Automatic production scheduling, machine scheduling and dynamic correction system based on MES

Abstract

The invention relates to an automatic production scheduling, machine scheduling and dynamic correction system based on MES, and the system comprises a data perception and collection module which collects static basic data and dynamic event data needed by machine scheduling; the production scheduling engine module is used for performing multi-dimensional production plan production scheduling through multiple production scheduling strategies based on the static basic data; the scheduling engine module is used for carrying out optimization calculation on the production plan of scheduling according to the constraint condition; the task scheduler is used for configuring and triggering a timed scheduling task and a periodically planned health examination task according to the optimized production plan; the plan correction module is used for evaluating the influence of the event on the current plan according to the dynamic event data and performing plan correction according to a preset rule decision; the plan issuing and synchronizing module is used for issuing the procedure plan generated after machine arrangement or correction as an executable instruction in the MES and synchronizing the executable instruction to the production terminal; and the manual regulation and control module is used for realizing man-machine cooperation based on intelligent machine tool scheduling, forced insertion and conflict detection.

Description

Technical Field

[0001] This invention relates to the field of manufacturing execution systems, and in particular to an automated production scheduling, machine scheduling and dynamic correction system based on MES. Background Technology

[0002] In the machining industry, Manufacturing Execution System (MES) is a key information system connecting the enterprise's planning layer and shop floor control layer. Advanced Planning and Scheduling (APS), as a core functional module of MES, is responsible for the refined scheduling of production tasks in terms of time, resources, and processes under limited capacity conditions. Existing APS systems typically suffer from the following shortcomings: 1) The scheduling process heavily relies on manual triggering and intervention by planners, failing to automatically respond to new orders or emergency insertions during off-peak hours (such as nighttime), resulting in delayed scheduling responses; 2) After the scheduling results are generated, if actual progress deviates from the schedule (such as equipment failure, material delays, or exceeding time limits), the system cannot automatically and promptly detect and reschedule, requiring manual detection and execution of the "rescheduling" operation. The real-time performance and accuracy of the correction are poor, leading to a disconnect between the scheduling plan and actual production, gradually losing its guiding value. Therefore, there is an urgent need for an APS system that can trigger scheduling manually, automatically, and on a timed basis, and can monitor production status in real time and automatically correct the plan, in order to improve the intelligence and automation level of scheduling and ensure consistency between planning and execution. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of existing technologies and provide an automated production scheduling, machine scheduling, and dynamic correction system based on MES. This system can automatically execute production scheduling according to preset strategies, monitor deviations in real time during production execution, and automatically trigger plan corrections, thereby achieving automated management of the entire production lifecycle and significantly improving the accuracy and executability of production plans.

[0004] To achieve the above objectives, the present invention provides the following solution: An MES-based automated production scheduling, machine scheduling, and dynamic correction system includes: The data sensing and acquisition module is used to collect static basic data and dynamic event data required by the scheduling system; The production scheduling engine module is used to perform multi-dimensional production planning and scheduling based on the static basic data and through various scheduling strategies. The scheduling engine module is used to optimize the production plan based on constraints. The task scheduler is used to configure and trigger scheduled machine scheduling tasks and periodic planned health check tasks based on the optimized production plan. The plan correction module is used to assess the impact of events on the current plan based on the dynamic event data, and to make plan correction decisions based on preset rules. The planning and synchronization module is used to publish the process plan generated after scheduling or correction as an executable instruction in the MES and synchronize it to the production terminal. The manual control module is used to force insertion and detect conflicts based on intelligent machine tool scheduling, so as to achieve human-machine collaboration.

[0005] Optionally, the static basic data includes: product process route, equipment capacity and calendar, material issuance time, and standard working hours; The dynamic event data includes: new orders / order changes, work order reporting progress, equipment status, and programming status.

