An event-aware based mixing plant production rescheduling method and system

Abstract

The application provides an event-aware-based intensive mixing workshop production rescheduling method and system, relates to the technical field of intelligent manufacturing and production scheduling, and comprises the following steps: obtaining production scheduling basic data, a current scheduling scheme and a scheduling constraint set; monitoring order states and equipment running states in real time, identifying disturbance events, and determining a rescheduling triggering time, affected objects and workshop production states; determining locked processes and rescheduling processes according to process execution states, and determining equipment availability information; combining the above information to reconstruct the scheduling constraint set, constructing and solving a rolling rescheduling model, and obtaining a Pareto frontier solution set; calculating the comprehensive satisfaction of each scheme, determining a target rescheduling scheme, converting the target rescheduling scheme into production scheduling control instructions and updating a production execution queue. The application improves the rescheduling response speed of the intensive mixing workshop, the feasibility of the scheme and the comprehensive benefits of multiple targets.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing and production scheduling technology, and in particular to a method and system for production rescheduling in a mixing workshop based on event awareness. Background Technology

[0002] As tire and rubber product manufacturing moves towards multi-variety, small-batch, and flexible delivery, production scheduling in mixing workshops is gradually shifting from manual, experience-based scheduling to digital scheduling that combines Manufacturing Execution Systems (MES) with intelligent optimization algorithms. Existing scheduling technologies typically establish scheduling models based on data such as order delivery requirements, rubber compound processing technology, equipment processing capacity, and production costs. They then employ heuristic algorithms, swarm intelligence algorithms, or multi-objective optimization algorithms to determine the allocation relationships between processes and equipment, as well as process execution times, to improve order delivery efficiency and equipment utilization. With the development of equipment networking, production status data acquisition, and workshop edge computing technologies, dynamically adjusting existing scheduling schemes using real-time production data has become an important development direction for intelligent production management in mixing workshops.

[0003] However, existing internal mixing workshop scheduling methods still have significant shortcomings in dealing with dynamic disturbances. First, they lack a real-time response mechanism for continuous dynamic disturbances. In actual production, internal mixing workshops frequently face dynamic events such as urgent order insertions, expedited order changes, and sudden internal mixer malfunctions. Traditional scheduling schemes are mostly generated once based on static order data. Once a disturbance occurs, the original plan becomes invalid, and there is a lack of ability to quickly reconstruct the scheduling scheme, which can easily cause the entire line to stop and subsequent processes to be in disarray. Second, the limitations of single-objective or simple weighted optimization are prominent. Most existing methods take minimizing completion time or maximizing equipment utilization as the single optimization objective, failing to incorporate energy consumption costs on the manufacturing side and inventory costs on the logistics side into a unified optimization framework, making it difficult to balance "rushing production to ensure delivery" with "cost reduction and energy consumption control". Furthermore, existing rescheduling methods often rearrange all processes from scratch, ignoring the actual execution status of already started processes and the reality that corresponding production equipment is already occupied. The new scheme is disconnected from the actual execution status on site, resulting in poor executability. Finally, the decision-making of multi-objective solutions lacks engineering orientation. There are numerous solutions in the Pareto front, each with its own advantages and disadvantages. Existing methods lack a decision-making mechanism that automatically selects the optimal compromise solution from an engineering practical perspective, relying on human experience for judgment, which is inefficient and makes it difficult to guarantee the quality of the solution.

[0004] How to solve the above-mentioned technical problems is the challenge facing this invention. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an event-aware method and system for rescheduling production in a mixing plant, which improves the response speed, feasibility, and multi-objective comprehensive benefits of rescheduling.

[0006] The technical solution adopted by this invention to solve its technical problem is as follows: This invention provides an event-aware method for production rescheduling in a mixing workshop, comprising the following steps: Step S1: Obtain the basic production scheduling data, current scheduling scheme, and scheduling constraint set of the internal mixing workshop; the basic production scheduling data includes order delivery attributes, process processing requirements, equipment processing capacity, and cost parameters; the current scheduling scheme includes the allocation relationship between processes and equipment and the planned execution time of processes; the scheduling constraint set includes process constraints, time constraints, and resource constraints. Step S2: Monitor the order status and equipment operation status of the mixing workshop in real time. When a dynamic disturbance event is identified, determine the rescheduling trigger time and the affected objects, and obtain the workshop production status information at the rescheduling trigger time. The dynamic disturbance events include order insertion, order expediting, and equipment failure. The affected objects include affected orders, processes, and equipment. The workshop production status information includes process execution status and equipment operation status. Step S3: Based on the process execution status in the workshop production status information, divide the processes in the current scheduling plan and the processes corresponding to newly inserted orders into locked processes and processes to be rescheduled, and determine the equipment availability information; Step S4: Based on the rescheduling trigger time, locked process, process to be rescheduled, and equipment availability information, combined with the production scheduling basic data, reconstruct the scheduling constraint set to obtain the reconstructed scheduling constraint set, and construct a rolling rescheduling model. Step S5: Using a multi-objective optimization algorithm and combining it with the reconstructed scheduling constraint set, solve the rolling rescheduling model to obtain the Pareto front solution set; each solution in the Pareto front solution set corresponds to a candidate rescheduling scheme. Step S6: Calculate the overall satisfaction of each candidate rescheduling scheme, select the candidate rescheduling scheme with the highest overall satisfaction as the target rescheduling scheme, convert the target rescheduling scheme into a production scheduling control instruction and issue it to update the production execution queue of the mixing workshop.

[0007] Preferably, step S2 specifically includes: Dynamic disturbance events can be identified by monitoring the order status and equipment operating status in the mixing workshop; When the dynamic disturbance event is an order insertion, obtain the order delivery attributes and process requirements of the newly inserted order and synchronize them to the production scheduling basic data; when the dynamic disturbance event is an order expediting, update the order delivery attributes of the corresponding order and synchronize them to the production scheduling basic data; when the dynamic disturbance event is an equipment failure, obtain the failure information and estimated repair time of the failed equipment. The system time when a dynamic disturbance event is detected is determined as the rescheduling trigger time. Based on the type of dynamic disturbance event, combined with the allocation relationship between processes and equipment in the current scheduling scheme, the processing requirements of the processes, and the processing capacity of the equipment, the affected objects are determined. Using the rescheduling trigger time as the state capture point, the process execution record and equipment operation status data are time-aligned to obtain the process execution status and equipment operation status at the rescheduling trigger time.

[0008] Preferably, step S3 specifically includes: Based on the process execution status at the rescheduling trigger time, completed processes and processes that have started but not yet completed in the current scheduling plan are identified as locked processes, and the allocation relationship between the locked processes and equipment and the process execution status remain unchanged; processes that have not yet started in the current scheduling plan as of the rescheduling trigger time and all processes corresponding to newly inserted orders are identified as processes to be rescheduled; and the allocation relationship between the processes and equipment corresponding to the processes that were originally planned to be assigned to faulty equipment but have not yet started in the current scheduling plan is released. When the equipment is in a faulty state, the estimated repair completion time of the faulty equipment is determined based on the fault occurrence time and estimated repair time in the fault information. Obtain the actual completion time of the completed process and use the actual completion time of the completed process as the fixed process connection boundary of the corresponding subsequent process; obtain or determine the estimated completion time of the process that has started but not yet completed, determine the estimated completion time as the completion time of the corresponding locked process, and determine the time period between the rescheduling trigger time and the estimated completion time as the equipment occupancy time period of the corresponding locked process; Based on the equipment operating status at the time of rescheduling trigger, the equipment occupancy period of the locked process, and the estimated repair completion time of the faulty equipment, the equipment availability information is determined; the equipment availability information includes the equipment availability status, the earliest initial availability time, and the equipment unavailability period.

[0009] Preferably, the process constraints are as follows: when there is a preceding process, the subsequent process begins after the preceding process is completed and the corresponding process cooling time is met; the time constraints limit the earliest start time and completion time of the process; the resource constraints limit the equipment that the process can select and the corresponding available time range, and limit the execution time periods of different processes on the same equipment to not overlap. The reconstruction of the scheduling constraint set in step S4 includes: The process constraints are restructured. If the preceding process is a locked process, the process to be rescheduled will start after the locked process is completed and the corresponding process cooling time is met. The time constraints are restructured, and the earliest start time of the process to be rescheduled is updated to be no earlier than the rescheduling trigger time. The resource constraints are restructured, and the equipment that the process can select and the corresponding available time range are updated based on the equipment availability information.

[0010] Preferably, the construction of the rolling rescheduling model in step S4 specifically includes: Using the allocation relationship between the processes and equipment corresponding to the processes to be rescheduled, as well as the execution order, start time and completion time of the processes to be rescheduled, as the content to be determined, a rolling rescheduling model is constructed under the constraints of the reconstructed scheduling set, with the optimization objectives of minimizing delivery performance and minimizing overall operating costs. The cost parameters include processing energy consumption cost coefficient, standby energy consumption cost coefficient, and inventory cost coefficient; the delivery performance is determined based on the final completion time and corresponding order weight of each scheduled workpiece in a set of workpieces to be rescheduled, where at least one process belongs to the set of processes to be rescheduled; the comprehensive operating cost includes manufacturing-side energy consumption cost and logistics-side inventory cost; the manufacturing-side energy consumption cost is determined based on the total effective processing time and total idle standby time of each piece of equipment, combined with the processing energy consumption cost coefficient and the standby energy consumption cost coefficient; the logistics-side inventory cost is determined based on the relationship between the completion time and delivery date of each scheduled workpiece, as well as the relationship between the actual waiting time and process cooling time between adjacent processes, combined with the inventory cost coefficient.

[0011] Preferably, step S5 specifically includes: The one-dimensional real-number encoding dimension of each individual in the population is determined based on the number of processes to be rescheduled. An initial population is generated, and the correspondence between each encoding position and each process to be rescheduled is established. The encoding positions are sorted in ascending order of the real-number encoding values ​​in each individual to obtain the order of the processes to be rescheduled for each individual. The current process to be decoded is determined sequentially according to the arrangement order and the process sequence relationship between the processes to be rescheduled. A dynamic greedy device selection strategy is used to determine the production equipment and execution time corresponding to the current process to be decoded, until all processes to be rescheduled are decoded, and a candidate rescheduling scheme is obtained. The dynamic greedy device selection strategy determines the completion time of the current process to be decoded on each candidate device based on the availability time of the candidate devices, the completion time of the preceding process of the current process to be decoded, and the process cooling time. The candidate device with the earliest completion time is selected first. When the completion time is the same, the candidate device with the shortest processing time is selected. When the completion time and processing time are the same, the candidate device is selected according to the preset device sorting relationship. The availability time of the corresponding production device is updated after each process to be rescheduled is determined. Based on the delivery performance and overall operating cost of each candidate rescheduling scheme, non-dominated sorting and congestion evaluation are performed. Individuals in the population are selected according to the non-dominated level and congestion. The one-dimensional real number encoding of the individuals in the population is updated using a population update strategy. The encoding position sorting, individual decoding, non-dominated sorting and congestion evaluation are repeated for the updated individuals in the population until the iteration termination condition is met, and the Pareto front solution set is obtained.

