Production scheduling method, device and equipment based on digital twinning and medium

Abstract

The invention discloses a production scheduling method and device based on digital twinning, equipment and a medium, and relates to the field of intelligent manufacturing, and the method comprises the steps: collecting production data, generating an initial production state, defining an initial production scheduling constraint condition, and constructing a digital twinning model; predicting the available capacity of the production resources through a digital twinborn model, identifying a risk period, and updating the twinborn state of the production resources; coupling the predicted production resource available capacity with the order set, generating a task priority queue and updating production scheduling constraint conditions; executing scheduling optimization and simulation detection conflicts in the digital twin model, if conflicts exist, executing local rearrangement based on conflict information until requirements are met, and generating an executable scheduling scheme; and issuing the executable scheduling scheme to a production and logistics link, monitoring execution deviation, rescheduling an affected task, completing a production task and updating the digital twin database. According to the invention, the real-time performance and accuracy of scheduling response are improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing, and in particular to a production scheduling method, device, equipment and medium based on digital twins. Background Art

[0002] In the manufacturing process, production scheduling is a key link in achieving optimal resource allocation and on-time order delivery. Traditional production scheduling methods are mostly based on rule-driven, static models or historical experience-based scheduling decisions, and preliminary task allocation and scheduling control are achieved through the production scheduling module in the manufacturing execution system. This type of method usually uses fixed parameters and constraints. Based on known resource status and order requirements, it relies on heuristics, genetic algorithms, and constraint satisfaction to generate scheduling plans, and completes the scheduling and execution loop with the support of production management systems, equipment control systems, and information collection systems. In addition, in recent years, some studies have attempted to introduce real-time perception data into static models, and achieve response adjustments to abnormal conditions through feedback control to improve the robustness and effectiveness of the scheduling system.

[0003] However, in a volatile manufacturing environment, traditional production scheduling methods still face challenges in terms of dynamic adaptability and multi-source data integration. For one thing, quantitative forecasts of future short-term available capacity and explicit identification of risk periods are limited, making it difficult to proactively avoid bottlenecks before production scheduling. Furthermore, constraints and priorities are often static, making it difficult to write back conflicting information and drive local rescheduling and rescheduling within the same round, resulting in policy updates lagging behind the evolution of physical processes. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a production scheduling method based on digital twins to solve the problems of insufficient forward-looking identification of production capacity fluctuations and insufficient dynamic adaptation of scheduling constraints to real-time status in the prior art.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a production scheduling method based on digital twins, which comprises: Collect production data, generate initial production status, define initial production scheduling constraints, and build a digital twin model; Use digital twin models to predict the available capacity of production resources, identify risk periods, and update the twin status of production resources; The predicted available production capacity of production resources is coupled with the order set issued by the production management link, the comprehensive order priority is calculated and a task priority queue is generated, the risk window is solidified and the initial production scheduling constraints are updated; Based on the task priority queue and the updated production scheduling constraints, scheduling optimization is performed in the digital twin model to generate candidate scheduling solutions. Timing simulation of the candidate scheduling solutions is then performed in the digital twin environment to detect conflicts. Based on the detected conflicts, the conflict information is written back to the updated production scheduling constraints, and local rescheduling is performed within the same round until the conflicts are eliminated and the performance indicators meet the preset requirements, and an executable scheduling plan is output; The executable scheduling plan is issued to the production and logistics links, execution deviations are monitored, local working condition packages are generated, affected tasks are rescheduled, the overall production tasks are completed, production data is written back to the digital twin database, and the production resource available capacity forecast data, production scheduling constraints and optimization weights are updated.

[0007] As a preferred solution of the production scheduling method based on digital twins described in the present invention, wherein: The construction of the digital twin model refers to collecting production data and aligning them by timestamp to generate a production data set, generating an initial production state based on the production data set, defining initial production scheduling constraints, and building a digital twin model by combining virtual production line topology, resource nodes, processing nodes, and logistics paths.

[0008] As a preferred solution of the production scheduling method based on digital twins described in the present invention, wherein: The specific steps of predicting the available capacity of production resources, identifying risk periods, and updating the twin status of production resources are as follows: Based on the digital twin model, the available capacity of production resources is predicted and the available capacity prediction results of production resources are generated; Based on the prediction results of the available capacity of the production resources, the time period below the capacity threshold is identified as the risk period and the twin status of the production resources is updated.

[0009] As a preferred solution of the production scheduling method based on digital twins described in the present invention, the steps of coupling the predicted available production resource capacity with the order set issued by the production management link, calculating the comprehensive order priority and generating a task priority queue, solidifying the risk window and updating the initial production scheduling constraints are as follows: The available capacity forecast results of production resources are coupled with the order set issued by the production management link to generate the coupling results of production resources and order sets; Calculate the comprehensive order priority of each order and generate the comprehensive order priority of all orders; Sort by order priority from high to low to generate a task priority queue; Solidify the risk window and update the initial production scheduling constraints.

[0010] As a preferred solution of the production scheduling method based on digital twins described in the present invention, wherein: The execution scheduling optimization, generation of candidate scheduling solutions, and timing simulation to detect conflicts are specifically performed as follows: Execute scheduling optimization operations in the digital twin model to generate a scheduling optimization operation set; Perform timing simulation on the scheduling optimization operation set in the digital twin environment to generate simulation results of candidate scheduling solutions; Detect processing sequence conflicts, resource competition conflicts, and logistics congestion conflicts in the simulation results of candidate scheduling plans and generate conflict detection results.

[0011] As a preferred solution of the production scheduling method based on digital twins described in the present invention, wherein: The conflict information is written back to the updated production scheduling constraints, and local rearrangement is performed until the conflict is eliminated and the performance indicators meet the preset requirements, and an executable scheduling plan is output. The specific steps are: Based on the conflict detection results, the conflict type, conflict task identifier, conflict resource identifier, and occurrence time and location are extracted and summarized to generate conflict information; Writing back the conflict information to the updated production scheduling constraint condition to generate an updated production scheduling constraint condition result including the conflict constraint; Based on the updated production scheduling constraint condition results of the conflicting constraints, a local rearrangement operation is performed in the same round to generate a local rearrangement operation result; Based on the results of the local rearrangement operation, the timing simulation is performed again to cyclically determine whether the conflict is eliminated and whether the performance indicators meet the preset requirements. When the conditions are met, an executable scheduling plan is generated.

[0012] As a preferred solution of the production scheduling method based on digital twins described in the present invention, wherein: The executable scheduling plan is sent to the production and logistics links, execution deviations are monitored, local working condition packages are generated, affected tasks are rescheduled, the overall production tasks are completed, production data is written back to the digital twin database, and the production resource available capacity forecast data, production scheduling constraints and optimization weights are updated. The specific steps are as follows: Send the executable scheduling plan to the production and logistics links, monitor the execution deviation during the execution of the executable scheduling plan, generate the execution deviation results, and generate the local working condition package based on the execution deviation results; Execute the rescheduling operation and generate the rescheduling operation result; Complete the overall production task and generate the production task completion results; The production data is written back to the digital twin database, and the available capacity forecast data of production resources, production scheduling constraints and optimization weights are updated to generate an updated digital twin database.