[0006] Optionally, the production scheduling engine module includes: The production scheduling strategy management unit is used to set up various production scheduling strategies and implement them. The intelligent capacity synchronization unit is used to collect equipment abnormal events in real time through IoT and MES, predict capacity loss with AI and dynamically correct theoretical capacity deviation, trigger the production scheduling engine to recalculate, and ensure that the deviation rate between planned and actual capacity is less than the preset value. The intelligent pre-scheduling unit integrates forward scheduling simulation and backward scheduling verification algorithms. It conducts scenario testing through Monte Carlo simulation, intelligently predicts the risk of capacity conflict, and outputs pre-scheduling solutions with early warnings and bottleneck optimization suggestions.

[0007] Optionally, the scheduling strategy management unit may set various scheduling strategies, including: Forward scheduling mode: It deduces forward from the current time, dynamically constructs a resource constraint matrix, uses a dual-engine drive to generate time / strategy constraints, iteratively searches for the optimal node through ant colony algorithm, uses a hybrid algorithm for intelligent screening, and outputs a global scheduling solution that balances efficiency and fairness through closed-loop recursion. Inverted ordering mode: Starting from the end of the delivery date, reverse the process to build an inverted resource constraint network; use dual engines to generate an inverted time model and policy matrix; use ant colony inverted search combined with a hybrid algorithm to select the best nodes; and output a closed-loop solution that meets the delivery date and has the best resource utilization. Insertion mode: Emergency orders are dynamically inserted to build multi-dimensional resource constraints; the insertion strategy is configured using a rule engine, the optimal node is located iteratively using the ant colony algorithm, overload conflicts are intelligently detected and reordered, and the final plan that balances efficiency and stability is output.

[0008] Optionally, the scheduling engine module has a built-in scheduling algorithm to optimize the production plan based on the input constraints. The scheduling algorithm employs a hybrid algorithm combining constraint programming and heuristic algorithms. First, it utilizes the hard and soft constraints defined by constraint programming. Then, it uses a heuristic algorithm to perform global optimization within the solution space that satisfies the hard constraints, aiming to minimize the total delay time, maximize equipment utilization, and minimize setup time. The hard constraints include: equipment uniqueness, process sequence, and material delivery time. The soft constraints include: delivery date and equipment changeover time.

[0009] Optionally, the task scheduler includes: The timed scheduling unit is used to configure one or more customizable timed tasks. When a task is triggered, the scheduling engine module is automatically invoked to generate a new round of master production plan using the latest order data and basic data. The scheduled work reporting unit is used to process the work reporting cache in batches on a regular basis, update the work order status, and drive the scheduling of subsequent processes. The timed correction unit is used to trigger periodic planned health check tasks.

[0010] Optionally, in the scheduled work reporting unit: The work order is updated by periodically consuming the reported work records in the cache through a scheduled task; the actual number of completed work orders, status, and process timestamp are updated. The status includes: started, partially completed, and closed. Based on real-time progress reporting, the pre-scheduling management of subsequent processes is dynamically driven: combining process routes, resource load and delivery constraints, the expected start / completion time of downstream tasks is automatically deduced, realizing rolling advance planning.

[0011] Optionally, the planned correction module includes: The work reporting correction unit is used to automatically compare and dynamically adjust the original scheduling plan based on the actual work reporting information in the workshop; when an abnormal work reporting is detected, an abnormal work order is automatically generated. The automatic correction unit is used to periodically check for deviations in plan execution and automatically trigger rearrangement calculations for correction.

[0012] Optionally, in the manual control module: Intelligent machine tool switching scheduling: Analyzes the current load, available time period, process capacity and priority rules of the target machine tool, and automatically calculates the optimal insertion position, that is, finds the earliest feasible time window without affecting high priority tasks; when the machine tool switching is executed, the relevant work orders are immediately locked to prevent interference from other operations during the rescheduling period, and ensure data consistency and scheduling atomicity. Forced insertion and conflict detection: For all affected work orders that are at risk of conflict or overdue due to insertion, an abnormal work order list will be automatically generated, the impact type will be marked, and the list will be pushed to the planner's or scheduler's workbench; users can then make manual decisions based on this: accept the adjustment, further optimize, split the work order, or coordinate resources.

