efficiency of the production plan.

[0041] Furthermore, the specific configuration of the task duration analysis module 60 also includes obtaining multiple sample target manufacturing process information, multiple sample task reliability assessment values ​​and multiple sample task duration tolerance thresholds as training samples; using the training samples to perform supervised learning on a framework constructed based on a feedforward neural network until the training converges, thereby obtaining the trained task execution time analyzer.

[0042] Preferably, multiple sample target manufacturing process information (process flow, resource allocation, equipment usage, etc. required for each task execution), multiple sample task reliability evaluation values ​​(task reliability calculated for each task in historical execution) and multiple sample task duration tolerance thresholds (statistical data of actual execution time range of historical tasks, including the shortest duration, the longest duration and the allowed time elasticity range of the task) are obtained from historical production data records, and these data are used as training samples, and then the training samples are used to supervise the learning of a framework built based on a feedforward neural network, wherein a feedforward neural network (FNN) is a common artificial neural network model, including an input layer, a hidden layer and an output layer. Specifically, the target manufacturing process information (such as task The model takes the task reliability evaluation value (such as the reliability of the task under the conditions of resources and equipment) as input features, and the task duration tolerance threshold as a label. The input data is calculated through the network framework, and the predicted task duration range is output. The task duration range predicted by the model and the actual duration tolerance threshold are compared, and the loss function value (such as mean square error) is calculated. The model parameters are adjusted by the gradient descent method to minimize the loss function value. The above process is repeated until the model training reaches a convergence state (the loss value no longer decreases significantly), and then a trained task execution time analyzer is obtained, which can predict the task duration range and tolerance threshold according to the new target manufacturing process information and task reliability evaluation value to ensure that the task is completed on time.

[0043] The conflict identification module 70 is used to perform conflict identification on the task execution time tolerance threshold queue, and adjust the task preset start time according to the conflict identification result to obtain the task adjustment start time queue.

[0044] Preferably, conflict identification is performed on the task execution time tolerance threshold queue, that is, by analyzing the task execution time tolerance threshold, it is determined whether there is a time conflict or overlap between tasks, specifically including task time range overlap, checking the tolerance thresholds of multiple tasks, and determining whether there is a time conflict. For example, if the tolerance threshold of task A is [08:00-10:00] and the tolerance threshold of task B is [09:30-11:30], there is a conflict between the two because there is an overlap in the time period [09:30-10:00]; resource dependency conflict, if multiple tasks need to share the same resource (such as equipment or manpower), check whether the task execution time overlaps. For example, if a certain device is scheduled by two tasks at the same time, an overlap in task time will cause a resource conflict; equipment maintenance, the execution time of the task is inconsistent with the equipment maintenance plan or other unplanned Unavailable time conflict; according to the conflict identification results, the preset start time of the task is adjusted to resolve the conflict after the conflict is identified, according to the type and severity of the conflict, such as adjusting tasks that can be flexibly changed (the preset start time of high-priority tasks should remain unchanged as much as possible, while low-priority tasks can be adjusted). When adjusting tasks, it is necessary to ensure that resources can be matched in time. For example, if the start time of a task is adjusted, it may cause a certain device to be unavailable during this time period, then the adjustment plan needs to be re-evaluated. For each task, a task adjustment time is obtained, and multiple task adjustment start times are combined into a task adjustment start time queue, recording the latest planned start time of each task, optimizing the time arrangement between tasks, ensuring that all tasks can be executed smoothly without conflicts, and ensuring efficient operation and stable output of the production line.

[0045] The production management module 80 is used to perform production management on the target manufacturing production line based on the task adjustment start time queue.

[0046] Preferably, production management of the target manufacturing production line is performed according to the task adjustment time of the task adjustment start time queue, that is, all resources, tasks and equipment in the manufacturing production line are coordinated and scheduled to ensure that production activities are carried out efficiently and orderly, including, according to the adjusted task start time, allocating each task to the appropriate position on the production line according to the latest schedule, ensuring that tasks are executed according to priority and sequence, and avoiding interruptions to the production process due to unreasonable scheduling; reasonably allocating production resources (such as raw materials, equipment, manpower, etc.) according to the adjusted task time, ensuring that each task can obtain the required resources when executing, dynamically monitoring the use of resources, and reallocating them according to actual needs; coordinating the production line settings in combination with the task adjustment time. The use of equipment can ensure efficient switching between tasks, and equipment maintenance or debugging time can be arranged in advance to avoid delays caused by equipment failure during task execution; based on the adjusted schedule, the progress of task execution can be tracked in real time to ensure that tasks are completed as planned. If task delays or insufficient resources are found during execution, the subsequent task time can be adjusted quickly and the task adjustment time queue can be regenerated; according to the adjusted task time schedule, the critical path in the production line is analyzed to identify possible bottleneck links (such as insufficient resource supply and excessive equipment load), and the bottleneck is alleviated by optimizing scheduling, dispatching spare resources, etc.; thereby ensuring the efficient use of resources, equipment and time, so that production activities can be carried out in an orderly manner and production efficiency can be improved.