[0013] In a second aspect, the present invention provides a production scheduling device based on digital twins, comprising: The data twin construction module is used to collect production data, generate the initial production state, define the initial production scheduling constraints, and build the digital twin model; The capacity forecasting and risk control module is used to predict the available capacity of production resources, identify risky periods, and update the twin status of production resources; The order priority queue module is used to couple the predicted available production resource capacity with the order set issued by the production management link, calculate the overall order priority and generate the task priority queue, solidify the risk window and update the initial production scheduling constraints; The optimization simulation conflict detection module performs scheduling optimization in the digital twin model based on the task priority queue and the updated production scheduling constraints, generates candidate scheduling solutions, and performs timing simulation on the candidate scheduling solutions in the digital twin environment to detect conflicts. The conflict write-back rescheduling module is used to write back the conflict information to the updated production scheduling constraints, perform local rescheduling within the same round until the conflict is resolved and the performance indicators meet the preset requirements, and output an executable scheduling plan; The execution monitoring and readjustment module is used to send executable scheduling plans to production and logistics links, monitor execution deviations, generate local working condition packages, reschedule affected tasks, complete overall production tasks, write production data back to the digital twin database, and update production resource available capacity forecast data, production scheduling constraints, and optimization weights.

[0014] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the production scheduling method based on digital twins as described in the first aspect of the present invention is implemented.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the production scheduling method based on digital twins as described in the first aspect of the present invention.

[0016] The present application has the beneficial effects that: by predicting the available capacity of production resources, identifying risk periods and updating the twin state of production resources based on the digital twin model, the forward-looking perception for future capacity fluctuations is established before production scheduling, potential risks are dynamically avoided during scheduling, and the continuity and reliability of resource allocation are ensured; by coupling the predicted available capacity of production resources and the order set issued by the production management link, calculating the order comprehensive priority and generating the task priority queue, solidifying the risk window and updating the initial production scheduling constraint conditions, the deep matching and sorting of order demand and resource capacity are realized, scheduling optimization and local rearrangement in the same round are carried out in the digital twin model, and through the iterative closed loop of "simulation-conflict detection-constraint write-back", the executable scheme is output, and the accuracy and timeliness of scheduling are improved. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0018] Figure 1 The flowchart of the production scheduling method based on digital twin.

[0019] Figure 2 The schematic diagram of the production scheduling system based on digital twin.

[0020] Figure 3 The flowchart of digital twin prediction and task priority queue generation.

[0021] Figure 4 The flowchart of scheduling optimization-simulation-conflict resolution and rescheduling. DETAILED DESCRIPTION

[0022] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification.

[0023] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the connotation of the present application, therefore the present application is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0025] Example 1, with reference to Figures 1 to 4 , is an embodiment of the present invention, which provides a production scheduling method based on digital twins, including the following steps: S1. Collect production data, generate the initial production status, define the initial production scheduling constraints, and build a digital twin model.

[0026] Collect production data and generate production data sets.

[0027] Furthermore, through the integrated call of equipment operation interface, distributed sensor collection device, manufacturing execution management information source and logistics status collection device, production data can be comprehensively collected.

[0028] Production data includes production equipment operating parameters, material work-in-progress status, inventory levels, logistics node status, and energy consumption data.

[0029] Furthermore, the collected production data is aligned according to timestamps to generate a production data set.

[0030] Based on the production dataset, an initial production state is generated.

[0031] The initial production state includes the resource availability matrix, process completion ratio, inventory vector, and logistics in-transit quantity.

[0032] Furthermore, based on the production data set, the resource availability matrix, process completion ratio vector, inventory vector, and logistics in-transit vector are calculated.

[0033] Among them, resource availability is expressed as: ; in, Indicates the Class resources in Availability of the process, Indicates the Class resources are in The available time of the process, Indicates the Class resources are in The planned duration of each process.

[0034] Determine the rows and columns (resource × process) and statistical period of the resource availability matrix, and take the planned required time and available time for each "resource × process" combination; calculate and fill in the matrix by dividing the available time by the planned required time, and make corrections for 0, missing, out-of-bounds and risk window situations to generate a resource availability matrix.

[0035] It should be noted that the process completion ratio vector is obtained by calculating the real-time task execution status data collected from the equipment operation interface and the manufacturing execution management information source.

[0036] The inventory vector is obtained from the material in and out records recorded in the manufacturing execution management information source and the inventory data in the warehousing system.

[0037] The logistics in-transit quantity vector obtains the current quantity of unfinished transportation by comparing the material loading departure time with the estimated arrival time.

[0038] Furthermore, the initial production state is generated based on the resource availability matrix, combined with the process completion ratio vector, inventory vector and logistics in-transit vector.

[0039] Using the virtual production line topology as the framework, the resource availability matrix, process completion ratio vector, inventory vector, and logistics in-transit vector are aligned by node / process, with unified time windows and units, to generate the state vector of the physical production line at t=0.

[0040] Furthermore, based on the initial production status, initial production scheduling constraints are defined.

[0041] It should be noted that the initial production scheduling constraints include resource capacity limitations, process routes, equipment maintenance plans, raw material supply delays, and energy caps.

[0042] Based on the rated capacity of the equipment, the resource capacity cap is formed as a capacity constraint according to the equipment maintenance plan and historical efficiency. The operation sequence and resource matching relationship are determined according to the process route of the order. The equipment maintenance plan is solidified as the period of resource unavailability, and the raw material supply delay constraint is generated by combining inventory, logistics in transit, and procurement delivery cycle. The energy upper limit is set based on the energy threshold / contracted power capacity, and the data statistical period and metering caliber are unified to generate the initial production scheduling constraints.

[0043] It should be noted that the process route is obtained by extracting standard process documents from the manufacturing execution management information source and analyzing the order product structure table.

[0044] The sources of equipment maintenance plans include preset regular maintenance plans, estimated repair times extracted from abnormal maintenance records, and the next maintenance time window predicted by historical maintenance cycle statistical models.

[0045] The sources of raw material supply delays include procurement cycle data recorded in the ERP system, statistical averages of historical supplier delivery times, and material shortage warning information and replenishment cycles in the warehousing system.

[0046] The sources of energy caps include energy scheduling thresholds for each time period provided by the energy monitoring platform, factory contracted power capacity or seasonal power restriction policies, and historical energy consumption models to estimate equipment energy consumption distribution.

[0047] Based on the initial production scheduling constraints, a digital twin model is constructed by combining virtual production line topology, processing nodes, resource nodes, and logistics paths.

[0048] Furthermore, based on the initial production scheduling constraints, according to the process route and resource deployment information, the task dependencies and resource distribution are extracted to generate a virtual production line topology; combined with the effective status of the equipment maintenance plan, the resource node set is determined from the resource capacity constraint list; the order process is decomposed according to the process route, and the processing node set is constructed based on the accessibility conditions of the raw material supply delay; using the resource node connection in the virtual production line topology as the skeleton, combined with the raw material supply delay and energy upper limit, the transmission capacity and constraints are verified to determine the logistics path set; a digital twin model with one-to-one correspondence between the virtual production line and the physical production line is established.