[0013] The beneficial effects of this invention are as follows: Automation and unattended operation: The automatic operation of the machine scheduling during non-working hours is realized through timed tasks, which reduces the reliance on manual labor and improves the response speed.

[0014] Real-time performance and high adaptability: Through an event-driven dynamic correction mechanism, production plans can quickly adapt to changes in the workshop, always maintaining a high degree of consistency between the plan and reality, thus enhancing the guiding value of the plan.

[0015] Intelligent decision-making: By combining a rule engine and optimization algorithms, the system can not only execute automatically, but also intelligently select the optimal correction strategy (local or global reordering) based on business rules, balancing computational efficiency and scheduling effectiveness. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 For the calculation of processing time in this embodiment of the invention; Figure 2 This is a schematic diagram of the overall process of an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the implementation process of the timed scheduling function in an embodiment of the present invention; Figure 4 This is a schematic diagram illustrating the implementation process of the event-driven work reporting correction function in an embodiment of the present invention; Figure 5 This is a schematic diagram illustrating the implementation process of the automatic correction function in an embodiment of the present invention; Figure 6 This is a schematic diagram illustrating the automatic rotation of the machine tool with the same material number according to an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] This embodiment proposes an automated production scheduling, machine scheduling, and dynamic correction system based on MES, including: The data sensing and acquisition module is used to collect static basic data and dynamic event data required by the scheduling system; The production scheduling engine module is used to perform multi-dimensional production planning and scheduling based on the static basic data and through various scheduling strategies. The scheduling engine module is used to optimize the production plan based on constraints. The task scheduler is used to configure and trigger scheduled machine scheduling tasks and periodic planned health check tasks based on the optimized production plan. The plan correction module is used to assess the impact of events on the current plan based on the dynamic event data, and to make plan correction decisions based on preset rules. The planning and synchronization module is used to publish the process plan generated after scheduling or correction as an executable instruction in the MES and synchronize it to the production terminal. The manual control module is used to force insertion and detect conflicts based on intelligent machine tool scheduling, so as to achieve human-machine collaboration.

[0021] Furthermore, the static basic data includes: product process route, equipment capacity and calendar, material issuance time, and standard working hours; The dynamic event data includes: new orders / order changes, work order reporting progress, equipment status, and programming status.

[0022] Specifically, in this embodiment, the data sensing and acquisition module interfaces with other modules of the MES (such as order management, programming management, equipment management, work order reporting, and APS scheduling management) to continuously collect the static basic data and dynamic event data required for machine scheduling. Static basic data includes product process routes, equipment capacity and calendar, material issuance time, standard working hours, etc., and is stored in the basic database. Dynamic event data includes new orders / order changes, work order reporting progress (start, completion, pause), equipment status (fault, repair, recovery), programming status, etc.

[0023] Furthermore, the production scheduling engine module includes: The production scheduling strategy management unit is used to set up various production scheduling strategies and implement them. The intelligent capacity synchronization unit is used to collect equipment abnormal events in real time through IoT and MES, predict capacity loss with AI and dynamically correct theoretical capacity deviation, trigger the production scheduling engine to recalculate, and ensure that the deviation rate between planned and actual capacity is less than the preset value. The intelligent pre-scheduling unit integrates forward scheduling simulation and backward scheduling verification algorithms. It conducts scenario testing through Monte Carlo simulation, intelligently predicts the risk of capacity conflict, and outputs pre-scheduling solutions with early warnings and bottleneck optimization suggestions.

[0024] Furthermore, the scheduling strategy management unit sets various scheduling strategies, including: Forward scheduling mode: It deduces forward from the current time, dynamically constructs a resource constraint matrix, uses a dual-engine drive to generate time / strategy constraints, iteratively searches for the optimal node through ant colony algorithm, uses a hybrid algorithm for intelligent screening, and outputs a global scheduling solution that balances efficiency and fairness through closed-loop recursion. Inverted ordering mode: Starting from the end of the delivery date, reverse the process to build an inverted resource constraint network; use dual engines to generate an inverted time model and policy matrix; use ant colony inverted search combined with a hybrid algorithm to select the best nodes; and output a closed-loop solution that meets the delivery date and has the best resource utilization. Insertion mode: Emergency orders are dynamically inserted to build multi-dimensional resource constraints; the insertion strategy is configured using a rule engine, the optimal node is located iteratively using the ant colony algorithm, overload conflicts are intelligently detected and reordered, and the final plan that balances efficiency and stability is output.