[0047] In the above, refer to Figure 1 A digital management platform based on intelligent manufacturing according to an embodiment of the present invention is described in detail. Figure 2 A digital management method based on intelligent manufacturing according to an embodiment of the present invention is described.

[0048] A digital management method based on intelligent manufacturing, such as Figure 2As shown, the method includes: obtaining a target manufacturing task queue of a target manufacturing production line, and performing a manufacturing process of each target manufacturing task according to the target manufacturing task queue to obtain a target manufacturing process information queue; extracting the task preset start time of the target manufacturing task queue respectively to obtain a task preset start time queue; traversing the target manufacturing task queue and the target manufacturing process information queue to perform resource reliability analysis to obtain a resource reliability coefficient queue; traversing the target manufacturing task queue and the target manufacturing process information queue to perform equipment failure probability analysis, and taking the difference between the analysis result and 1 as the equipment reliability coefficient to obtain an equipment reliability coefficient queue; mapping the resource reliability coefficient queue and the equipment reliability coefficient queue to perform task reliability maximum likelihood estimation to obtain a task reliability evaluation value queue; performing task duration analysis based on the target manufacturing process information queue, the task reliability evaluation value queue and the task preset start time queue to obtain a task execution time tolerance threshold queue; performing conflict identification on the task execution time tolerance threshold queue, and performing task preset start time association adjustment according to the conflict identification result to obtain a task adjustment start time queue; and performing production management on the target manufacturing production line based on the task adjustment start time queue.

[0049] In a possible implementation, the digital management method based on intelligent manufacturing also performs the following processing: constructing a resource demand matrix according to the target manufacturing task queue and the target manufacturing process information queue, respectively, to obtain a resource demand matrix queue; obtaining a resource configuration list of the target manufacturing production line; performing a resource reliability analysis on the resource demand matrix queue based on the resource configuration list, to obtain the resource reliability coefficient queue.

[0050] In a possible implementation, the digital management method based on intelligent manufacturing also performs the following processing: pre-constructing a task reliability likelihood function; performing a one-to-one mapping between the resource reliability coefficient queue and the equipment reliability coefficient queue to obtain a mapped task reliability analysis result queue; using the maximum likelihood method to perform likelihood estimation on the mapped task reliability analysis result queue respectively through the task reliability likelihood function to obtain a task reliability maximum likelihood estimator queue; and using the task reliability maximum likelihood estimator queue as the task reliability evaluation value queue.

[0051] In a possible implementation, the digital management method based on intelligent manufacturing further performs the following processing: the task reliability likelihood function is: ; in, is the resource reliability coefficient, is the equipment reliability factor, It's a sample The estimated value of It's a sample The estimated value of is the sample value The corresponding probability value is is the number of subsamples, .

[0052] In a possible implementation, the digital management method based on intelligent manufacturing also performs the following processing: obtaining a sub-sample sequence queue based on the mapping task reliability analysis result queue; inputting the sub-sample sequence queues into the task reliability likelihood function for maximum likelihood estimation, and obtaining a task reliability maximum likelihood estimation queue.

[0053] In a possible implementation, the digital management method based on intelligent manufacturing also performs the following processing: pre-constructing a task execution time analyzer; using the task execution time analyzer to analyze the target manufacturing process information queue and the task reliability assessment value queue to obtain a task duration tolerance threshold queue; combining the task duration tolerance threshold queue and the task preset start time queue for time integration to obtain the task execution time tolerance threshold queue.

[0054] In a possible implementation, the digital management method based on intelligent manufacturing also performs the following processing: obtaining multiple sample target manufacturing process information, multiple sample task reliability assessment values ​​and multiple sample task duration tolerance thresholds as training samples; using the training samples to perform supervised learning on a framework constructed based on a feedforward neural network until the training converges, thereby obtaining the trained task execution time analyzer.

[0055] Although the present application makes various references to certain modules in the platform according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be implemented; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0056] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A digital management platform based on intelligent manufacturing, characterized in that: The platform includes: An information queue acquisition module is used to acquire a target manufacturing task queue of a target manufacturing production line, and to perform a manufacturing process of each target manufacturing task according to the target manufacturing task queue to acquire a target manufacturing process information queue; A start time queue acquisition module, used to extract the task preset start time of the target manufacturing task queue respectively, and obtain the task preset start time queue; A resource reliability analysis module, used for traversing the target manufacturing task queue and the target manufacturing process information queue to perform resource reliability analysis and obtain a resource reliability coefficient queue; An equipment failure probability analysis module, used to traverse the target manufacturing task queue and the target manufacturing process information queue to perform equipment failure probability analysis, and use the difference between the analysis result and 1 as the equipment reliability coefficient to obtain an equipment reliability coefficient queue; A reliability estimation module, used for performing a maximum likelihood estimation of task reliability on the resource reliability coefficient queue and the device reliability coefficient queue to obtain a task reliability evaluation value queue; A task duration analysis module, used to perform task duration analysis based on the target manufacturing process information queue, the task reliability evaluation value queue and the task preset start time queue, and obtain a task execution time tolerance threshold queue; A conflict identification module is used to identify conflicts in the task execution time tolerance threshold queue, and adjust the task preset start time according to the conflict identification result to obtain the task adjustment start time queue; A production management module is used to perform production management on the target manufacturing production line based on the task adjustment start time queue.