[0012] Preferably, step S6 specifically includes: Determine the maximum and minimum values ​​of the objective function in the Pareto front solution set; Using a membership function based on fuzzy set theory, and based on the delivery performance and overall operating cost of each candidate rescheduling scheme, as well as the maximum and minimum values ​​of the optimization objectives, the delivery performance and overall operating cost are linearly normalized according to the rule that the smaller the objective value, the higher the satisfaction, to obtain the satisfaction membership value of each candidate rescheduling scheme on the two objectives. The satisfaction membership values ​​of each candidate rescheduling scheme on the two objectives are summarized and standardized to obtain the comprehensive satisfaction of each candidate rescheduling scheme. The candidate rescheduling scheme with the highest comprehensive satisfaction is determined as the target rescheduling scheme. The allocation relationship, execution order, start time and completion time of the processes and equipment corresponding to the processes to be rescheduled in the target rescheduling scheme are converted into production scheduling control instructions and issued. The production execution queue of the mixing workshop is updated according to the production scheduling control instructions, and the target rescheduling scheme is updated to the current scheduling scheme.

[0013] The present invention also provides an event-aware production rescheduling system for internal mixing workshops, comprising: The basic data acquisition module is used to acquire the basic production scheduling data, current scheduling scheme and scheduling constraint set of the mixing workshop; The event perception and status acquisition module is used to monitor the order status and equipment operation status of the mixing workshop in real time. When a dynamic disturbance event is detected, the rescheduling trigger time and the affected objects are determined, and the workshop production status information at the rescheduling trigger time is obtained. The process division and equipment availability information determination module is used to divide the processes in the current scheduling plan and the processes corresponding to newly inserted orders into locked processes and processes to be rescheduled, based on the process execution status in the workshop production status information, and to determine the equipment availability information. The constraint reconstruction and model building module is used to reconstruct the scheduling constraint set based on the rescheduling trigger time, locked process, process to be rescheduled, and equipment availability information, combined with the production scheduling basic data, to obtain the reconstructed scheduling constraint set and build a rolling rescheduling model. The multi-objective optimization solution module is used to solve the rolling rescheduling model by employing a multi-objective optimization algorithm and combining the reconstructed scheduling constraint set to obtain the Pareto front solution set. The comprehensive satisfaction decision and instruction issuance module is used to calculate the comprehensive satisfaction of each candidate rescheduling scheme, select the candidate rescheduling scheme with the highest comprehensive satisfaction as the target rescheduling scheme, convert the target rescheduling scheme into a production scheduling control instruction, issue it, and update the production execution queue of the mixing workshop.

[0014] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the above-described event-aware rescheduling method for rescheduling production in a mixing plant.

[0015] The present invention also provides a computer storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the above-described event-aware rescheduling method for rescheduling production in a mixing workshop.

[0016] The beneficial effects of this invention are: improving the response speed of rescheduling in the mixing workshop, the feasibility of the scheme, and the comprehensive benefits of multiple objectives. By monitoring order status and equipment operating status in real time, the invention enables event perception of order insertion, expedited orders, and equipment failures. When dynamic disturbance events are identified, the rescheduling trigger time and affected objects are determined, and the workshop production status information at the corresponding time is obtained. This allows the rescheduling process to respond promptly to changes in production conditions, providing a data foundation adapted to the on-site conditions for rapid adjustment of the scheduling scheme. By treating delivery performance and comprehensive operating cost as independent optimization objectives, and incorporating manufacturing-side energy consumption costs and logistics-side inventory costs into the comprehensive operating cost, the invention obtains the Pareto front solution set through a multi-objective optimization algorithm. This preserves different trade-offs between delivery performance and operating costs, avoiding evaluation biases caused by using a single production indicator or a simple fixed weighting method, and is conducive to balancing order delivery efficiency and workshop operating costs. Based on the process execution status at the rescheduling trigger time, locked processes and processes to be rescheduled are determined. Combined with equipment operating status, equipment availability information is determined. While retaining the existing process execution results and equipment occupancy status, process constraints, time constraints, and resource constraints are reconstructed. This ensures that subsequent optimization only targets processes that can be rescheduled, reducing the interference of rescheduling on already executed and currently executing processes, and improving the consistency, stability, and executability of the rescheduling scheme with the actual production status in the workshop. The overall satisfaction is calculated based on the delivery performance and comprehensive operating cost of each candidate rescheduling scheme in the Pareto front solution set, and the candidate rescheduling scheme with the highest overall satisfaction is determined as the target rescheduling scheme. This automatically determines the appropriate execution scheme from multiple candidate schemes with varying advantages and disadvantages, reducing reliance on manual experience-based decision-making. By converting the target rescheduling scheme into production scheduling control instructions and updating the production execution queue, the scheduling calculation results are transformed into the workshop production execution process, improving the decision-making and issuance efficiency of rescheduling schemes. Attached Figure Description

[0017] Figure 1 This is a diagram illustrating the method steps of the present invention.

[0018] Figure 2 This is a system module diagram of the present invention.

[0019] Figure 3 This is a diagram of the internal structure of a computer device according to Embodiment 3 of the present invention. Detailed Implementation

[0020] To clearly illustrate the technical features of this solution, the following detailed implementation method will be used to explain the solution.

[0021] Example 1: See Figure 1 As shown, this embodiment is an event-aware method for production rescheduling in a mixing workshop, including the following steps: Step S1: Obtain the basic production scheduling data, current scheduling scheme, and scheduling constraint set of the mixing workshop; the basic production scheduling data includes order delivery attributes, process processing requirements, equipment processing capacity, and cost parameters; the current scheduling scheme includes the allocation relationship between processes and equipment and the planned execution time of processes; the scheduling constraint set includes process constraints, time constraints, and resource constraints. Specifically, before entering the dynamic disturbance event monitoring process, the scheduling control device establishes data interfaces with the manufacturing execution system, process parameter database, equipment parameter database, and cost parameter database of the mixing workshop to obtain basic production scheduling data; it retrieves the currently executing scheduling scheme from the production execution queue or scheduling scheme database of the manufacturing execution system; and it obtains a pre-established set of scheduling constraints from the scheduling rule database. The scheduling control device is deployed on the workshop edge server or production scheduling server to associate, store, and retrieve the acquired data.

[0022] It should be noted that the basic production scheduling data is the overall data used to generate, evaluate, and adjust scheduling schemes within the current production task scope of the mixing workshop, and does not correspond one-to-one with individual candidate scheduling schemes. Different candidate scheduling schemes can be formed based on the same set of basic production scheduling data. The current scheduling scheme is a specific production scheduling scheme that has been issued and is being executed, formed based on the basic production scheduling data and the set of scheduling constraints.

[0023] Furthermore, in the scheduling model of this embodiment, the first... The production task corresponding to each rubber material order is abstracted as a scheduling workpiece. And establish scheduling workpieces through order identifiers. The correspondence between the order and the corresponding rubber compound order is established. Order delivery attributes include order identifier, order weight, order priority, and delivery date. The order identifier is used to associate the corresponding scheduled workpiece and its process data; the order priority characterizes the urgency of the order delivery and is used to determine the order weight; the order weight characterizes the importance of the corresponding order in the delivery performance evaluation; and the delivery date indicates the required delivery time for the corresponding order. When order expediting causes a change in order priority, the scheduling control device synchronously updates the order weight of the corresponding order according to the updated order priority.

[0024] Furthermore, the process requirements describe the processing procedure of the corresponding rubber compound for the workpiece, including process identification, process sequence, optional processing equipment, and cooling time between adjacent processes. Equipment processing capacity describes the processing tasks that each production piece of equipment in the mixing workshop can undertake, including equipment identification, the types of rubber compounds the equipment can process, and the types of processes the equipment can execute. The equipment identification includes equipment numbers used to uniquely identify and sort each production piece of equipment, with these equipment numbers having a preset numerical sorting relationship. Equipment processing capacity does not include the processing status, standby status, fault status, or maintenance status of the production equipment at the time of rescheduling trigger, nor does it include the time slots occupied by the production equipment due to executing assigned processes. The operating status of the production equipment at the time of rescheduling trigger is obtained in step S2, and the equipment availability status, initial earliest available time, and equipment unavailability period are determined in step S3.

[0025] For scheduling workpieces The The process is assigned to the first process. The processing time required when using the production equipment is denoted as: The processing time is a matching parameter determined by the processing requirements of the process and the processing capacity of the equipment. The scheduling and control device associates the sequence of processes, candidate equipment, processing time, and process cooling time with the corresponding workpiece based on the order identifier, process identifier, and equipment identifier, thereby determining the production equipment that can be selected for each process and its corresponding processing time.

[0026] Furthermore, cost parameters are used to quantify the overall operating costs corresponding to different scheduling schemes, including processing energy consumption cost coefficient, standby energy consumption cost coefficient, and inventory cost coefficient, denoted as follows: , and Among them, the processing energy consumption cost coefficient represents the processing energy consumption cost corresponding to the unit effective processing time of the production equipment; the standby energy consumption cost coefficient represents the standby energy consumption cost corresponding to the unit idling standby time of the production equipment; and the inventory cost coefficient represents the inventory cost corresponding to the unit inventory occupancy time of the rubber material. All cost parameters adopt a unified time measurement unit and cost measurement unit.

[0027] It should also be noted that the current scheduling scheme is a production scheduling scheme generated, issued, and executed by the mixing workshop before the occurrence of dynamic disturbance events. It includes the allocation relationship between processes and equipment, and the planned execution time of each process. Specifically, the allocation relationship between processes and equipment records the production equipment corresponding to each process and its planned processing sequence on that equipment; the planned execution time of each process includes the scheduled workpieces. The The planned start and finish times for each process are specified. The scheduling and control device uses order identifiers, process identifiers, and equipment identifiers as association indexes to associate equipment allocation results, planned processing sequence, planned start time, and planned finish time, forming a production execution queue that can be read separately for each production equipment.