[0025] Specifically, in this embodiment, the production scheduling engine module: achieves multi-dimensional production planning and scheduling through various production scheduling methods and strategies to meet personalized order demands; Forward row mode: Dynamic resource modeling: Real-time synchronization of data such as equipment capacity, material inventory, and process routes to construct a multi-dimensional resource constraint matrix; Dual-engine drive: Time engine: generates time constraint models based on user-defined rules (such as delivery priority and process dependency); Strategy engine: encapsulates production scheduling strategies (such as load balancing and emergency order insertion) to form a configurable strategy library; Ant colony collaborative optimization: Through the dynamic pheromone evaporation mechanism, iteratively search for a set of feasible production scheduling nodes under dual-engine constraints; Intelligent node selection: A hybrid algorithm of "priority weight + round-robin compensation" is used to select the optimal solution from candidate nodes to avoid local convergence; Closed-loop recursive scheduling: The above process is executed cyclically for all work orders, outputting a global production scheduling plan that balances efficiency and fairness, and supports dynamic rescheduling.

[0026] Inverted format: Dynamic resource preloading: Real-time synchronization of data such as equipment capacity, material availability, and process routes to construct an inverse resource constraint network; Dual-engine architecture: Reverse timing engine: Generates a time-reverse constraint model based on user-defined rules (such as backward delivery date calculation and fronting up bottleneck processes); Strategy Engine: Encapsulates scheduling strategies (such as critical chain buffers and elastic resource allocation) to form a strategy matrix; Ant colony reverse search: Through the pheromone reverse evaporation mechanism, feasible production scheduling paths are iteratively searched from the end point of delivery to the starting point under the constraints of dual engines; Intelligent node selection: Employs a hybrid algorithm of "dynamic priority + load balancing" to select the optimal solution from candidate nodes, avoiding resource overload; Closed-loop recursive scheduling: The above process is executed cyclically for all work orders, and a reverse scheduling plan that meets the delivery constraints and has the best resource utilization is output, supporting dynamic adjustment.

[0027] Power strip mode: Dynamic resource synchronization: Real-time collection of data such as equipment capacity, material inventory, and human skills to build a multi-dimensional resource constraint model; Rule engine construction: Based on user-defined priority (such as EDD / SRPT), mold change time and other rules, a configurable interpolation time engine is generated; Ant colony optimization scheduling: Through a dynamic pheromone update mechanism combined with iterative capacity constraints, the optimal global scheduling node is accurately located; Intelligent conflict detection: By comparing the scheduling results with the resource pool, it automatically identifies overloaded nodes and triggers rescheduling to ensure the feasibility of the solution; Closed-loop output: After eliminating invalid schedules, the output is a final production schedule that balances efficiency and stability, and supports visual adjustment via Gantt chart.

[0028] In the intelligent production capacity synchronization unit: Real-time anomaly detection: By connecting the device's IoT sensors to the MES system, abnormal events such as equipment downtime, malfunctions, and mold changes are collected periodically, and AI algorithms are used to predict potential capacity losses; Dynamic data calibration: The collected abnormal data is written into the digital twin model in real time to automatically correct the deviation between theoretical and actual capacity; Closed-loop scheduling correction: Based on the calibrated capacity data, the APS scheduling engine is triggered to recalculate work order allocation to ensure that the deviation rate between planned and actual capacity is less than 2%, thus guaranteeing delivery reliability.