[0049] It should be noted that the resource node set corresponds to physical equipment, warehousing, and logistics hubs, and is determined based on the list of all production and logistics resources listed in the resource capacity constraint and combined with the effective status screening results in the equipment maintenance plan.

[0050] Furthermore, through the data mapping function The physical production line at time The state vector of is mapped to the virtual production line state vector, which is expressed as: ; in, Indicates the physical production line at time The state vector of Indicates that the virtual production line is at time The state vector of Represents a moment on a unified timeline.

[0051] Physical production line at all times The state vector is constructed from data collected on-site, which is continuously collected as production is executed, forming a state vector sequence of the physical production line that evolves over time.

[0052] It should be noted that the mapping function The physical production line state vectors at the same time are time-aligned and standardized, resource / node identifiers are encoded, and aggregated according to the virtual production line topology to generate the virtual production line at time The state vector of .

[0053] Furthermore, bidirectional synchronization of physical and virtual states is achieved through data mapping functions, so that each element of the virtual production line has a clear semantic mapping in the virtual production line topology, resource nodes, processing nodes and logistics paths. Combined with the resource capacity constraints, equipment maintenance plans, etc. in the initial production scheduling constraints, the actual production status is mapped into a complete constraint state vector that can be used for simulation and scheduling optimization in the digital twin model.

[0054] S2. Use the digital twin model to predict the available capacity of production resources, identify risk periods, and update the twin status of production resources.

[0055] Based on the digital twin model, the available capacity of production resources is predicted by combining the historical operation data of production resources, maintenance plan data and logistics bottleneck data, and the available capacity prediction results of production resources are generated.

[0056] It should be noted that the units of historical operation data of production resources, maintenance plan data and logistics bottleneck data are unified as pieces / hour, and the time variable is measured in hours; all data are standardized before entering the digital twin model to ensure the consistency of the integral item dimensions.

[0057] Furthermore, in the established digital twin model, the production resource available capacity prediction formula is used to generate the predicted production resource available capacity based on the production resource historical operation data, maintenance plan data, and logistics bottleneck data. Furthermore, the predicted average production resource available capacity is expressed as: ; in, Indicates the resources in the time interval [t0, t0+Δt] The forecast average available capacity of production resources, Indicates time The theoretical maximum output, Represents a resource At the moment Loss of production capacity due to planned maintenance, Represents a resource Loss of production capacity due to logistics delays, represents the mean time between failures of the equipment, λ represents the attenuation coefficient of the failure impact, represents the length of the prediction time window, Indicates that the starting point of the prediction time window has a value range greater than 0. represents the integration variable, ∈[t0,t0+Δt].

[0058] It should be noted that the time The theoretical maximum yield is calculated by the resource capacity in the initial scheduling constraints, combined with the virtual production line topology and the rated capacity of resource nodes at different times.

[0059] Resource At time The capacity loss due to planned maintenance is estimated based on the equipment maintenance plan, historical maintenance records, and equipment operating parameters to estimate the loss window and amplitude caused by maintenance, subscript represents the capacity loss item corresponding to the planned maintenance.

[0060] Resource The capacity loss due to logistics delay is calculated by referring to the logistics node state, energy consumption data, and logistics in-transit vector, combined with the real-time congestion of each logistics path. represents the capacity loss item corresponding to the logistics delay. The average failure-free interval of the equipment is derived by combining the historical data of the equipment operation and the maintenance plan statistics. The failure influence decay coefficient is calibrated by minimizing the deviation between the actual available capacity in the historical period and the model prediction value, and is updated in a sliding time window.

[0061] It should be noted that the integral variable is the time point that changes in a small time interval [t0, t0+Δt] in the future, which is used to accumulate and weight the average production resource available capacity change in the future time window.

[0062] The capacity loss due to logistics delay is calculated by referring to the logistics node state, energy consumption data, and logistics in-transit vector, combined with the real-time congestion of each logistics path.

[0063] The predicted average production resource available capacity considers multiple variables such as resource capacity, equipment maintenance, logistics bottleneck, and resource reliability, realizes dynamic quantitative expression of the average production resource available capacity, and improves the scientificity and responsiveness of task allocation and scheduling plan.

[0064] Further, the integral operation of the average production resource available capacity prediction adopts a sliding window method, and the prediction result is output in steps of hours, forming an average production resource available capacity prediction result set.

[0065] It should be noted that the average production resource available capacity prediction result set is a set formed after dynamically estimating the average production resource available capacity of the future n prediction time windows.

[0066] Based on the average production resource available capacity prediction result, the time period below the capacity threshold is identified as the risk period and the twin state of the production resource is updated.

[0067] Furthermore, a capacity threshold is set. When the average available capacity of production resources corresponding to a time node is less than the capacity threshold, the time interval where the time node is located is recorded as a risk period.

[0068] Among them, the capacity threshold is calculated by performing statistical analysis on the mean and standard deviation of the actual historical production resource available capacity data, combined with the set fault tolerance coefficient, to serve as a benchmark value for judging whether the capacity is in a risky state.

[0069] Based on the risk period, a risk period set is generated and the twin status of the production resources is updated.

[0070] Furthermore, the production resource status within the corresponding risk period is updated in the digital twin model to form a new set of production resource twin statuses.

[0071] In the digital twin model, status patches are written to the corresponding resource nodes at the granularity of "resource × risk period" to generate a new production resource twin status set. Each status patch contains at least a derated operation mark and a capacity limit mark. The generated new production resource twin status set can be regarded as a set of resource status records with validity periods (such as: resource ID, time window start and end, derated operation mark, capacity upper limit value, etc.), which are directly read and executed in subsequent production scheduling to limit the allocation and output of resources within the corresponding time window.

[0072] It should be noted that when generating a new production resource twin state, some production resource available capacity limit marker parameters for scheduling optimization are added to the original physical state. These parameters are derived based on the average production resource available capacity forecast results. The average production resource available capacity forecast result within the time window is calculated for each "resource × forecast time window"; a capacity threshold is then set based on historical data statistics to determine whether the time window falls within a risk period. If it does, the average production resource available capacity forecast result is used as the benchmark value for the maximum allowable available capacity of the resource within the time window. This value is then reduced by a preset tolerance factor or the smaller value is taken from the rated capacity to generate the production resource available capacity limit marker parameters. Simultaneously, the derated operation flag is set.

[0073] It should be noted that the derated operation mark means that the production resources are not running at full capacity during certain forecast periods, but can only work at a lower efficiency due to factors such as maintenance, failure or bottleneck.

[0074] The production resource available capacity limit mark indicates the maximum allowable output capacity of the production resource in a certain period of time, which is derived from the average production resource available capacity forecast result.

[0075] The twin state in the risk period not only reflects the current production resource available capacity limit, but also directly participates in the scheduling optimization calculation as a constraint condition in the subsequent scheduling stage, so as to realize the prediction-driven production scheduling strategy.