[0028] When the internal mixing workshop starts production for the first time, the current scheduling scheme is the initial scheduling scheme generated and issued by the scheduling control device at the initial moment. When the internal mixing workshop has completed one or more reschedulings, the current scheduling scheme is the most recently determined and issued scheduling scheme that is still in the execution process at the current moment. This step is used to obtain the existing current scheduling scheme, and the rolling rescheduling model is not reconstructed in this step.

[0029] It should also be noted that the planned start time and planned completion time in the current scheduling scheme are planned data prior to the occurrence of dynamic disturbance events and cannot be directly used as the actual production status at the rescheduling trigger time. Whether each process is in a completed, started but not completed, or not started state at the rescheduling trigger time should be determined in step S2 based on the real-time collected process execution status.

[0030] The scheduling constraint set is a pre-established set of overall constraint rules based on the requirements of the internal mixing production process and the operating rules of the production equipment. It includes process constraints, time constraints, and resource constraints. Process constraints are used to limit the processing sequence between different processes in the same scheduled workpiece and the process cooling time required between adjacent processes. Time constraints are used to limit the start time of the process, the completion time of the process, and the arrangement relationship of each process on the scheduling time axis. Resource constraints are used to limit the matching relationship of processing capacity between processes and equipment, and to limit that only one process can be executed by the same production equipment at any given time.

[0031] Among them, for scheduling workpieces The In a process, the process cooling constraint between adjacent processes is represented as: in, Indicates scheduling workpieces The The completion time of each process. Indicates the first The required cooling time after each process step is completed. Indicates scheduling workpieces The The start time of the next process step. This constraint allows the first... The first process is in the The process begins after each step is completed and the specified cooling time has been reached.

[0032] Furthermore, resource constraints are used to limit each process to production equipment that meets the processing requirements of that process, and to limit different processes on the same production equipment from overlapping in time. Equipment failures and their maintenance periods are not included in the static resource constraints pre-set in this step. When step S2 identifies an equipment failure, the scheduling control device obtains the faulty equipment identifier and the estimated maintenance time. Step S3 determines the corresponding equipment availability information based on the operating status of the faulty equipment and the estimated maintenance time, and step S4 updates the corresponding resource constraints.

[0033] Step S2: Monitor the order status and equipment operation status of the mixing workshop in real time. When a dynamic disturbance event is identified, determine the rescheduling trigger time and the affected objects, and obtain the workshop production status information at the rescheduling trigger time. Dynamic disturbance events include order insertion, order expediting, and equipment failure. Affected objects include affected orders, processes, and equipment. Workshop production status information includes process execution status and equipment operation status. Specifically, during the execution of the current scheduling plan obtained in step S1, the scheduling control device maintains data connectivity with the manufacturing execution system, equipment controller, and equipment management system. It periodically collects the current status of orders and equipment, and receives order status change messages and equipment fault alarm messages. Order status includes at least order identifier, order placement status, order execution progress, order priority, and order delivery requirements. Equipment operating status includes at least equipment identifier, processing status, standby status, fault status, maintenance status, current processing step, and fault information. Based on the order identifier, process identifier, and equipment identifier, the scheduling control device correlates the real-time collected data with the production scheduling basic data and the current scheduling plan obtained in step S1.

[0034] It should be noted that order status and equipment operating status are provided by the Manufacturing Execution System (MES) and the Equipment Controller, respectively. The MES records order placement, order changes, and process start and completion information; the Equipment Controller collects operating signals, fault alarm signals, and current processing information from the production equipment. The scheduling control device continues to monitor order status and equipment operating status during rescheduling processes, and the data acquisition is not interrupted by the initiation of the rescheduling process.

[0035] Furthermore, the scheduling control device identifies dynamic disturbance events based on changes in order status and equipment operating status. Dynamic disturbance events include order insertion, order expediting, and equipment failure. Order insertion refers to a new order that has been placed in the mixing workshop but is not included in the current scheduling plan in the manufacturing execution system; order expediting refers to changes in the order weight, priority, or delivery date of existing orders in the current scheduling plan that tighten delivery requirements; equipment failure refers to the equipment controller outputting a fault alarm signal, or the equipment operating status changing from processing or standby to fault status, causing the corresponding equipment to be unable to continue performing processing tasks according to the current scheduling plan.

[0036] When an order insertion event is detected, the scheduling control device retrieves the corresponding order delivery attributes and process requirements from the manufacturing execution system and process parameter database based on the order identifier of the newly inserted order, and synchronizes the retrieved data to the production scheduling base data. When an order expedited event is detected, the scheduling control device retrieves the process arrangement of the corresponding order in the current scheduling plan based on the order identifier, and simultaneously retrieves the changed order priority, order weight, and delivery date. When expedited ordering causes a change in order priority, the scheduling control device synchronously updates the order weight of the corresponding order based on the updated order priority; if the expedited ordering directly provides an updated order weight, the updated order weight replaces the original order weight. The updated order weight is used in step S4 to calculate delivery performance, and the updated delivery date is used in step S4 to calculate logistics-side inventory costs. Therefore, the production scheduling base data called in step S4 includes data for newly inserted orders and orders with expedited changes.

[0037] When a equipment failure event is detected, the scheduling control device obtains the currently processing operation of the failed equipment based on the equipment identifier, as well as the operations that have been assigned to the equipment but have not yet started. Simultaneously, the scheduling control device obtains the corresponding equipment failure type and estimated repair time. The estimated repair time is determined by querying the corresponding maintenance parameters in the equipment management system based on the failure code output by the equipment controller; when equipment maintenance personnel enter the estimated recovery time, the estimated repair time is determined based on the failure occurrence time and the estimated recovery time. Step S2 only obtains the failed equipment and its estimated repair time; the determination of the unavailable period of the failed equipment and the updating of corresponding resource constraints are completed in steps S3 and S4, respectively.

[0038] Furthermore, when any dynamic disturbance event is detected, the scheduling control device reads the current system time and determines this time as the rescheduling trigger time. The rescheduling trigger time serves as the time reference between the historical execution part of the current scheduling scheme and the subsequent rescheduling processing part, and is used to subsequently determine the execution status of the process and update the time constraints of the process to be rescheduled.

[0039] Furthermore, the scheduling and control device determines the affected objects based on the type of dynamic disturbance event and the order, process, and equipment relationships established in step S1. The affected objects include the affected orders, processes, and equipment.

[0040] For order insertion events, the newly inserted order and its corresponding process are identified as the affected orders and processes, and the equipment that needs to be verified for availability is determined based on the matching relationship between the processing capacity of the process and the equipment. For order expedited events, the order that has undergone expedited changes and its incomplete processes are identified as the affected orders and processes, and the corresponding equipment is determined based on the equipment allocation relationship in the current scheduling plan. For equipment failure events, the failed equipment, the processes currently being processed by the failed equipment, and the processes that have been allocated to the equipment but have not yet started processing are identified as the affected equipment and processes, and the affected orders are determined based on the correspondence between the processes and the orders.

[0041] After determining the rescheduling trigger time, the scheduling control device uses the rescheduling trigger time as the state capture point, retrieves the process execution record from the manufacturing execution system, and obtains equipment status data from the equipment controller to form the workshop production status information at the rescheduling trigger time. The workshop production status information includes the process execution status and the equipment operating status.

[0042] Furthermore, the execution status of a process includes completed, started but not completed, and not started status. For processes that have generated an actual completion record by the time of rescheduling trigger, they are defined as completed; for processes that have generated an actual start record but not yet an actual completion record by the time of rescheduling trigger, they are defined as started but not completed; and for processes that have not generated an actual start record by the time of rescheduling trigger, they are defined as not started.

[0043] For processes that are in a completed state, obtain their actual processing equipment and actual completion time; for processes that are in a started but not completed state, obtain their actual processing equipment, actual start time, current processing progress, and estimated completion time; for processes that are not in a started state, record their execution status as where no actual start record has been generated. The planned processing equipment and planned execution time corresponding to processes that have not started are still retrieved from the current scheduling scheme obtained in step S1 and used as reference data for determining processes to be rescheduled in step S3.

[0044] Equipment operating status includes processing status, standby status, fault status, and maintenance status. For equipment in processing status, the current processing step and the estimated completion time of the corresponding step are obtained; for equipment in standby status, the time it entered standby status is obtained; for equipment in fault or maintenance status, the time of fault occurrence, fault type, and estimated maintenance duration are obtained. The above equipment operating status is used in step S3 to determine the availability range and availability time of each production equipment after the rescheduling trigger time.

[0045] It should also be noted that, to ensure that the process execution status and equipment operating status correspond to the same production time, the scheduling control device aligns the process execution records in the manufacturing execution system and the equipment status data in the equipment controller according to a unified time base. For data whose acquisition time is earlier than the rescheduling trigger time, the latest valid status record of the corresponding order, process, or equipment before the rescheduling trigger time is selected; for data whose status changes within the same acquisition cycle, the record with the status change timestamp is used.

[0046] Step S3: Based on the process execution status in the workshop production status information, divide the processes in the current scheduling plan and the processes corresponding to newly inserted orders into locked processes and processes to be rescheduled, and determine the equipment availability information; Specifically, the scheduling control device uses the rescheduling trigger time determined in step S2 as the state division benchmark, reads the current scheduling scheme obtained in step S1 and the process execution status, equipment operation status and affected objects obtained in step S2, and classifies the processes in the current scheduling scheme according to the actual execution status of each process at the rescheduling trigger time.

[0047] Furthermore, locked processes refer to processes that have already yielded actual results or are in progress, and whose equipment and execution time are not allowed to be rescheduled in this rescheduling process. This includes processes that have been completed as of the rescheduling trigger time and processes that have started but not yet been completed. Processes awaiting rescheduling refer to processes that have not yet started execution as of the rescheduling trigger time, but whose equipment allocation relationships and execution times can be re-determined in subsequent rescheduling processes.

[0048] For processes that have been completed by the time of rescheduling triggering, the scheduling control device classifies them as locked processes and records the actual processing equipment, actual start time, and actual completion time of the process. Completed processes are no longer the optimization decision objects of the subsequent rolling rescheduling model, but their actual completion time and the process connection relationship with subsequent processes serve as fixed boundary conditions for reconstructing process constraints and time constraints in step S4.

[0049] For production equipment that is in a faulty or under-maintenance state, the scheduling and control device determines the estimated maintenance completion time based on the fault occurrence time and the estimated maintenance duration.