2. A digital management platform based on intelligent manufacturing as claimed in claim 1, characterized in that: The resource reliability analysis module includes: A resource requirement matrix construction unit, used to construct a resource requirement matrix according to the target manufacturing task queue and the target manufacturing process information queue, respectively, to obtain a resource requirement matrix queue; A resource configuration list acquisition module is used to acquire a resource configuration list of a target manufacturing production line; A resource reliability analysis unit is used to perform resource reliability analysis on the resource demand matrix queue based on the resource configuration list to obtain the resource reliability coefficient queue.

3. A digital management platform based on intelligent manufacturing as claimed in claim 2, characterized in that: The reliability estimation module comprises: A task reliable likelihood function pre-construction unit, used to pre-construct a task reliable likelihood function; A coefficient queue mapping unit, used for performing one-to-one mapping between the resource reliability coefficient queue and the device reliability coefficient queue to obtain a mapping task reliability analysis result queue; A likelihood estimation unit, used to use the maximum likelihood method to perform likelihood estimation on the mapping task reliability analysis result queues respectively through the task reliability likelihood function to obtain a task reliability maximum likelihood estimation amount queue; The task reliability evaluation value queue obtaining unit is used to use the task reliability maximum likelihood estimation value queue as the task reliability evaluation value queue.

4. A digital management platform based on intelligent manufacturing as claimed in claim 3, characterized in that: The task reliability likelihood function is: ； in, is the resource reliability coefficient, is the equipment reliability factor, It's a sample The estimated value of It's a sample The estimated value of is the sample value The corresponding probability value is is the number of subsamples, .

5. A digital management platform based on intelligent manufacturing as claimed in claim 3, characterized in that: The reliability estimation module comprises: A sub-sample sequence queue acquisition unit, used to acquire a sub-sample sequence queue according to the mapping task reliable analysis result queue; The maximum likelihood estimation unit is used to input the sub-sample sequence queues into the task reliability likelihood function to perform maximum likelihood estimation and obtain a task reliability maximum likelihood estimation amount queue.

6. A digital management platform based on intelligent manufacturing as claimed in claim 5, characterized in that: The task duration analysis module includes: A task execution time analyzer pre-building unit, used to pre-build a task execution time analyzer; A task duration tolerance threshold queue obtaining unit is used to analyze the target manufacturing process information queue and the task reliability evaluation value queue using the task execution time analyzer to obtain a task duration tolerance threshold queue; The time integration unit is used to combine the task duration tolerance threshold queue and the task preset start time queue to perform time integration to obtain the task execution time tolerance threshold queue.

7. A digital management platform based on intelligent manufacturing as claimed in claim 6, characterized in that: The task duration analysis module includes: A training sample acquisition unit, used to acquire a plurality of sample target manufacturing process information, a plurality of sample task reliability evaluation values ​​and a plurality of sample task duration tolerance thresholds as training samples; The task execution time analyzer obtaining unit is used to use the training samples to perform supervised learning on a framework constructed based on a feedforward neural network until the training reaches convergence, thereby obtaining the task execution time analyzer that has been trained.

8. A digital management method based on intelligent manufacturing, the method being applied to a digital management platform based on intelligent manufacturing as claimed in any one of claims 1 to 7, the method comprising: Acquire a target manufacturing task queue of a target manufacturing production line, and perform a manufacturing process of each target manufacturing task according to the target manufacturing task queue to obtain a target manufacturing process information queue; Extract the task preset start time of the target manufacturing task queue respectively to obtain the task preset start time queue; Traversing the target manufacturing task queue and the target manufacturing process information queue to perform resource reliability analysis and obtain a resource reliability coefficient queue; Traversing the target manufacturing task queue and the target manufacturing process information queue to perform equipment failure probability analysis, and taking the difference between the analysis result and 1 as the equipment reliability coefficient to obtain an equipment reliability coefficient queue; Performing a mapping task reliability maximum likelihood estimation on the resource reliability coefficient queue and the device reliability coefficient queue to obtain a task reliability evaluation value queue; Perform task duration analysis based on the target manufacturing process information queue, the task reliability evaluation value queue and the task preset start time queue to obtain a task execution time tolerance threshold queue; Conflict identification is performed on the task execution time tolerance threshold queue, and task preset start time association adjustment is performed according to the conflict identification result to obtain a task adjustment start time queue; Production management is performed on the target manufacturing production line based on the task adjustment start time queue.

Images