[0076] S3, coupling the predicted production resource available capacity with the order set issued by the production management link, calculating the order comprehensive priority and generating the task priority queue, solidifying the risk window and updating the initial production scheduling constraint condition.

[0077] Coupling the production resource available capacity prediction result with the order set issued by the production management link, generating the production resource and order set coupling result.

[0078] Further, each order in the order set issued by the production management link includes the delivery time Order quantity, customer level coefficient and required process resource path.

[0079] Mapping and matching the order set issued by the production management link with the average production resource available capacity prediction result set in the time dimension and the resource dimension, forming the production resource and order set coupling result structure, represented as: ; Among them, represents the set of production resource and order coupling results, represents the th order, represents the order corresponding production time demand set, represents traversing all production time points belonging to the order , represents the predicted average production resource available capacity of resource in the time interval [t1, t1+Δt].

[0080] Based on the coupling result of the average production resource available capacity prediction result and the order set, the order comprehensive priority of each order is calculated, and the order comprehensive priority of all orders is generated.

[0081] When generating the task priority queue, the cross-time-domain sliding window mechanism is adopted, so as to consider the current capacity prediction and also take into account the resource supply and demand matching situation in the future multiple periods.

[0082] Further, from the current time, a plurality of time windows are generated according to the preset window length and step, each window covering the production resource supply and demand situation in a future period of time.

[0083] In each time window, based on the average production resource available capacity prediction results of the digital twin model, the available capacity values ​​of all average production resource available capacities in the time window are counted to form a window-level average production resource available capacity set.

[0084] Furthermore, the order demand is apportioned within the relevant time windows according to the delivery deadline of the order, and the nominal demand for the order in different time windows is generated.

[0085] In each time window, the window-level average available production resource capacity and the nominal demand of the order are compared to obtain the resource supply and demand adequacy of the order placed in the time window.

[0086] It should be noted that if the average available production capacity of window-level production resources is insufficient, a supply-demand gap will be generated.

[0087] Furthermore, the supply and demand gaps of multiple time windows are weighted and accumulated to perform cross-time risk penalty. To highlight the importance of adjacent windows, attenuation weights are used to distinguish between near and far windows, expressed as: ; ; in, represents the cross-temporal risk penalty, Indicates the number of rolling time windows, represents the time window weight function, Display window The risk gap items, Indicates order In the time window The adequacy of resources, represents the time window index, Represents the decay coefficient, which controls the rate at which the weight decreases with the window index.

[0088] It should be noted that when the resource supply and demand adequacy is less than 1, the risk gap term is equal to the gap ratio. When the resource adequacy is greater than or equal to 1, the risk gap item is set to 0 and no penalty is calculated.

[0089] The decay coefficient determines the rate of decrease, with closer windows receiving greater weights and farther windows receiving smaller weights. It is set based on the comprehensive rolling horizon strategy, the decrease in prediction confidence over time, window configuration, and business KPI preferences. The value range is greater than 0.

[0090] Based on the order delivery urgency and customer level, combined with the cross-time risk penalty, the overall order priority is calculated.

[0091] Further, the order delivery urgency is calculated according to a unified standard and normalized, the customer level is mapped to a customer level coefficient, the cross-time domain risk penalty score is calculated by time window weighted accumulation, the delivery urgency and the customer level coefficient are combined into a basic score according to a configuration weight, and the cross-time domain risk penalty score is modified to generate an order comprehensive priority.

[0092] The task priority queue is generated by sorting all orders according to the order comprehensive priority from high to low.

[0093] Further, the task priority queue across time domains is generated by sorting all orders according to the order comprehensive priority from high to low.

[0094] It should be noted that the priority queue across time domains is recalculated at each time rolling window update, so that the scheduling strategy takes into account both short-term and long-term.

[0095] Based on the task priority queue and the risk window, the risk window is solidified and the initial production scheduling constraint condition is updated.

[0096] Further, based on the identified risk period set, the task priority queue under the current time round and the risk window are jointly mapped, it is analyzed whether there is risk coverage in the expected processing period of each task, and the risk information is written into the scheduling constraint structure, and finally the updated production scheduling constraint condition is formed.

[0097] S4, based on the task priority queue and the updated production scheduling constraint condition, scheduling optimization is performed in the digital twin model to generate a candidate scheduling scheme, and the candidate scheduling scheme is simulated in time sequence in the digital twin environment to detect conflicts.

[0098] Based on the task priority queue and the updated production scheduling constraint condition, scheduling optimization operation is performed in the digital twin model to generate a scheduling optimization operation set.

[0099] Further, the scheduling optimization algorithm is called in the digital twin model, taking the task priority queue and the updated production scheduling constraint condition as input, and making optimization decisions on task allocation, starting time and processing resources in the time dimension, resource dimension and path dimension, outputting multiple scheduling results, and the scheduling results are composed of multiple candidate scheduling schemes.

[0100] It should be noted that each candidate scheduling scheme includes task identification, time allocated resource number, task start time and task end time.

[0101] It should be noted that the multiple scheduling results outputted are scheduling schemes that meet the updated production scheduling constraint condition generated by inputting the task priority queue and the updated production scheduling constraint condition into the digital twin model through multiple iterations of the optimization algorithm.

[0102] It should be noted that the candidate scheduling scheme includes a task-resource-time three-dimensional mapping structure.

[0103] The task identifier is a unique identifier of a specific task in the scheduling process, and the task is uniquely indexed and bound with scheduling information in the scheduling scheme.

[0104] In the digital twin environment, a timing simulation is performed on the scheduling optimization operation set to generate a candidate scheduling scheme simulation result.

[0105] Further, the timing simulation process considers the precedence relationship between tasks, resource usage exclusivity, and logistics transmission time, performs scheduling operations on a virtual time axis, and outputs a candidate scheduling scheme simulation result set.

[0106] In the constructed digital twin model, a discrete event simulation is deployed to drive time advancement with an event queue; the initial production state is loaded, the updated production scheduling constraint conditions and task priority queue are assembled, and the average production resource available capacity prediction result at the current time is injected as a resource capability parameter within the simulation time window.

[0107] Each scheduling scheme in the scheduling optimization operation set is input into the simulation as a "dispatching and allocation strategy": when a task meets the prerequisite completion condition and enters the executable queue, the target resource allocation and logistics path are selected according to the process route, resource locking and occupation duration calculation are performed, start / end events are generated and written into the event queue.

[0108] Three types of conflicts are detected in real time at key hooks such as task start, resource application, and logistics departure according to rules: checking job precedence, identifying processing sequence conflicts, checking resource occupation mutual exclusion at the same time window, identifying resource competition conflicts, calculating time delay overruns based on path remaining capacity and in-transit queue, and identifying logistics blocking conflicts.

[0109] For triggered conflicts, record the conflict position, time node, involved task and resource identifiers, and conflict type, and write them into the simulation log of the current scheme. The simulation timeline, resource occupation record, logistics path delay, and task state marker are generated synchronously during the simulation advancement process. After the simulation is completed, the scheduling execution trajectory and conflict list of the candidate scheduling scheme are output, and the complete candidate scheduling scheme simulation result is generated. Repeat the process to obtain the simulation results of all candidate scheduling schemes, and aggregate them to form the candidate scheduling scheme simulation result set.