[0050] For processes that have started but not yet completed by the rescheduling trigger time, the scheduling control device classifies them as locked processes, freezes the actual processing equipment and actual start time of the process, and maintains the allocation relationship between the process and the current processing equipment. For locked processes where the equipment is in normal processing status, the estimated completion time obtained in step S2 is used as the estimated completion time of the process. For processes that have started but not completed where the equipment is in a fault or maintenance status, if processing can be resumed after interruption in the corresponding rubber processing technology, the process is classified as a locked process and its original equipment allocation relationship is maintained. The scheduling control device determines the estimated remaining processing time based on the processing time of the corresponding process obtained in step S1 and the current processing progress obtained in step S2, and determines the estimated completion time of the locked process based on the estimated maintenance completion time and the estimated remaining processing time.

[0051] It should also be noted that freezing a process that has started but not yet been completed means not reselecting processing equipment during subsequent rescheduling, nor changing the actual start time and execution status of the process, rather than stopping the processing of that process. Although the locked process is not used as an optimization decision variable in the subsequent rolling rescheduling model, its expected completion time and equipment occupancy period are still used as fixed inputs for the time and resource constraints of step S4.

[0052] Furthermore, for processes that have not yet generated an actual start record by the time of rescheduling triggering, the scheduling control device classifies them as processes awaiting rescheduling. The original equipment allocation relationship, planned start time, and planned completion time of processes awaiting rescheduling in the current scheduling scheme are only used as historical scheduling references and are not conditions that must remain unchanged in subsequent rescheduling processes.

[0053] For order insertion events, since the process corresponding to the newly inserted order is not included in the current scheduling scheme obtained in step S1, the scheduling control device does not divide its historical execution status. Instead, based on the process processing requirements synchronized to the production scheduling basic data in step S2, it directly determines all processes corresponding to the newly inserted order as processes to be rescheduled and retains the process sequence relationship between each process.

[0054] For equipment failure events, completed processes and started but incomplete processes on the failed equipment are classified as locked processes according to their actual execution status. Processes originally planned to be assigned to the failed equipment but not yet started are considered unstarted processes in the current scheduling scheme, and are classified as processes awaiting rescheduling, and their original equipment allocation relationship with the failed equipment is terminated. Terminating the original equipment allocation relationship means allowing subsequent steps to reselect processing equipment based on equipment processing capacity and equipment availability information, and does not indicate that a new equipment allocation has been completed in step S3.

[0055] Furthermore, the scheduling control device aggregates all unstarted processes in the current scheduling plan, as well as the processes corresponding to newly inserted orders, to form a set of processes to be rescheduled, denoted as... Specifically, for unstarted processes originally planned to be assigned to faulty equipment, their original equipment allocation relationship is marked as terminated when they are added to the set of processes awaiting rescheduling; for other unstarted processes, their original equipment allocation relationship and planned execution time are only used as historical scheduling references. This avoids repeatedly adding unstarted processes on faulty equipment to the set of processes awaiting rescheduling as an independent category.

[0056] The set of processes to be rescheduled records the order identifier, process identifier, process sequence relationship, candidate equipment, processing time, and process cooling time for each process. This information is used in step S4 to determine the applicable process constraints, candidate equipment range, and scheduling time relationships for the processes to be rescheduled. This step only forms the set of processes to be rescheduled; it does not set the earliest start time for the processes or modify the scheduling constraint set.

[0057] After the process division is completed, the scheduling control device determines the equipment availability information of each production equipment after the rescheduling trigger time based on the equipment processing capacity obtained in step S1, the equipment operating status obtained in step S2, the equipment occupancy of locked processes, and the estimated repair time of faulty equipment. Equipment availability information includes equipment identifier, equipment availability status, initial earliest available time, and equipment unavailability period. The initial earliest available time indicates the time boundary at which the corresponding production equipment can initially undertake the rescheduled process before the current rolling rescheduling scheme begins execution.

[0058] For production equipment that is in standby mode and not occupied by a locked process at the rescheduling trigger time, it is identified as available equipment, and the rescheduling trigger time is determined as the initial earliest available time of the equipment.

[0059] For production equipment that is in the processing state and is performing a locking operation, it is identified as a delayed available equipment, and the estimated completion time of the corresponding locking operation is determined as the initial earliest available time of the equipment; the period between the rescheduling trigger time and the estimated completion time of the locking operation is marked as the equipment unavailable period.

[0060] For production equipment in a faulty or under-maintenance state, the period between the rescheduling trigger time and the aforementioned estimated maintenance completion time is marked as the equipment unavailable period. When the faulty equipment does not have any locking procedures that need to be completed, the estimated maintenance completion time is determined as the initial earliest available time for the equipment; when the faulty equipment has locking procedures that need to be completed after maintenance, the aforementioned estimated completion time of the locking procedures is taken as the end time of the corresponding equipment unavailable period, and this estimated completion time is determined as the initial earliest available time for the faulty equipment.

[0061] Step S4: Based on the rescheduling trigger time, locked process, process to be rescheduled, and equipment availability information, combined with the production scheduling basic data, reconstruct the scheduling constraint set to obtain the reconstructed scheduling constraint set, and construct a rolling rescheduling model. Specifically, the scheduling control device reads the rescheduling trigger time determined in step S2, as well as the locked processes, the set of processes to be rescheduled, and the equipment availability information output in step S3. It then calls the production scheduling basic data and scheduling constraint set obtained in step S1 and updated synchronously after identifying order disturbances in step S2. The scheduling control device uses the set of processes to be rescheduled as the optimization object for this rolling rescheduling, and uses the actual execution status, completion time, and equipment occupancy period of the locked processes as fixed boundaries. It updates the process objects, time boundaries, and available equipment range affected by the original scheduling constraint set to obtain the reconstructed scheduling constraint set.

[0062] Furthermore, regarding process constraints, the scheduling control device determines the process connection relationship between adjacent processes based on the sequence of processes in the set of processes to be rescheduled and the process cooling time. When two adjacent processes are both processes to be rescheduled, according to the sequence of processes and process cooling constraints given in step S1, the latter process is limited to start execution after the former process is completed and the corresponding process cooling time is met.

[0063] When the current process is a locked process and the next process is a process awaiting rescheduling, the scheduling control device uses the completion time of the previous locked process as the fixed process connection boundary for the next process awaiting rescheduling. For locked processes that have already been completed, the actual completion time obtained in step S2 is used; for locked processes that have started but not yet been completed, the estimated completion time determined in step S3 is used. This ensures that the next process awaiting rescheduling begins execution after the previous locked process has actually been completed or is expected to be completed and meets the corresponding process cooling time, instead of using the original planned completion time in the current scheduling scheme.

[0064] Furthermore, regarding time constraints, the scheduling control device uses the rescheduling trigger time as the starting point of the current rolling rescheduling time axis, updating the earliest start time of the processes to be rescheduled to be no earlier than the rescheduling trigger time. For any process in the set of processes to be rescheduled, its start time satisfies: in, Indicates scheduling workpieces The The start time of the process awaiting rescheduling in the candidate rescheduling scheme. The rescheduling trigger time is defined by the above time constraints, which limit the production scheduling range of processes to be rescheduled to after the rescheduling trigger time, without changing the production facts that have already been established before the rescheduling trigger time.

[0065] When a process awaiting rescheduling has a preceding locked process, the start time of the process awaiting rescheduling is also limited by the actual or expected completion time of the preceding locked process and the process cooling time. When the candidate equipment corresponding to the process awaiting rescheduling has an unavailable period, the execution time of the process awaiting rescheduling should also avoid the unavailable period of the corresponding equipment. Therefore, the actual start time of the process awaiting rescheduling is jointly limited by the rescheduling trigger time, the completion time of the preceding process, the corresponding process cooling time, and the earliest available time of the candidate equipment.

[0066] Furthermore, regarding resource constraints, the scheduling control device first determines the candidate equipment that can be selected for each process to be rescheduled based on the process requirements and equipment processing capacity obtained in step S1; then, based on the equipment availability information determined in step S3, it determines the time range within which each candidate equipment can undertake the corresponding process after the rescheduling trigger time. Each process to be rescheduled can only be assigned to one production equipment that simultaneously meets the processing capacity requirements and dynamic availability conditions, and the same production equipment can only execute one process at a time.

[0067] For equipment performing a locking operation at the rescheduling trigger time, the equipment occupancy period from the rescheduling trigger time to the expected completion time of the corresponding locking operation is set as an unallocable period. During this period, no rescheduling operation is allowed to be assigned to the corresponding equipment; after the expected completion of the locking operation, the corresponding equipment will participate in the allocation of subsequent operations according to the equipment availability information determined in step S3.

[0068] For equipment in a faulty or maintenance state, the fault maintenance period determined in step S3 is set as an unavailable resource block. Within the time range corresponding to the unavailable resource block, no rescheduling operations are allowed to be assigned to the faulty equipment. When the faulty equipment does not have any locking operations that need to be performed after maintenance, the estimated maintenance completion time is used as the earliest time for the equipment to participate in the rescheduling operation allocation; when the faulty equipment has any locking operations that need to be performed after maintenance, the estimated completion time of the locking operations is used as the earliest time for the corresponding equipment to participate in the rescheduling operation allocation.

[0069] For processes awaiting rescheduling that have had their original equipment allocation relationships removed in step S3, the scheduling control device re-determines the range of candidate equipment based on the processing requirements, equipment processing capacity, and equipment availability information of the corresponding process. For other processes awaiting rescheduling, their original equipment allocation relationships, planned execution order, and planned execution time in the current scheduling scheme are only used as historical scheduling references and are not constraints that must remain unchanged in this rolling rescheduling.

[0070] Furthermore, after obtaining the reconstructed scheduling constraint set, the scheduling control device uses the equipment allocation relationship, execution order, start time, and completion time of the processes to be rescheduled as the content to be determined. Under the condition of satisfying the reconstructed scheduling constraint set, it constructs a system to deliver performance. and overall operating costs To optimize the objective, a rolling rescheduling model is proposed. The objective function vector of the rolling rescheduling model is represented as:

[0071] Among them, delivery performance Used to characterize the completion time of each scheduling task involved in this rolling rescheduling; comprehensive operating cost Used to characterize the manufacturing-side energy consumption cost and logistics-side inventory cost corresponding to the candidate rescheduling scheme.

[0072] Furthermore, at least one process will belong to the set of processes to be rescheduled. The set of scheduling workpiece indices is denoted as For any Schedule the workpiece The final completion time in the candidate rescheduling scheme is denoted as . Suppose we schedule the workpiece. Include The next process involves scheduling the workpieces. The final completion time is the completion time of its last process in the candidate rescheduling scheme, that is: in, Indicates scheduling workpieces The final completion time in the candidate rescheduling scheme, superscript A fixed marker to represent the result of rescheduling; Indicates scheduling workpieces The completion time of the final process; Indicates scheduling workpieces The number of processes involved.