[0029] In the intelligent pre-scheduling unit: Dual-algorithm fusion engine: Forward simulation: Based on constraints such as equipment capacity and material availability, it forward-deduces the theoretical delivery cycle of the work order from the current time; Backward verification: It reverse-deduces from the customer's delivery date to identify resource bottlenecks and buffer times on the critical path. Dynamic conflict resolution: The dual-output results are tested more than 1,000 times through Monte Carlo simulation, and the risk points of capacity conflict are automatically marked. Intelligent predictive output: Generates a pre-production scheduling plan with risk warnings, including: the theoretical delivery date of each work order (forward scheduling result), the actual achievable delivery date (after reverse scheduling correction), and suggestions for optimizing bottleneck processes (such as equipment allocation and outsourcing).

[0030] Furthermore, the scheduling engine module has a built-in scheduling algorithm that optimizes the production plan based on the input constraints. The scheduling algorithm employs a hybrid algorithm combining constraint programming and heuristic algorithms. First, it utilizes the hard and soft constraints defined by constraint programming. Then, it uses a heuristic algorithm to perform global optimization within the solution space that satisfies the hard constraints, aiming to minimize the total delay time, maximize equipment utilization, and minimize setup time. The hard constraints include: equipment uniqueness, process sequence, and material delivery time. The soft constraints include: delivery date and equipment changeover time.

[0031] Specifically, in this embodiment, the ranking engine module is the system's computing unit, which has a built-in ranking algorithm and performs optimization calculations based on the input constraints and objectives.

[0032] Algorithm Implementation Logic: A hybrid algorithm combining constraint programming (CP) and heuristic algorithms is employed. First, constraint programming is used to define strict hard constraints (such as equipment uniqueness, process sequence, and material delivery time) and soft constraints (such as delivery date and equipment changeover time). Then, a heuristic algorithm is used to perform global optimization within the solution space that satisfies the hard constraints, aiming to minimize total delay time, maximize equipment utilization, and minimize setup time. The algorithm supports finite capacity scheduling and forward / backward scheduling modes.

[0033] Furthermore, the task scheduler includes: The timed scheduling unit is used to configure one or more customizable timed tasks. When a task is triggered, the scheduling engine module is automatically invoked to generate a new round of master production plan using the latest order data and basic data. The scheduled work reporting unit is used to process the work reporting cache in batches on a regular basis, update the work order status, and drive the scheduling of subsequent processes. The timed correction unit is used to trigger periodic planned health check tasks.

[0034] Furthermore, in the scheduled work reporting unit: The work order is updated by periodically consuming the reported work records in the cache through a scheduled task; the actual number of completed work orders, status, and process timestamp are updated. The status includes: started, partially completed, and closed. Based on real-time progress reporting, the pre-scheduling management of subsequent processes is dynamically driven: combining process routes, resource load and delivery constraints, the expected start / completion time of downstream tasks is automatically deduced, realizing rolling advance planning.

[0035] Specifically, in this embodiment, the task scheduler is a "timer" and "trigger" that the system runs automatically.

[0036] Scheduled production scheduling function implementation: Configure one or more customizable scheduled tasks (e.g., execute once every 10 minutes). When a task is triggered, the scheduling engine module is automatically invoked, using the latest order data and basic data to generate a new round of master production plan.

[0037] Scheduled work reporting implementation: Work report data is first written to a highly available cache pool to decouple it from the main business system, ensuring that the collection terminal can continuously and with low latency receive work report requests from operation terminals even under high concurrency or network fluctuation scenarios. The system periodically consumes work report records in the cache through robust scheduled jobs, atomically updating the actual number of completed work orders, status (such as "started", "partially completed", "closed"), and process timestamps. Building upon this foundation, the system dynamically drives the pre-scheduling management of subsequent processes based on real-time work progress reports. By combining process routes, resource loads, and delivery constraints, it automatically extrapolates the expected start / completion times of downstream tasks, enabling rolling plan advancement and improving production line collaboration efficiency. The entire processing chain adopts an asynchronous, idempotent, and retryable event-driven architecture. Even in the event of network interruptions or temporary service unavailability, work data can be reliably replayed and compensated after fault recovery, ensuring data integrity and production continuity, and meeting the industrial-grade requirements of Manufacturing Execution Systems (MES) for high reliability and strong consistency.