[0110] Detecting three types of conflicts in the candidate scheduling scheme simulation result includes processing sequence conflicts, resource competition conflicts, and logistics blocking conflicts.

[0111] It should be noted that processing sequence conflict means that the order of task execution is inconsistent with the process route; resource competition conflict means that the same resource is occupied by multiple tasks in the same time period; logistics congestion conflict means that the logistics path capacity is over-called or the path delay exceeds the upper limit.

[0112] Each conflict detection result includes the conflict type, the task ID involved in the conflict, the resource ID where the conflict occurs, and the time and location of the conflict.

[0113] During the simulation process, the execution time, resource occupancy status and logistics path of each task are tracked in real time to identify possible processing sequence conflicts, resource competition conflicts and logistics blockage conflicts. By traversing the simulation trajectory, the task identification, allocated resource number, conflict type and conflict time of each type of conflict are recorded to form the conflict information under a single plan. In this way, the same operation is performed on all candidate scheduling plans to finally construct a complete set of conflict detection results.

[0114] S5. Based on the detected conflicts, the conflict information is written back to the updated production scheduling constraints, and local rescheduling is performed within the same round until the conflicts are eliminated and the performance indicators meet the preset requirements, and an executable scheduling plan is output.

[0115] Based on the conflict detection result set, a conflict information set is generated.

[0116] The total conflict information set is composed of each individual conflict information data.

[0117] Furthermore, starting from each individual conflict information data, similar items are first merged and the field naming and time sequence are unified, and a conflict detection result set is generated. The task identifier, resource identifier, conflict time and conflict type are extracted one by one, and arranged into standard entries in a fixed order. Duplication, missing items and consistency checks are performed, and finally a conflict information set is output.

[0118] The conflict information data includes conflict task identifiers, conflict resource identifiers, conflict time, and conflict type.

[0119] The conflict information is written back to the updated production scheduling constraint condition to generate an updated scheduling constraint condition result including the conflict constraint.

[0120] Furthermore, the conflict information set is written back to the updated production scheduling constraint condition to form a newly added conflict constraint condition, thereby generating an updated production scheduling constraint condition including the conflict constraint.

[0121] The updated production scheduling constraints including the conflicting constraints are generated by combining the newly added conflicting constraints and the updated production scheduling constraints.

[0122] It should be noted that the newly added conflict constraints are used to avoid resource usage conflicts, processing sequence errors, and logistics congestion.

[0123] Furthermore, it is necessary to first perform structured extraction on the conflict task identifier, resource identifier, conflict time and conflict type included in the conflict detection results to generate new conflict constraints. The new conflict constraints clearly restrict certain resources from being occupied by conflicting tasks within a specific time period, prohibit the execution of tasks that violate the established processing sequence, or restrict the triggering of tasks under logistics congestion conditions, thereby realizing the regulation and avoidance of conflict scenarios; the updated production scheduling constraints and the new conflict constraints are merged to generate updated production scheduling constraints including conflict constraints.

[0124] Based on the updated production scheduling constraint condition results including the conflicting constraints, a local rearrangement operation is performed in the same round to generate a local rearrangement operation result.

[0125] Furthermore, based on the updated production scheduling constraints including conflicting constraints, a local reordering operation is performed in the same round, and a local search algorithm is used to reorder the affected task subsets to generate a set of local reordering solutions.

[0126] It should be noted that the local rescheduling plan set consists of multiple records, each of which includes a task identifier, a rescheduled resource identifier, a rescheduled task start time, and a rescheduled task end time.

[0127] Furthermore, based on the local rescheduling solution set and taking the original scheduling solution as a benchmark, conflicting task records are removed and replaced or supplemented with corresponding rescheduling records in the local rescheduling solution set to generate an updated scheduling solution.

[0128] Furthermore, the local rescheduling operation result set is used as the starting point, including the rescheduled resources and start and end times of the affected tasks. The old arrangements of the corresponding conflicting tasks are located and removed in the original scheduling plan. The vacant positions after removal are used as input, and the new arrangements in the local rescheduling operation result set are inserted and occupied in chronological order and resource numbers. The inserted temporary plan is used as input, and sequence verification, resource occupancy verification, and logistics verification are performed based on the updated production scheduling constraints including conflict constraints. If there is a violation, a new constraint is generated and the local search is returned to continue the rescheduling. The temporary plan that passes the verification is used as input, and the existing arrangements of the unaffected tasks are merged to form a scheduling plan after local rescheduling, and then enter the simulation verification.

[0129] Based on the results of the local rearrangement operation, the timing simulation is performed again to cyclically determine whether the conflict is eliminated and whether the performance indicators meet the preset requirements. When the conditions are met, an executable scheduling plan is generated.

[0130] Furthermore, the scheduling scheme after local rearrangement is simulated to detect conflicts and perform performance evaluation. If the conflict detection result is empty and the performance evaluation meets the preset performance indicator threshold, the scheduling scheme after local rearrangement is output as the final executable scheduling scheme.

[0131] Furthermore, using the partially rescheduled schedule as input, the execution process is reviewed in the digital twin environment, and the simulation timeline, resource utilization records, and logistics path delays are summarized to generate simulation results for performance evaluation. Comprehensive performance is calculated based on indicators such as delivery date achievement rate, total construction period, resource load balance, logistics achievement rate, and energy consumption to form a performance evaluation result. Each indicator is compared with the pre-set performance threshold to determine whether the partially rescheduled schedule is acceptable.

[0132] Based on the simulation timeline, the completion time of each order is determined and compared with the delivery deadline. The percentage of orders completed on time is calculated to generate the delivery achievement rate. The total duration is calculated as the span between the earliest start time and the latest completion time of all tasks. Based on resource occupancy records, the average utilization of each resource within the window is calculated. The fluctuation of each resource utilization rate (such as variance or coefficient of variation) is used to measure resource load balancing. The smaller the fluctuation, the higher the score. Based on logistics route delays and in-transit records, the percentage of logistics events that arrive on schedule or within the tolerance threshold is calculated to generate the logistics achievement rate. Energy consumption is calculated by accumulating the resource running / idle periods and the power consumption coefficient.

[0133] Indicators such as delivery date achievement rate, total construction period, resource load balance, logistics achievement rate, and energy consumption are normalized to the same standard: larger indicators (delivery date achievement rate, logistics achievement rate) are directly mapped to scores, while smaller indicators (total construction period, energy consumption, load imbalance) are reversely mapped or converted to scores based on thresholds. The configured indicator weights are weighted and synthesized to generate a comprehensive performance, which is then compared item by item with the preset performance indicator thresholds to determine whether the local rescheduling plan is approved.

[0134] If each indicator reaches the pre-set performance indicator threshold and the conflict detection is empty, the scheduling plan after local rearrangement is confirmed to be executable and output.

[0135] The pre-set performance indicator threshold can be determined by performing mean and standard deviation statistics on relevant historical data and combining it with the fault tolerance coefficient.