[0073] The delivery performance target is expressed as the weighted sum of the completion times of all relevant scheduled artifacts: in, Indicates scheduling workpieces The order weight corresponds to the order weight. The higher the order weight, the greater the proportion of the final completion time of the corresponding scheduled job in the delivery performance evaluation. Therefore, under the condition of satisfying the reconstructed scheduling constraint set, the multi-objective optimization algorithm is more inclined to shorten the final completion time of the scheduled job corresponding to the high-weight order.

[0074] It should be noted that the set of processes to be rescheduled is a set of process levels, while the set of workpiece indexes is a set of indexes at the workpiece level. The scheduling control device determines the set of workpiece indexes based on the correspondence between the processes to be rescheduled and their respective workpieces, and then determines the corresponding final completion time based on the completion time of the last process of each workpiece, thereby avoiding directly representing the workpieces as elements in the set of processes to be rescheduled.

[0075] Furthermore, overall operating costs Including energy consumption costs on the manufacturing side and logistics-side inventory costs Manufacturing-side energy costs consist of processing energy costs incurred by production equipment performing processing tasks and standby energy costs incurred by production equipment in idling or standby mode, expressed as: in, This indicates the total number of production equipment included in this rolling rescheduling. This represents the production equipment index. Indicates the first Total effective processing time of the production equipment. Indicates the first Total idle time of production equipment; and These represent the processing energy consumption cost coefficient and the standby energy consumption cost coefficient, respectively.

[0076] It should be noted that for each candidate rescheduling scheme, the scheduling control device determines the actual start time of the earliest process among the relevant scheduled tasks involved in this rolling rescheduling as the cost statistics starting point; when none of the relevant scheduled tasks have started execution before the rescheduling trigger time, the rescheduling trigger time is determined as the cost statistics starting point; and the latest final completion time of all relevant scheduled tasks involved in the candidate rescheduling scheme is determined as the cost statistics ending point. For each production equipment included in this rolling rescheduling, the total effective processing time and total idling standby time of the corresponding equipment are calculated according to the same cost statistics starting point and cost statistics ending point.

[0077] The effective processing time and idling standby time before the rescheduling trigger time are determined based on the actual execution data recorded by the manufacturing execution system; the effective processing time after the rescheduling trigger time is determined based on the remaining processing time of the production equipment continuing to execute the locked process and the processing time of executing the process to be rescheduled. The period during which the production equipment is available but does not perform processing tasks during the cost statistics period is included in the total idling standby time. The period of equipment failure maintenance and other equipment unavailability is not included in the total idling standby time.

[0078] It should also be noted that when calculating logistics-side inventory costs, for completed locked-in processes, the actual start time and actual completion time obtained in step S2 are used; for locked-in processes that have started but not yet completed, their actual start time and the estimated completion time determined in step S3 are used; for processes awaiting rescheduling, the start time and completion time determined by the corresponding candidate rescheduling scheme are used. Thus, logistics-side inventory costs simultaneously reflect the fixed inventory holding status formed before the rescheduling trigger time and the inventory holding status determined by the candidate rescheduling scheme after the rescheduling trigger time.

[0079] Logistics-side inventory costs characterize the additional waiting costs of rubber compounds exceeding the necessary cooling time between adjacent processes, as well as the finished goods inventory costs incurred due to the early completion of scheduled workpieces and the resulting wait for delivery. They are expressed as follows:

[0080] in, Indicates scheduling workpieces The corresponding order delivery date, Indicates scheduling workpieces The completion time of the final process, Indicates scheduling workpieces The number of processes included. Indicates the first The start time of each process, This indicates the completion time of the previous process. This indicates the required cooling time after the previous process is completed.

[0081] The first item in the aforementioned logistics-side inventory cost is used to calculate the waiting time for finished products to be delivered after the scheduled workpiece is completed before the delivery date. When the completion time of the last process is not earlier than the delivery date, this item is zero; when the completion time of the last process is earlier than the delivery date, the time difference between the delivery date and the completion time of the last process is taken as the finished product inventory holding time.

[0082] The second item in the aforementioned logistics-side inventory costs is used to calculate the additional waiting time between adjacent processes exceeding the process cooling time. When the... The first process is in the When the first process step is completed and the specified cooling time has been reached, this item is set to zero; when the second process step begins... When a process is delayed due to equipment occupancy or other production scheduling reasons, the portion exceeding the specified process cooling time shall be treated as additional inventory occupancy time between processes.

[0083] Processing costs, standby costs, and inter-process waiting costs incurred before the rescheduling trigger time are determined based on the corresponding actual execution data and are included as historical fixed cost items in all candidate rescheduling schemes. Manufacturing-side energy consumption costs and logistics-side inventory costs after the rescheduling trigger time are determined according to the corresponding candidate rescheduling scheme. The historical fixed cost items have the same value in all candidate rescheduling schemes and do not affect the relative evaluation results between different candidate rescheduling schemes.

[0084] Furthermore, the overall operating cost target is expressed as follows: Therefore, the rolling rescheduling model, under the constraints of the restructured scheduling set, simultaneously minimizes the delivery performance objective. and overall operating cost targets Reduce delivery performance targets This is aimed at shortening the final completion time of scheduled workpieces corresponding to high-weight orders and reducing overall operating costs. This is used to reduce manufacturing-side energy consumption costs caused by processing and idling of production equipment, and to reduce logistics-side inventory costs caused by rubber materials waiting for delivery between adjacent processes and after scheduling workpieces.

[0085] It should be noted that the delivery performance target With overall operating cost target As two independent optimization objectives, they are not merged into a single objective in step S4 using preset weights. Different candidate rescheduling schemes may have advantages in delivery performance or overall operating cost. Step S5 uses a multi-objective optimization algorithm to evaluate the candidate rescheduling schemes for non-dominance relationships, thereby obtaining a Pareto front solution set that reflects the trade-off between the two objectives.

[0086] Step S5: Using a multi-objective optimization algorithm and combining it with the reconstructed scheduling constraint set, solve the rolling rescheduling model to obtain the Pareto front solution set; each solution in the Pareto front solution set corresponds to a candidate rescheduling scheme. Specifically, the scheduling control device reads the data output in steps S3 and S4, and uses the equipment allocation relationship, execution order, start time and completion time of the process to be rescheduled as the content to be determined, and uses a multi-objective optimization algorithm to iteratively solve the problem.

[0087] It should be noted that in this embodiment, the population initialization, population partitioning, one-dimensional real position vector update, and position boundary processing within the continuous search space all follow the existing population search rules of the current continuous domain dung beetle optimization algorithm, without changing its population search operators. Based on the existing population search rules, this embodiment introduces a GAOE mechanism, a dynamic greedy decoding mechanism, a non-dominated sorting mechanism, a crowding evaluation mechanism, and an elite retention mechanism to form a multi-objective optimization algorithm for production rescheduling in the nitrifying workshop. The encoding mechanism used to map one-dimensional real position vectors to the order of processes to be rescheduled is denoted as Greedy Ascending Order Encoding (GAOE). The GAOE mechanism is used to convert the one-dimensional real position vectors obtained from population updates into the order of processes to be rescheduled. The dynamic greedy decoding mechanism is used to determine the production equipment and execution time corresponding to each process to be rescheduled based on the order of processes to be rescheduled, the process sequence, equipment processing capacity, and equipment availability information. The non-dominated sorting mechanism and the congestion evaluation mechanism are used to perform multi-objective comparisons based on the delivery performance and overall operating cost of candidate rescheduling schemes. The elite retention mechanism is used to retain the high-quality non-dominated solutions that have already been obtained.

[0088] Furthermore, the scheduling control device determines the encoding dimension of each population individual based on the number of processes to be rescheduled, representing each population individual as a one-dimensional real-valued position vector, and establishes a one-to-one correspondence between the position of each vector element and each process to be rescheduled. This correspondence remains unchanged during the rescheduling optimization process. The vector element positions are sorted in ascending order of value. When two or more vector element values ​​are the same, they are arranged according to the order of their corresponding positions. Based on the sorted vector element positions, the corresponding processes to be rescheduled are read sequentially to obtain the initial arrangement order of each process. This initial arrangement order serves as the priority basis for determining the decoding order of each process to be rescheduled, combined with the process sequence relationship. The equipment allocation result and process execution time are not written into the individual code as fixed content, but are dynamically determined during the decoding process based on the process requirements, equipment processing capacity, equipment availability information, and the reconstructed scheduling constraint set.

[0089] When order insertion causes a change in the number of processes to be rescheduled, the scheduling control device redetermines the one-dimensional real-valued position vector dimension of the population individuals during the current optimization process based on the updated set of processes to be rescheduled, and establishes a one-to-one correspondence between the position of each vector element and each process to be rescheduled. When equipment failure causes some inactive processes to have their original equipment allocation relationships removed, the vector element positions corresponding to the rescheduled processes are retained, and the candidate equipment range for the corresponding processes is re-determined during the decoding process based on the updated equipment availability information. Thus, the one-dimensional real-valued encoding dimension of the population individuals can be dynamically adjusted with changes in the set of processes to be rescheduled, instead of using a fixed encoding length and original equipment allocation relationships that are no longer applicable before the dynamic disturbance event.

[0090] Furthermore, the scheduling control device generates an initial population according to a preset population size. When generating the initial individuals, a one-to-one correspondence is established between the position of each vector element in the one-dimensional real-valued position vector and each rescheduling procedure, and the vector element positions are sorted in ascending order of vector element values ​​to obtain the initial arrangement order of each rescheduling procedure.

[0091] The scheduling control device performs priority-preserving topology adjustments to the initial arrangement order based on the process sequence relationships between the rescheduled processes. Specifically, from the rescheduled processes that have not yet been added to the decoding order, it selects optional processes that have no preceding processes, whose preceding processes are locked processes, or whose preceding rescheduled processes have all been added to the decoding order. When multiple optional processes exist, the optional process with the highest position in the initial arrangement is selected and added to the decoding order, and this process is removed from the rescheduled processes that have not yet been added to the decoding order. This process is repeated until all rescheduled processes are added to the decoding order. After the above processing, each initial individual corresponds to all processes in the set of rescheduled processes, each rescheduled process corresponds to only one vector element position, and the preceding process enters the dynamic greedy decoding process before its corresponding subsequent process.

[0092] It should also be noted that the initial permutation order obtained by the GAOE mechanism is used to represent the decoding priority of each process to be rescheduled. The scheduling control device performs a priority-preserving topology adjustment on the initial permutation order according to the process sequence, resulting in the actual decoding order used for dynamic greedy decoding. Different decoding orders may lead to different equipment occupancy states, sequential process connection times, and next available time for equipment, thus enabling different populations to form candidate rescheduling schemes with differences. By dynamically determining the equipment allocation relationship and process execution time in the decoding process, the encoding dimension of the population can be reduced, and the search difficulty of the algorithm in the discrete scheduling decision space can be lowered.