[0038] The scheduled correction function is implemented as follows: In addition to real-time correction triggered by response events, a periodic "planned health check" task can also be set (executed once a day at 21:00). This task actively scans the overall deviation between the plan and the reported progress and automatically triggers the plan correction module to perform global correction.

[0039] Furthermore, the planned correction module includes: The work reporting correction unit is used to automatically compare and dynamically adjust the original scheduling plan based on the actual work reporting information in the workshop; when an abnormal work reporting is detected, an abnormal work order is automatically generated. An automatic correction unit for periodically checking the deviation of plan execution and automatically triggering rearrangement calculation for correction.

[0040] Specifically, in this embodiment, the plan correction module is the core calculation unit of the system, responsible for processing dynamic events, calculating their impact on the current plan, and deciding whether to trigger and how to trigger correction.

[0041] Work reporting correction: Work reporting correction means that the system automatically compares and dynamically adjusts the original machine arrangement plan according to the process information actually reported in the workshop; when detecting work reporting anomalies (such as process skipping, overtime, out-of-order, etc.), the system will automatically generate an exception work order and push it to the relevant responsible person for manual intervention in verification and processing to ensure the accuracy and executability of the production plan.

[0042] Automatic correction: Correction mode: Global correction; Implementation logic: Based on the configuration strategy and taking the running time as a node, calculate the production time of all processes without changing the production plan of the remote machine tool. Correction evaluation: According to the machine tool status (such as equipment failure) and the scope of influence (affected equipment, processes, time), combined with the current plan, quickly evaluate the degree of deviation. For example, calculate the expected delay time of the affected processes. Correction decision: Make a decision according to the preset rule library. Rule examples: IF (Machine arrangement start time > Scheduling start time) AND (Machine arrangement end time < PMC delivery date) THEN the plan is postponed; IF (Machine arrangement start time < Scheduling start time) AND (Machine arrangement end time < PMC delivery date) THEN the plan is advanced; IF (Machine arrangement end time > PMC delivery date) THEN the plan is overdue.

[0043] Furthermore, in the manual regulation module: Intelligent transfer machine tool scheduling: Analyze the current load, available time period, process capabilities and priority rules of the target machine tool, and automatically calculate the optimal insertion position, that is, find the earliest feasible time window without affecting high-priority tasks; when performing machine transfer, immediately lock the involved work orders to prevent interference by other operations during rearrangement, and ensure data consistency and scheduling atomicity. Force insertion and detect conflicts: For all affected work orders that have conflicts or overtime risks due to insertion, an exception work order list will be automatically generated, marking the type of influence, and pushed to the workbench of the planner or dispatcher; the user makes a manual decision based on this: accept the adjustment, further optimize, split the work order or coordinate resources.

[0044] Specifically, in this embodiment, the manual control module: Machine tool transfer is a key function in Advanced Planning and Scheduling (APS) systems, designed to address resource adjustment needs arising from equipment failures, uneven loads, process changes, or urgent order insertions during production. This operation is not simply transferring work orders from one machine to another; rather, it is intelligently executed based on pre-defined scheduling strategies. The system first analyzes the target machine's current load, available time slots, process capabilities, and priority rules, automatically calculating the optimal insertion position—that is, finding the earliest feasible time window without affecting high-priority tasks, ensuring minimal disruption to the overall schedule.

[0045] When a machine transfer is executed, the system immediately locks the relevant work orders to prevent interference from other operations during rescheduling, ensuring data consistency and scheduling atomicity. Furthermore, the machine transfer function supports manual intervention by specifying a date, such as when a user wants to force a work order to be scheduled on a machine three days later. In this case, the system not only verifies the feasibility of that time period but also intelligently identifies all related work orders affected by the insertion—including subsequent tasks with the occupied time and downstream processes that depend on the output of this work order—and assesses whether it violates delivery deadlines, capacity, or process constraints.