[0136] If the result of the simulation conflict detection of the scheduling scheme after local rearrangement does not meet the conflict detection result and is empty, or the performance evaluation result does not meet the preset performance indicator threshold, the simulation evaluation is performed in sequence, the conflict information set is generated, the scheduling constraints are written back and updated, the affected tasks are determined and locally rearranged, the scheduling scheme is updated and evaluated again, and the above steps are repeated until the conflict is eliminated and the performance indicator reaches the preset performance indicator threshold.

[0137] It should be noted that through dynamic feedback and iterative updating of production scheduling constraints based on conflict information, the adaptability and real-time response capability of scheduling results are improved, ensuring that key conflicts are avoided while maintaining scheduling efficiency, and achieving the stability and executability of the final scheduling plan.

[0138] S6. Send the executable scheduling plan to the production and logistics links, monitor the execution deviation, generate local working condition packages, reschedule the affected tasks, complete the overall production tasks, write the production data back to the digital twin database, and update the production resource available capacity forecast data, production scheduling constraints and optimization weights.

[0139] The executable scheduling plan is sent to the production and logistics links, the execution deviation during the execution of the executable scheduling plan is monitored, and the execution deviation results are generated.

[0140] Furthermore, the generated executable scheduling plan is sent to the actual production and logistics links, a monitoring cycle is set, feedback status data is periodically collected, and deviations are compared with the scheduling plan values ​​to form an execution deviation set.

[0141] The execution deviation set consists of multiple records, each of which includes the task where the deviation occurred, the deviation type, and the deviation time; the deviation time is the difference between the actual time and the planned time; the deviation type includes start delay, logistics interruption, and resource unavailability.

[0142] Based on the set of execution deviations, local working condition packages are constructed for all affected tasks.

[0143] It should be noted that the local working condition package consists of multiple records, each of which includes the impact scope of the task where the deviation occurs, the deviation type, and the deviation duration.

[0144] Furthermore, based on the execution deviation results fed back from the production or logistics process, the specific tasks where the deviation occurred are identified. A deviation propagation analysis model is invoked, inputting the current task's scheduling relationships, resource usage information, and dependency chains to dynamically calculate the potential impact of the deviation. Based on the collected deviation information, the deviation type (such as delay, resource unavailability, path blockage, etc.) and the deviation duration are recorded and assembled into a structure, forming a local working condition package instance. All deviation tasks are individually identified, propagated, analyzed, information recorded, and local working condition packages assembled, ultimately forming a collection of local working condition packages to support subsequent execution rescheduling.

[0145] It should be noted that the construction process of the deviation propagation analysis model includes determining the input and output. The input includes the scheduling relationship of tasks, resource usage information, material / BOM dependency chain, and deviation events observed in the digital twin model; the output is the risk assessment of the affected task set, resource overload time window and critical path / delivery node, and records the deviation type, deviation time, affected task identification, and impact scope.

[0146] Taking tasks as nodes, establish multiple types of edges according to process precedence and following relationships, shared resource relationships, and material / BOM chains; record the earliest / latest start and completion times of tasks on task nodes, and record resource capacity and occupancy information on the resource side to construct a task-resource multi-relationship graph.

[0147] The deviation events observed in the digital twin model are normalized into node equivalent disturbances (changes in duration / quality rework / resource occupancy), and the disturbed task nodes and time windows are located on the task-resource multi-relationship graph.

[0148] Three types of propagation cores are defined: the process predecessor and successor relationships update the earliest / latest time of the affected successor tasks; the shared resource relationship detects resource capacity overload and performs local rescheduling of conflicting tasks, so that the delay is transmitted between tasks with the same resources; the material / BOM chain transmits changes in the start time point to downstream tasks along the material dependency.

[0149] The disturbed task nodes are added to the event queue, and the updates of the three types of propagation cores are promoted in the topological direction. After each update, the newly affected tasks are re-added to the event queue. When, after one propagation and local resource rearrangement, the earliest / latest start and completion times of all tasks are consistent with the previous round, no new resource capacity overload time window is generated, no further local rearrangement is required, and the event queue is empty, the iteration is stopped.

[0150] Output the affected task identifiers and the start, completion, and margin changes of each task; output the impact scope, including the affected task set, the capacity overload time window of each resource, and the changes and worst-case delays of the critical path and delivery nodes; record the deviation type and deviation time; and write into the local working condition package.

[0151] The local working condition package set includes the affected task identifier, impact range, deviation type and deviation time.

[0152] Based on the local working condition package set, the rescheduling operation is performed to generate the rescheduling operation result.

[0153] Furthermore, based on the local working condition package set, the affected task set is extracted and expressed as: ; in, represents the set of affected tasks extracted from the local working condition package, represents a set of local operating condition packages, Express All records are combined and summed up. Indicates the specific task where the deviation occurs, subscript Indicates the index for performing a union operation on each record in the local working condition package set. represents the set of tasks affected by the deviation, Indicates the deviation type, Indicates the deviation duration.

[0154] Furthermore, in order to extract all tasks that need to be rescheduled, it is necessary to perform a collective merging operation on the tasks involved in all local working condition packages.

[0155] Specifically, the local working condition package set is traversed, the tasks where deviations occur and the impact range of the tasks where deviations occur are unioned, and all the union results are merged again to form a task set in a local working condition package, providing accurate task boundaries for the subsequent construction of local rescheduling constraints and rescheduling calculations.

[0156] Furthermore, local rescheduling constraints are constructed, and the rescheduling algorithm is executed in combination with the current remaining resource status to form a rescheduling plan.

[0157] The rescheduling plan consists of multiple records, each of which includes the rescheduled task, the new resource assigned to the task, the new start time of the task, and the new end time of the task.

[0158] Furthermore, local rescheduling constraints are first constructed based on the latest scheduling execution status, remaining production resources and current scheduling constraints; and a rescheduling algorithm is used to efficiently reschedule the tasks in the task set in the local working condition package.

[0159] It should be noted that the specific method of efficient rescheduling is "time window sliding + greedy insertion": for each task affected by the rescheduling, slide its expected time window, and give priority to finding insertion slots on the new resources allocated to the task in the rescheduling; on the premise of meeting the timing constraints and resource constraints, arrange the start time and end time of the tasks affected by the rescheduling as early as possible. Finally, the scheduling information of all successfully rescheduled tasks is uniformly recorded to form a complete set of rescheduling plans, thereby realizing dynamic scheduling adjustments without task interruption.

[0160] The rescheduling plan is merged into the remaining unaffected scheduling plan to form a complete update plan, which is expressed as: ; in, Represents the latest executable task scheduling scheme generated after rescheduling, Represents the new task scheduling set generated by rescheduling, Indicates the set of scheduling schemes currently being executed. Represents the set of affected tasks extracted from the local workload package.

[0161] Furthermore, based on the affected task set identified by the local operating condition package, the task scheduling records are first removed from the currently executing schedule to eliminate the disturbed unstable plan. Subsequently, the newly generated task schedule set is combined with the currently executing schedule after the removal, achieving a seamless replacement of the task schedules. During this merging process, it is necessary to ensure that the resource allocation and time schedules of each task in the newly generated task schedule set do not conflict with those of the remaining tasks in the currently executing schedule set, thereby constructing a complete update solution that is both consistent and executable.