[0093] Furthermore, for each individual in the population, the scheduling control device obtains the initial arrangement order of the processes to be rescheduled according to the GAOE mechanism, and forms a decoding order that satisfies the process sequence relationship according to the aforementioned priority-preserving topology adjustment method; the current process to be decoded is determined sequentially according to the decoding order, and dynamic greedy decoding is performed. For the current process to be decoded, the scheduling control device first determines the candidate equipment that can perform the process based on the process processing requirements and equipment processing capabilities obtained in step S1; then, based on the equipment availability information determined in step S3, it identifies the earliest available time and unavailable period of each candidate equipment, and obtains the available candidate equipment and its available time range corresponding to the current process.

[0094] For each available candidate device, the scheduling control device determines the corresponding completion time of the preceding process based on the status of the preceding process: when the preceding process is a completed locking process, its actual completion time is used; when the preceding process is a locked process that has started but not yet completed, its estimated completion time is used; when the preceding process is a process that has been decoded and is awaiting rescheduling, the completion time of the preceding process that has been determined in the current candidate rescheduling scheme is used.

[0095] The scheduling and control device determines the feasible start time of the current process to be decoded on the corresponding candidate equipment based on the rescheduling trigger time, the completion time of the preceding process, the process cooling time between adjacent processes, the earliest initial availability time of the candidate equipment, the unavailability period of the candidate equipment, and the next available time of the equipment; and then determines the corresponding completion time based on the processing time of the current process to be decoded on the corresponding candidate equipment.

[0096] When there is no preceding process in the current decoding process, no lower limit for the start time is set based on the completion time of the preceding process and the process cooling time. The scheduling control device determines the corresponding feasible start time based on the rescheduling trigger time, the earliest initial available time of the candidate equipment, the unavailable period of the candidate equipment, and the next schedulable time of the equipment.

[0097] The scheduling and control device will trial-assign the current process to be decoded to each available candidate device. Based on the trial-assignment results, it will determine the feasible start and finish times for each process and prioritize selecting the candidate device with the earliest finish time. If two or more candidate devices have the same finish time, the candidate device with the shortest processing time will be selected. If both finish time and processing time are the same, the candidate device with the largest device number will be selected. The scheduling and control device will then designate the selected candidate device as the production equipment corresponding to the current process to be decoded and determine the start and finish times for that process.

[0098] After determining the equipment allocation result and execution time of the current process to be decoded, the scheduling control device updates the production sequence of the corresponding production equipment and the next available time of the equipment in the candidate rescheduling scheme, and determines the next process to be decoded according to the decoding order, and continues to execute dynamic greedy decoding until all processes in the set of processes to be rescheduled are decoded.

[0099] During the decoding process, when all candidate devices for a certain decoding step are temporarily unavailable, the scheduling control device, based on the earliest initial availability time of each candidate device determined in step S3 and the end time of the device's unavailability period, postpones the start time of the current step until at least one candidate device becomes available and meets the connection requirements of the preceding step. If there are no candidate devices that meet the processing capacity requirements of the corresponding step, or if the candidate devices become available but still cannot form a device allocation result that satisfies the reconstructed scheduling constraint set, the corresponding individual in the population is marked as an infeasible individual, and target evaluation and non-dominated sorting are not performed on that individual.

[0100] It should also be noted that the aforementioned priority-preserving topology adjustment applies to individuals in the initial population and individuals formed after population updates. The topology adjustment only alters the order in which rescheduled processes enter the dynamic greedy decoding process; it does not change the element values ​​of the one-dimensional real-valued position vector, nor does it change the correspondence between the vector element positions and the rescheduled processes. For multiple rescheduled processes that simultaneously meet the process sequence conditions, the decoding priority is determined according to their relative order in the initial GAOE arrangement.

[0101] When there are still rescheduled processes that have not yet been added to the decoding sequence, but there are no alternative processes that meet the conditions of the preceding processes, it indicates that there is a cycle or missing data in the corresponding process sequence. The scheduling control device will mark the corresponding individual as an infeasible individual. This processing is only used to resolve conflicts between the encoded mapping results and the process sequence, and is not used to handle situations where the production equipment does not have the corresponding processing capacity or the equipment cannot be restored to usability.

[0102] After decoding, the scheduling control device merges the locked process status determined in step S3 with the decoding results of the processes to be rescheduled to form a complete candidate rescheduling scheme. The candidate rescheduling scheme includes the matching relationship between each process to be rescheduled and the production equipment, the execution order of each process on the corresponding equipment, the start time and the completion time, and the fixed equipment occupancy period formed by the locked process.

[0103] Furthermore, the scheduling control device calculates the delivery performance of each candidate rescheduling scheme according to the rolling rescheduling model constructed in step S4. and overall operating costs Delivery performance is calculated based on the final completion time and order weight of each relevant scheduled task in the candidate rescheduling scheme; comprehensive operating cost is calculated based on the effective processing time of each production equipment, idle standby time, additional waiting time between rubber processing steps, and waiting time for delivery after the scheduled task is completed. Therefore, each candidate rescheduling scheme corresponds to a dual-objective evaluation result consisting of delivery performance and comprehensive operating cost.

[0104] Furthermore, the scheduling control device performs non-dominated sorting based on the bi-objective evaluation results of each candidate rescheduling scheme. For any two candidate rescheduling schemes... and When candidate rescheduling schemes It is no worse than candidate rescheduling solutions in terms of both delivery performance and overall operating cost. And it outperforms candidate rescheduling schemes in at least one objective. At that time, candidate rescheduling schemes are determined. Dominant candidate rescheduling scheme The determination criteria are expressed as follows: in, Indicates the sequence number of the optimization objective; when hour, Indicates delivery performance; when hour, This represents the total operating cost.

[0105] The scheduling control device compares each candidate rescheduling scheme in the current population pairwise. Schemes not dominated by other candidate rescheduling schemes are assigned to the first non-dominated layer. After temporarily removing schemes from the first non-dominated layer, schemes not dominated by other remaining schemes are assigned to the second non-dominated layer. This process is repeated until all candidate rescheduling schemes in the current population have obtained their corresponding non-dominated levels. The higher the non-dominated level, the better the candidate rescheduling scheme's dominance in terms of delivery performance and overall operating cost.

[0106] Furthermore, for candidate rescheduling schemes located within the same non-dominated layer, the scheduling control device calculates the congestion degree to evaluate the sparsity of the distribution of each candidate rescheduling scheme in the dual-objective space. For the first... One candidate rescheduling scheme Its congestion level is expressed as: in, and These respectively represent the order within the current non-dominated layer according to the first... After sorting the objective function values ​​of the optimization objectives, the adjacent candidate rescheduling schemes located on both sides of the candidate rescheduling scheme correspond to the th One objective function value; and These represent the current non-dominated layer. The maximum and minimum values ​​of the objective function; This represents the total number of optimization targets, in this embodiment. .

[0107] When calculating congestion, the scheduling control device sorts candidate rescheduling schemes within the same non-dominated layer according to delivery performance and overall operating cost, and accumulates the distances between adjacent candidate rescheduling schemes in the normalized target space. Candidate rescheduling schemes located at the current target boundary are prioritized for retention. When the maximum and minimum values ​​of a certain objective function within the current non-dominated layer are equal, the congestion component corresponding to that objective is set to zero. A higher congestion level indicates a sparser target space region where the corresponding candidate rescheduling scheme resides; therefore, candidate rescheduling schemes with higher congestion levels are prioritized for retention within the same non-dominated level.

[0108] Furthermore, the scheduling control device prioritizes individuals in the current population based on their non-dominance level and crowding level. For individuals with different non-dominance levels, those with higher non-dominance levels are designated as higher-priority individuals; for individuals with the same non-dominance level, those with higher crowding levels are designated as higher-priority individuals. Based on the priority comparison results, the scheduling control device determines the individuals used to guide the population search and the individuals requiring position updates, according to the population selection rules of the existing continuous domain dung beetle optimization algorithm.

[0109] The scheduling control device updates the one-dimensional real-valued position vectors of individuals in the continuous search space according to the population update rules of the existing continuous domain dung beetle optimization algorithm, and processes vector element values ​​that exceed the position value range according to the boundary handling rules of the existing algorithm. This embodiment does not change the population partitioning method, position update formula, and position boundary handling method of the existing algorithm. For the updated one-dimensional real-valued position vectors, the scheduling control device adopts the GAOE mechanism to sort the vector element positions in ascending order of vector element values, and obtains the initial arrangement order corresponding to the updated individuals based on the correspondence between each vector element position and each rescheduled process; the initial arrangement order is checked for process integrity, and a decoding order that satisfies the process sequence relationship is formed according to the aforementioned priority-preserving topology adjustment method; then, dynamic greedy decoding, target evaluation, non-dominated sorting, and crowding evaluation are re-executed.

[0110] Furthermore, the scheduling and control device merges the current population with the updated population, recalculates the delivery performance and overall operating cost of the merged population, and re-executes the non-dominated sorting and crowding calculation. Then, individuals are selected to enter the next generation population in descending order of non-dominated level; when the number of individuals in the same non-dominated level exceeds the remaining capacity of the next generation population, a corresponding number of individuals are selected according to their crowding degree from highest to lowest. Through this elite retention method, the currently obtained high-quality non-dominated solutions are preserved to the next generation, while maintaining the uniform distribution of the next generation population in the dual-objective space.

[0111] The scheduling control device iterates through the aforementioned GAOE mapping, dynamic greedy decoding, target evaluation, non-dominated sorting, crowding calculation, population update, and elite retention processes, updating the current population and its non-dominated level after each iteration. Iteration stops when the preset maximum number of iterations is reached. The preset population size and preset maximum number of iterations are set based on the number of processes to be rescheduled and the available computing resources of the scheduling control device.

[0112] Finally, the scheduling control device extracts all feasible solutions from the first non-dominated layer at the end of the iteration and merges duplicate solutions with the same equipment allocation relationship, process execution order, and process execution time to obtain the Pareto front solution set. Each non-dominated solution in the Pareto front solution set corresponds to a candidate rescheduling scheme, and records the delivery performance, overall operating cost, process-equipment matching relationship, process execution order, process start time, and process completion time of the candidate rescheduling scheme.