[0162] Based on the rescheduling calculation results, the overall production task is completed and the production task completion result is generated.

[0163] Furthermore, the latest executable task scheduling plan generated after rescheduling is fully executed to the final state, and the task completion status, start and completion time, resource usage records, etc. are recorded to constitute the production task completion result.

[0164] Based on the completion results of the production tasks, the production data is written back to the digital twin database, and the production resource available capacity forecast data, production scheduling constraints and optimization weights are updated to generate an updated digital twin database.

[0165] Furthermore, the completion results of the production tasks are written back to the digital twin database to update the available capacity forecast data of production resources, which is expressed as: ; in, Represents a resource At the moment The updated historical occupancy record includes the latest task execution status. Represents a resource At the moment The original historical occupancy record, that is, the occupancy information set that existed before the update, Represents an occupancy record triple, indicating resources The actual time when the task starts executing Actual execution time The occupancy situation, Indicates a specific production resource identifier, such as a machine tool, a production line, or a logistics vehicle. The actual start time of the task The actual end time of the task execution.

[0166] Furthermore, the actual execution time period of each completed task is determined by the actual start time and the actual end time, forming a new resource occupancy record triple. The new resource occupancy record triple will be combined with the current resource at time The original historical occupancy record set is merged to form an updated resource occupancy history. The update process ensures the continuity and real-time nature of resource status information, providing more accurate basic data for subsequent production resource available capacity prediction, thereby improving the prediction model's responsiveness to resource bottleneck identification and scheduling strategy adjustment.

[0167] Furthermore, production scheduling constraints are updated, such as dynamic energy consumption constraints, logistics delays, resource maintenance cycles, etc.

[0168] The constraint condition set of this round and the production execution results of this round are comprehensively calculated to generate the scheduling constraint set of the next round.

[0169] It should be noted that the completion results of this round of production tasks include information such as task execution status, actual resource occupancy, and task start and end time.

[0170] The constraint set of this round is the final generated scheduling constraint set including optimization weights, conflict constraints and feasibility rules.

[0171] Furthermore, based on resource usage records during task execution, parameters such as resource maintenance cycles, available time windows, and energy consumption boundaries are updated to reflect the latest equipment health and energy consumption. In combination with logistics execution paths and completion timelines, the system dynamically adjusts the delay prediction rules and availability windows for logistics node status, strengthening the ability to control material flow bottlenecks.

[0172] If resource reconstruction or plan adjustment is detected during task execution, the process path and resource alternatives must be updated.

[0173] The task status, actual execution time, and conflict feedback involved in the completion results of the current round of production tasks are comprehensively applied. The set of production scheduling constraints of the previous round is mapped and transformed through functions to generate new production scheduling constraints including conflict avoidance, resource updating, energy consumption limitation, logistics adjustment, and optimization weights, providing constraint support for the next round of scheduling optimization.

[0174] In a preferred embodiment, the process of updating the optimization weights not only corrects a single optimization objective based on the execution deviation, but also adopts a dynamic weight adaptive evolution mechanism of the optimization objectives so that the weights of each optimization objective can evolve round by round with the scheduling performance.

[0175] Furthermore, after each round of scheduling is completed, multiple types of indicator data are collected to generate a scheduling performance set for the current round.

[0176] It should be noted that the multiple indicator data include delivery delay duration, overall construction period, energy consumption, logistics on-time rate and resource load balance.

[0177] Furthermore, the multi-category indicator data are normalized and standardized through sliding statistics to eliminate the dimensional differences between different indicators and obtain the relative deviation values ​​of the multi-category indicators.

[0178] It should be noted that the relative deviation values ​​of multiple indicators can reflect the fluctuation of each indicator relative to the historical average level.

[0179] A comprehensive driving signal is constructed based on the relative deviation values ​​of multiple indicators to generate a weight update vector.

[0180] Furthermore, a driving vector consisting of the relative deviation values ​​of multiple indicators is generated, and the weight update amount is calculated by optimizing the target weight and the driving vector, which is expressed as: ; in, Represents the optimization goal In the currently ongoing The weight update amount of the round, Indicates the The original target weight during round-robin scheduling, Indicates the Normalized index deviation of the optimization objective of the round, Represents the sensitivity coefficient.

[0181] It should be noted that the sensitivity coefficient controls the degree of influence of the deviation signal on the weight update amplitude. It is set according to the dynamics and stability of the production environment and the simulation scheduling results. The example value range is (0.05, 1). A value that is too small (for example, ) will cause the weight update to be almost unchanged, losing its adaptive meaning, and taking a value that is too large (e.g. ) will cause the weight update to be too drastic and the scheduling results to be unstable.

[0182] Furthermore, the weight updates of all optimization objectives are normalized, and a minimum weight threshold is set for each optimization objective to ensure that no optimization objective is completely ignored.

[0183] It should be noted that the final normalized weights satisfy the requirement that the sum is 1 and are not less than the preset minimum weight threshold, which can be expressed as: ; in, Indicates the During the round of iteration, The final normalized weight of the optimization objective, Indicates the During the round of iteration, The initial weight value of the optimization objective, Indicates in In the round iteration, the target in the target set is optimized The initial weight value of represents the target index of the optimization target set, Indicates the preset minimum weight threshold.

[0184] The new weight vector is used for the scheduling optimization objective function of the next round, so that the scheduling optimization can adaptively focus on the optimization objectives that performed poorly in the previous round. The scheduling optimization objectives can be continuously adjusted between different rounds, thus forming a continuously evolving dynamic weight mechanism.

[0185] This embodiment also provides a production scheduling device based on digital twins, including: The data twin construction module is used to collect production data, generate the initial production status, define the initial production scheduling constraints, and build a digital twin model.

[0186] The capacity forecasting and risk control module is used to predict the available capacity of production resources, identify risk periods, and update the twin status of production resources.

[0187] The order priority queue module is used to couple the predicted available production resource capacity with the order set issued by the production management link, calculate the comprehensive priority of the orders and generate the task priority queue, solidify the risk window and update the initial production scheduling constraints.

[0188] The optimization simulation conflict detection module performs scheduling optimization in the digital twin model based on the task priority queue and the updated production scheduling constraints, generates candidate scheduling schemes, and performs timing simulation on the candidate scheduling schemes in the digital twin environment to detect conflicts.

[0189] The conflict write-back rescheduling module is used to write back the conflict information to the updated production scheduling constraints, perform local rescheduling within the same round until the conflict is eliminated and the performance indicators meet the preset requirements, and output an executable scheduling plan.

[0190] The execution monitoring and readjustment module is used to send executable scheduling plans to production and logistics links, monitor execution deviations, generate local working condition packages, reschedule affected tasks, complete overall production tasks, write production data back to the digital twin database, and update production resource available capacity forecast data, production scheduling constraints, and optimization weights.

[0191] The embodiment also provides a computer device suitable for the production scheduling method based on digital twinning, including a memory and a processor, the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to realize the production scheduling method based on digital twinning proposed in the above embodiment.