[0113] Step S6: Calculate the overall satisfaction of each candidate rescheduling scheme, select the candidate rescheduling scheme with the highest overall satisfaction as the target rescheduling scheme, convert the target rescheduling scheme into a production scheduling control instruction and issue it to update the production execution queue of the mixing workshop.

[0114] Specifically, the scheduling control device reads the Pareto front solution set output in step S5, as well as the candidate rescheduling schemes, delivery performance, and overall operating costs corresponding to each non-dominated solution in the Pareto front solution set. It performs satisfaction normalization processing on the delivery performance and overall operating costs of different dimensions, calculates the overall satisfaction of each candidate rescheduling scheme, and determines the target rescheduling scheme based on the overall satisfaction evaluation results.

[0115] Furthermore, suppose the Pareto front solution set output in step S5 includes The non-dominated solution will be the first... The candidate rescheduling schemes corresponding to the non-dominated solutions are denoted as follows: For candidate rescheduling schemes, place them in the first... The objective function value for each optimization objective is denoted as . , Among them, when hour, Indicates the delivery performance of candidate rescheduling schemes ;when hour, This represents the overall operating cost of the candidate rescheduling scheme. .

[0116] The scheduling control unit compares the delivery performance and overall operating cost of each candidate rescheduling scheme in the Pareto front solution set to determine the first one. The minimum objective function value corresponding to each optimization objective and the maximum objective function value The minimum objective function value represents the th element in the Pareto front solution set. The optimal value of the _i_i optimization objective is represented by the maximum objective function value, which indicates the _i ... The worst value of each optimization objective.

[0117] Since both delivery performance and overall operating cost are minimized objectives, the smaller the objective function value of the candidate rescheduling scheme on the corresponding objective, the higher its target satisfaction. The scheduling control device calculates the target rescheduling scheme's performance on the [number]th [period] based on the objective function values, minimum objective function value, and maximum objective function value of the candidate rescheduling scheme. The satisfaction membership value on each optimization objective is represented as follows: in, The value of is between zero and one. When the candidate rescheduling scheme is in the _____, When the objective function value of an optimization objective is equal to the minimum objective function value in the current Pareto front solution set, the corresponding satisfaction membership value is one; when its objective function value is equal to the maximum objective function value in the current Pareto front solution set, the corresponding satisfaction membership value is zero; when the objective function value is between the maximum and minimum objective function values, it is linearly normalized according to its relative position in the corresponding objective value interval.

[0118] It should be noted that when the candidate rescheduling schemes in the Pareto front solution set are at the , When the objective function values ​​for all optimization objectives are the same, the following condition is satisfied: In this case, each candidate rescheduling scheme is in the... There are no evaluation differences on the optimization objectives. The scheduling control device will uniformly determine the satisfaction membership value of each candidate rescheduling scheme on the optimization objective as one, thereby avoiding the situation where the denominator is zero in the satisfaction calculation process.

[0119] Since the satisfaction membership value has normalized delivery performance and overall operating cost to the same range, the scheduling control device summarizes the satisfaction membership values ​​of candidate rescheduling schemes on the two optimization objectives and standardizes them across all candidate rescheduling schemes to obtain the overall satisfaction of the candidate rescheduling schemes, which is expressed as: in, This represents the number of optimization objectives, in this embodiment. ; This indicates the number of non-dominated solutions in the Pareto front solution set; This represents the summation index in the denominator used to iterate through the candidate rescheduling schemes; Indicates the first The candidate rescheduling schemes are in the... Satisfaction membership value on each optimization objective; Indicates the first The standardized overall satisfaction of each candidate rescheduling scheme.

[0120] It should be noted that the overall satisfaction is used to rank the candidate rescheduling schemes in the Pareto front solution set, without changing the non-dominance relationship determined in step S5.

[0121] Furthermore, the scheduling control device compares the overall satisfaction of each candidate rescheduling scheme in the Pareto front solution set and determines the candidate rescheduling scheme with the highest overall satisfaction as the target rescheduling scheme. When multiple candidate rescheduling schemes have the same maximum standardized overall satisfaction, the scheduling control device selects one of them as the target rescheduling scheme according to the order of the candidate rescheduling scheme identifiers. The above parallel processing is only used to ensure the uniqueness of the output result of the target rescheduling scheme and does not change the evaluation relationship between delivery performance and overall operating cost in the satisfaction calculation.

[0122] It should also be noted that the target rescheduling scheme includes the locked process status determined in step S3 and the production scheduling results of the processes to be rescheduled determined in step S5. The equipment allocation relationship, actual execution status, and fixed equipment occupancy time of the locked processes remain unchanged; the equipment allocation relationship, execution order, start time, and completion time of the processes to be rescheduled adopt the production scheduling results already determined in the target rescheduling scheme.

[0123] Furthermore, the scheduling control device reads the equipment allocation results and process execution time recorded in the target rescheduling scheme, extracts the target equipment, processing sequence within the equipment, planned start time and planned completion time for each process to be rescheduled according to the order identifier, process identifier and equipment identifier, and arranges each process to be rescheduled according to the production equipment to form an updated production scheduling queue corresponding to each production equipment.

[0124] The scheduling control device converts the target rescheduling plan into production scheduling control instructions. These instructions include a rescheduling plan identifier, rescheduling trigger time, order identifier, process identifier, target equipment identifier, process execution sequence, planned start time, and planned completion time. For adjacent processes requiring material transfer, the production scheduling control instructions also include the equipment identifiers for the preceding and following processes and the planned transfer time, ensuring that the rubber material is transferred to the production equipment corresponding to the following process after the preceding process is completed and meets the process cooling requirements.

[0125] Furthermore, the production scheduling control instructions only update the pending rescheduling processes that have not yet been executed, without canceling the execution records of completed processes or changing the current processing equipment and execution status of processes that have started but not yet been completed. For the processes that were not started in the current production execution queue, the scheduling control device updates the production execution queue of the equipment to which the process belongs based on the equipment allocation relationship and execution time determined in the target rescheduling plan; for the processes corresponding to newly inserted orders, they are added to the production execution queue of the corresponding production equipment according to the equipment allocation relationship and execution order determined in the target rescheduling plan; for processes that were originally planned to be assigned to faulty equipment but whose original equipment allocation relationship was terminated in step S3, they are removed from the original production execution queue of the faulty equipment and added to the production execution queue of the production equipment determined in the target rescheduling plan.

[0126] For rescheduled processes assigned to fault recovery equipment, their planned start time in the target rescheduled scheme is no earlier than the earliest initial availability time of the corresponding equipment determined in step S3; for equipment currently performing a locking process, newly assigned rescheduled processes are arranged after the expected completion time of the locking process. The above scheduling results already satisfy the reconfigured scheduling constraint set obtained in step S4.

[0127] The scheduling control device issues production scheduling control commands through the data interface with the Manufacturing Execution System (MES). Upon receiving the production scheduling control commands, the MES writes the production scheduling information for each process into the production execution queue of the corresponding production equipment according to the equipment identifier, and updates the target rescheduling scheme to the current scheduling scheme. Each production equipment executes its corresponding process sequentially according to the updated production execution queue, and the equipment controller feeds back the process start status, process completion status, and equipment operating status to the MES.

[0128] It should also be noted that after the target rescheduling scheme is updated to the current scheduling scheme, this target rescheduling scheme becomes the current scheduling scheme being executed by the mixing workshop before the next dynamic disturbance event occurs. The scheduling control device continues to monitor the order status and equipment operating status in real time according to step S2. When an order insertion, order expediting, or equipment failure is detected again, the target rescheduling scheme that is currently being executed is used as the new current scheduling scheme, and steps S2 to S6 are re-executed, thereby forming a rolling rescheduling closed loop for continuous dynamic disturbance events.

[0129] Through the above processing, the scheduling control device can retain the execution results of the formed processes based on the actual production status at the time of rescheduling trigger after order insertion, order expediting, or equipment failure. It can also re-determine the equipment allocation relationship and execution time for processes that have not yet been executed. By incorporating the dynamic availability information of equipment, the completion time of the preceding process, and the process cooling time into the decoding process, it can avoid conflicts between the rescheduling process and the time period occupied by the locked process or the time period when the equipment is unavailable. By conducting multi-objective evaluation of delivery performance and overall operating costs, it can determine the target rescheduling scheme for updating the production execution queue from the feasible candidate rescheduling schemes.

[0130] Example 2: See Figure 2 As shown, this embodiment is an event-aware production rescheduling system for a mixing workshop, including: The basic data acquisition module is used to acquire the basic production scheduling data, current scheduling scheme and scheduling constraint set of the mixing workshop; The event perception and status acquisition module is used to monitor the order status and equipment operation status of the mixing workshop in real time. When a dynamic disturbance event is detected, the rescheduling trigger time and the affected objects are determined, and the workshop production status information at the rescheduling trigger time is obtained. The process division and equipment availability information determination module is used to divide the processes in the current scheduling plan and the processes corresponding to newly inserted orders into locked processes and processes to be rescheduled, based on the process execution status in the workshop production status information, and to determine the equipment availability information. The constraint reconstruction and model building module is used to reconstruct the scheduling constraint set based on the rescheduling trigger time, locked process, process to be rescheduled, and equipment availability information, combined with the production scheduling basic data, to obtain the reconstructed scheduling constraint set and build a rolling rescheduling model. The multi-objective optimization solution module is used to solve the rolling rescheduling model by employing a multi-objective optimization algorithm and combining the reconstructed scheduling constraint set to obtain the Pareto front solution set. The comprehensive satisfaction decision and instruction issuance module is used to calculate the comprehensive satisfaction of each candidate rescheduling scheme, select the candidate rescheduling scheme with the highest comprehensive satisfaction as the target rescheduling scheme, convert the target rescheduling scheme into a production scheduling control instruction, issue it, and update the production execution queue of the mixing workshop.

[0131] Example 3: This embodiment provides a computer device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0132] This computer device can be a server, and its internal structure diagram can be as follows: Figure 3 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores server data. The network interface communicates with external terminals via a network connection. When executed by the processor, the computer program implements an event-aware rescheduling method for production in a mixing plant.

[0133] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0134] Example 4: This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0135] If the functions implemented by the method are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art or the current technical solution, can be embodied in the form of a software product. This current computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0136] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0137] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0138] The technical features of this invention not described can be implemented by or using existing technology, and will not be repeated here. Of course, the above description is not a limitation of this invention, and this invention is not limited to the examples above. Any changes, modifications, additions or substitutions made by those skilled in the art within the scope of this invention should also be within the protection scope of this invention.