[0192] The computer device can be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be achieved through WIFI, an operator network, NFC (Near Field Communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0193] The embodiment also provides a storage medium having a computer program stored thereon, the program is executed by a processor to realize the production scheduling method based on digital twinning proposed in the above embodiment. The storage medium can be realized by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a magnetic disk or an optical disk.

[0194] In summary, the present invention predicts the available capacity of production resources based on the digital twin model, identifies risk periods and updates the twin status of production resources, thereby establishing a forward-looking perception of future capacity fluctuations before scheduling, thereby dynamically avoiding potential risks during the scheduling process and ensuring the continuity and reliability of resource allocation; by coupling the predicted available capacity of production resources with the order set issued by the production management link, calculating the comprehensive priority of orders and generating a task priority queue, solidifying the risk window and updating the initial production scheduling constraints, it achieves deep matching and sorting of order requirements and resource capabilities, and then optimizes and rearranges the scheduling, thereby improving the accuracy and timeliness of scheduling.

[0195] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A production scheduling method based on digital twins, characterized by: include, Collect production data, generate initial production status, define initial production scheduling constraints, and build a digital twin model; Use digital twin models to predict the available capacity of production resources, identify risk periods, and update the twin status of production resources; The predicted available production capacity of production resources is coupled with the order set issued by the production management link, the comprehensive order priority is calculated and a task priority queue is generated, the risk window is solidified and the initial production scheduling constraints are updated; Based on the task priority queue and the updated production scheduling constraints, scheduling optimization is performed in the digital twin model to generate candidate scheduling solutions. Timing simulation of the candidate scheduling solutions is then performed in the digital twin environment to detect conflicts. Based on the detected conflicts, the conflict information is written back to the updated production scheduling constraints, and local rescheduling is performed within the same round until the conflicts are eliminated and the performance indicators meet the preset requirements, and an executable scheduling plan is output; The executable scheduling plan is issued to the production and logistics links, execution deviations are monitored, local working condition packages are generated, affected tasks are rescheduled, the overall production tasks are completed, production data is written back to the digital twin database, and the production resource available capacity forecast data, production scheduling constraints and optimization weights are updated.

2. The production scheduling method based on digital twin according to claim 1, characterized in that: The construction of the digital twin model refers to collecting production data and aligning them by timestamp to generate a production data set, generating an initial production state based on the production data set, defining initial production scheduling constraints, and building a digital twin model by combining virtual production line topology, resource nodes, processing nodes, and logistics paths.

3. The production scheduling method based on digital twin according to claim 2, characterized in that: The specific steps of predicting the available capacity of production resources, identifying risk periods, and updating the twin status of production resources are as follows: Based on the digital twin model, the available capacity of production resources is predicted and the available capacity prediction results of production resources are generated; Based on the prediction results of the available capacity of the production resources, the time period below the capacity threshold is identified as the risk period and the twin status of the production resources is updated.

4. The production scheduling method based on digital twin according to claim 3, characterized in that: The method couples the predicted available production capacity of production resources with the order set issued by the production management link, calculates the comprehensive priority of the orders and generates a task priority queue, solidifies the risk window and updates the initial production scheduling constraints. The specific steps are: The available capacity forecast results of production resources are coupled with the order set issued by the production management link to generate the coupling results of production resources and order sets; Calculate the comprehensive order priority of each order and generate the comprehensive order priority of all orders; Sort by order priority from high to low to generate a task priority queue; Solidify the risk window and update the initial production scheduling constraints.

5. The production scheduling method based on digital twin according to claim 4, characterized in that: The execution scheduling optimization, generation of candidate scheduling solutions, and timing simulation to detect conflicts are specifically performed as follows: Execute scheduling optimization operations in the digital twin model to generate a scheduling optimization operation set; Perform timing simulation on the scheduling optimization operation set in the digital twin environment to generate simulation results of candidate scheduling solutions; Detect processing sequence conflicts, resource competition conflicts, and logistics congestion conflicts in the simulation results of candidate scheduling plans and generate conflict detection results.

6. The production scheduling method based on digital twin according to claim 5, characterized in that: The conflict information is written back to the updated production scheduling constraints, and local rearrangement is performed until the conflict is eliminated and the performance indicators meet the preset requirements, and an executable scheduling plan is output. The specific steps are: Based on the conflict detection results, the conflict type, conflict task identifier, conflict resource identifier, and occurrence time and location are extracted and summarized to generate conflict information; Writing back the conflict information to the updated production scheduling constraint condition to generate an updated production scheduling constraint condition result including the conflict constraint; Based on the updated production scheduling constraint condition results of the conflicting constraints, a local rearrangement operation is performed in the same round to generate a local rearrangement operation result; Based on the results of the local rearrangement operation, the timing simulation is performed again to cyclically determine whether the conflict is eliminated and whether the performance indicators meet the preset requirements. When the conditions are met, an executable scheduling plan is generated.

7. The production scheduling method based on digital twin according to claim 6, characterized in that: The executable scheduling plan is sent to the production and logistics links, execution deviations are monitored, local working condition packages are generated, affected tasks are rescheduled, the overall production tasks are completed, production data is written back to the digital twin database, and the production resource available capacity forecast data, production scheduling constraints and optimization weights are updated. The specific steps are as follows: Send the executable scheduling plan to the production and logistics links, monitor the execution deviation during the execution of the executable scheduling plan, generate the execution deviation results, and generate the local working condition package based on the execution deviation results; Execute the rescheduling operation and generate the rescheduling operation result; Complete the overall production task and generate the production task completion results; The production data is written back to the digital twin database, and the available capacity forecast data of production resources, production scheduling constraints and optimization weights are updated to generate an updated digital twin database.

8. A production scheduling device based on digital twins, based on the production scheduling method based on digital twins according to any one of claims 1 to 7, characterized in that: include, The data twin construction module is used to collect production data, generate the initial production state, define the initial production scheduling constraints, and build the digital twin model; The capacity forecasting and risk control module is used to predict the available capacity of production resources, identify risky periods, and update the twin status of production resources; The order priority queue module is used to couple the predicted available production resource capacity with the order set issued by the production management link, calculate the overall order priority and generate the task priority queue, solidify the risk window and update the initial production scheduling constraints; The optimization simulation conflict detection module performs scheduling optimization in the digital twin model based on the task priority queue and the updated production scheduling constraints, generates candidate scheduling solutions, and performs timing simulation on the candidate scheduling solutions in the digital twin environment to detect conflicts. The conflict write-back rescheduling module is used to write back the conflict information to the updated production scheduling constraints, perform local rescheduling within the same round until the conflict is resolved and the performance indicators meet the preset requirements, and output an executable scheduling plan; The execution monitoring and readjustment module is used to send executable scheduling plans to production and logistics links, monitor execution deviations, generate local working condition packages, reschedule affected tasks, complete overall production tasks, write production data back to the digital twin database, and update production resource available capacity forecast data, production scheduling constraints, and optimization weights.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the production scheduling method based on digital twins described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the production scheduling method based on digital twins described in any one of claims 1 to 7 are implemented.

Images