Claims

1. A method for rescheduling production in a mixing plant based on event awareness, characterized in that, Includes the following steps: Step S1: Obtain the basic production scheduling data, current scheduling scheme, and scheduling constraint set of the internal mixing workshop; the basic production scheduling data includes order delivery attributes, process processing requirements, equipment processing capacity, and cost parameters; the current scheduling scheme includes the allocation relationship between processes and equipment and the planned execution time of processes; the scheduling constraint set includes process constraints, time constraints, and resource constraints. Step S2: Monitor the order status and equipment operation status of the mixing workshop in real time. When a dynamic disturbance event is identified, determine the rescheduling trigger time and the affected objects, and obtain the workshop production status information at the rescheduling trigger time. The dynamic disturbance events include order insertion, order expediting, and equipment failure. The affected objects include affected orders, processes, and equipment. The workshop production status information includes process execution status and equipment operation status. Step S3: Based on the process execution status in the workshop production status information, divide the processes in the current scheduling plan and the processes corresponding to newly inserted orders into locked processes and processes to be rescheduled, and determine the equipment availability information; Step S4: Based on the rescheduling trigger time, locked process, process to be rescheduled, and equipment availability information, combined with the production scheduling basic data, reconstruct the scheduling constraint set to obtain the reconstructed scheduling constraint set, and construct a rolling rescheduling model. Step S5: Using a multi-objective optimization algorithm and combining it with the reconstructed scheduling constraint set, solve the rolling rescheduling model to obtain the Pareto front solution set; each solution in the Pareto front solution set corresponds to a candidate rescheduling scheme. Step S6: Calculate the overall satisfaction of each candidate rescheduling scheme, select the candidate rescheduling scheme with the highest overall satisfaction as the target rescheduling scheme, convert the target rescheduling scheme into a production scheduling control instruction and issue it to update the production execution queue of the mixing workshop.

2. The event-aware production rescheduling method for the internal mixing workshop according to claim 1, characterized in that, Step S2 specifically includes: By monitoring the order status and equipment operating status in the mixing workshop, dynamic disturbance events can be identified; When the dynamic disturbance event is an order insertion, obtain the order delivery attributes and process requirements of the newly inserted order and synchronize them to the production scheduling basic data; when the dynamic disturbance event is an order expediting, update the order delivery attributes of the corresponding order and synchronize them to the production scheduling basic data; when the dynamic disturbance event is an equipment failure, obtain the failure information and estimated repair time of the failed equipment. The system time when a dynamic disturbance event is detected is determined as the rescheduling trigger time. Based on the type of dynamic disturbance event, combined with the allocation relationship between processes and equipment in the current scheduling scheme, the processing requirements of the processes, and the processing capacity of the equipment, the affected objects are determined. Using the rescheduling trigger time as the state capture point, the process execution record and equipment operation status data are time-aligned to obtain the process execution status and equipment operation status at the rescheduling trigger time.

3. The event-aware production rescheduling method for the internal mixing workshop according to claim 2, characterized in that, Step S3 specifically includes: Based on the process execution status at the rescheduling trigger time, completed processes and processes that have started but not yet completed in the current scheduling plan are identified as locked processes, and the allocation relationship between the locked processes and equipment and the process execution status remain unchanged; processes that have not yet started in the current scheduling plan as of the rescheduling trigger time and all processes corresponding to newly inserted orders are identified as processes to be rescheduled; and the allocation relationship between the processes and equipment corresponding to the processes that were originally planned to be assigned to faulty equipment but have not yet started in the current scheduling plan is released. When the equipment is in a faulty state, the estimated repair completion time of the faulty equipment is determined based on the fault occurrence time and estimated repair time in the fault information. Obtain the actual completion time of the completed process and use the actual completion time of the completed process as the fixed process connection boundary of the corresponding subsequent process; obtain or determine the estimated completion time of the process that has started but not yet completed, determine the estimated completion time as the completion time of the corresponding locked process, and determine the time period between the rescheduling trigger time and the estimated completion time as the equipment occupancy time period of the corresponding locked process; Based on the equipment operating status at the time of rescheduling trigger, the equipment occupancy period of the locked process, and the estimated repair completion time of the faulty equipment, the equipment availability information is determined; the equipment availability information includes the equipment availability status, the earliest initial availability time, and the equipment unavailability period.

4. The event-aware production rescheduling method for the internal mixing workshop according to claim 3, characterized in that, The process constraints are as follows: when there is a preceding process, the subsequent process begins after the preceding process is completed and the corresponding process cooling time is met; the time constraints limit the earliest start time and completion time of the process; the resource constraints limit the equipment that the process can select and the corresponding available time range, and limit the execution time periods of different processes on the same equipment to not overlap. The reconstruction of the scheduling constraint set in step S4 includes: The process constraints are restructured. If the preceding process is a locked process, the process to be rescheduled will start after the locked process is completed and the corresponding process cooling time is met. The time constraints are restructured, and the earliest start time of the process to be rescheduled is updated to be no earlier than the rescheduling trigger time. The resource constraints are restructured, and the equipment that the process can select and the corresponding available time range are updated based on the equipment availability information.

5. The event-aware production rescheduling method for the internal mixing workshop according to claim 4, characterized in that, The construction of the rolling rescheduling model in step S4 specifically includes: Using the allocation relationship between the processes and equipment corresponding to the processes to be rescheduled, as well as the execution order, start time and completion time of the processes to be rescheduled, as the content to be determined, a rolling rescheduling model is constructed under the constraints of the reconstructed scheduling set, with the optimization objectives of minimizing delivery performance and minimizing overall operating costs. The cost parameters include processing energy consumption cost coefficient, standby energy consumption cost coefficient, and inventory cost coefficient; the delivery performance is determined based on the final completion time and corresponding order weight of each scheduled workpiece in a set of workpieces to be rescheduled, where at least one process belongs to the set of processes to be rescheduled; the comprehensive operating cost includes manufacturing-side energy consumption cost and logistics-side inventory cost; the manufacturing-side energy consumption cost is determined based on the total effective processing time and total idle standby time of each piece of equipment, combined with the processing energy consumption cost coefficient and the standby energy consumption cost coefficient; the logistics-side inventory cost is determined based on the relationship between the completion time and delivery date of each scheduled workpiece, as well as the relationship between the actual waiting time and process cooling time between adjacent processes, combined with the inventory cost coefficient.

6. The event-aware production rescheduling method for the internal mixing workshop according to claim 5, characterized in that, Step S5 specifically includes: The one-dimensional real-number encoding dimension of each individual in the population is determined based on the number of processes to be rescheduled. An initial population is generated, and the correspondence between each encoding position and each process to be rescheduled is established. The encoding positions are sorted in ascending order of the real-number encoding values ​​in each individual to obtain the order of the processes to be rescheduled for each individual. The current process to be decoded is determined sequentially according to the arrangement order and the process sequence relationship between the processes to be rescheduled. A dynamic greedy device selection strategy is used to determine the production equipment and execution time corresponding to the current process to be decoded, until all processes to be rescheduled are decoded, and a candidate rescheduling scheme is obtained. The dynamic greedy device selection strategy determines the completion time of the current process to be decoded on each candidate device based on the availability time of the candidate devices, the completion time of the preceding process of the current process to be decoded, and the process cooling time. The candidate device with the earliest completion time is selected first. When the completion time is the same, the candidate device with the shortest processing time is selected. When the completion time and processing time are the same, the candidate device is selected according to the preset device sorting relationship. The availability time of the corresponding production device is updated after each process to be rescheduled is determined. Based on the delivery performance and overall operating cost of each candidate rescheduling scheme, non-dominated sorting and congestion evaluation are performed. Individuals in the population are selected according to the non-dominated level and congestion. The one-dimensional real number encoding of the individuals in the population is updated using a population update strategy. The encoding position sorting, individual decoding, non-dominated sorting and congestion evaluation are repeated for the updated individuals in the population until the iteration termination condition is met, and the Pareto front solution set is obtained.

7. The event-aware production rescheduling method for the internal mixing workshop according to claim 6, characterized in that, Step S6 specifically includes: Determine the maximum and minimum values ​​of the objective function in the Pareto front solution set; Using a membership function based on fuzzy set theory, and based on the delivery performance and overall operating cost of each candidate rescheduling scheme, as well as the maximum and minimum values ​​of the optimization objectives, the delivery performance and overall operating cost are linearly normalized according to the rule that the smaller the objective value, the higher the satisfaction, to obtain the satisfaction membership value of each candidate rescheduling scheme on the two objectives. The satisfaction membership values ​​of each candidate rescheduling scheme on the two objectives are summarized and standardized to obtain the comprehensive satisfaction of each candidate rescheduling scheme. The candidate rescheduling scheme with the highest comprehensive satisfaction is determined as the target rescheduling scheme. The allocation relationship, execution order, start time and completion time of the processes and equipment corresponding to the processes to be rescheduled in the target rescheduling scheme are converted into production scheduling control instructions and issued. The production execution queue of the mixing workshop is updated according to the production scheduling control instructions, and the target rescheduling scheme is updated to the current scheduling scheme.

8. An event-aware production rescheduling system for a mixing workshop, characterized in that, The steps for performing the event-aware rescheduling method for internal mixing plant production according to any one of claims 1 to 7 include: The basic data acquisition module is used to acquire the basic production scheduling data, current scheduling scheme and scheduling constraint set of the mixing workshop; The event perception and status acquisition module is used to monitor the order status and equipment operation status of the mixing workshop in real time. When a dynamic disturbance event is detected, the rescheduling trigger time and the affected objects are determined, and the workshop production status information at the rescheduling trigger time is obtained. The process division and equipment availability information determination module is used to divide the processes in the current scheduling plan and the processes corresponding to newly inserted orders into locked processes and processes to be rescheduled processes based on the process execution status in the workshop production status information, and to determine the equipment availability information. The constraint reconstruction and model building module is used to reconstruct the scheduling constraint set based on the rescheduling trigger time, locked process, process to be rescheduled, and equipment availability information, combined with the production scheduling basic data, to obtain the reconstructed scheduling constraint set and build a rolling rescheduling model. The multi-objective optimization solution module is used to solve the rolling rescheduling model by employing a multi-objective optimization algorithm and combining the reconstructed scheduling constraint set to obtain the Pareto front solution set. The comprehensive satisfaction decision and instruction issuance module is used to calculate the comprehensive satisfaction of each candidate rescheduling scheme, select the candidate rescheduling scheme with the highest comprehensive satisfaction as the target rescheduling scheme, convert the target rescheduling scheme into a production scheduling control instruction, issue it, and update the production execution queue of the mixing workshop.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the event-aware rescheduling method for rescheduling production in a mixing plant as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the event-aware rescheduling method for the internal mixing workshop as described in any one of claims 1 to 7.

Images