[0046] For all work orders affected by insertion that create conflicts or risks of delays, the system will automatically generate a list of abnormal work orders, detailing the impact type (such as "delay," "resource conflict," "broken prerequisite dependencies," etc.), and push it to the planner's or scheduler's workbench. Users can then make manual decisions based on this: accept adjustments, further optimize, split work orders, or coordinate resources. The entire process organically combines automated rescheduling with manual collaboration, improving scheduling flexibility while ensuring the plan's executability and transparency, effectively supporting the manufacturing site's ability to respond quickly to dynamic changes.

[0047] This embodiment also proposes an automated production scheduling, machine scheduling, and dynamic correction method based on MES, applied to the above system, including the following steps: System initialization, configuring scheduled machine scheduling tasks, scheduling strategies, and correction rules.

[0048] Execute scheduled production scheduling: The task scheduler is triggered at a preset time point. The scheduling engine module reads all currently scheduled and unscheduled orders and basic data, runs a hybrid optimization algorithm, generates the globally optimal production plan, and issues it through the plan release module.

[0049] Real-time production monitoring and event collection: During production execution, the data sensing module collects event data such as work reports and equipment status in real time.

[0050] Dynamic corrective decision-making and execution: The planning correction module receives events and evaluates their impact based on the rules engine.

[0051] If the correction trigger condition is met, the scheduling engine is invoked, and the affected part (local or global) is re-scheduled and optimized, using the current actual progress as a new starting point.

[0052] The revised plan and change instructions will be synchronized to relevant positions and terminals through the plan release module.

[0053] Automatic correction: The task scheduler triggers an automatic correction task, calculates the overall execution deviation of the plan, and executes the correction process.

[0054] Figure 1 This embodiment illustrates the calculation of processing time. Figure 2 This embodiment of the system presents a schematic diagram of its overall flow. Figure 3 The implementation process of the timed machine scheduling function is demonstrated. Figure 4 The implementation process of the event-driven work reporting correction function is demonstrated. Figure 5 The implementation process of the automatic correction function is demonstrated; Figure 6 The machine tool operation was demonstrated to automatically rotate with the same material number.

[0055] The APS Advanced Scheduling System comprises two sub-modules: APS Production Scheduling and APS Machine Scheduling. APS Production Scheduling, based on the process route and relevant production data such as the Bill of Materials (BOM), tools, and auxiliary materials, selects the most suitable resources for production scheduling according to optimal principles. When resources are insufficient, the system automatically selects alternative resources or suggests outsourcing to ensure the plan can be executed normally. If the customer's delivery deadline cannot be met, the system automatically issues an alarm, and PMC personnel then adjust the plan or resources. APS Machine Scheduling, based on the APS Production Scheduling plan, searches for the fastest available machine tools within the planned time frame, considering factors such as material arrival time and machine tool processing capacity to calculate the most suitable machine tool for scheduling. If a suitable machine tool cannot be found within the given planned resources and time frame, the APS Scheduling system automatically reports a scheduling failure to APS Production, which then searches for other suitable data for a second scheduling.

[0056] Taking the APS system of a CNC machining workshop as an example: The system automatically performs a full plant machine scheduling every 4 hours. Once triggered, the scheduling engine calls up all "unscheduled" work orders, combines them with the equipment maintenance calendar and employee shifts, and schedules machines with "order delivery date" as the highest priority target to generate a scheduling plan.

[0057] During daytime production, one CNC machine needs to be stopped for maintenance, and production personnel set the maintenance time. The timed calibration function process is as follows: The equipment data acquisition system pushes "equipment maintenance" events to the MES event bus.

[0058] In this embodiment, the data sensing module captures the event and forwards it to the correction module (automatic correction).

[0059] The scheduling engine is activated to recalculate the production time of the processes already scheduled on the equipment.

[0060] After the rearrangement, the system pushes the plan change information to the original equipment operators, team leaders and material handlers through the Kanban board, and prompts them with the new equipment arrangements and start-up time.

[0061] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. An automated production scheduling, machine scheduling, and dynamic correction system based on MES, characterized in that, include: The data sensing and acquisition module is used to collect static basic data and dynamic event data required by the scheduling system; The production scheduling engine module is used to perform multi-dimensional production planning and scheduling based on the static basic data and through various scheduling strategies. The scheduling engine module is used to optimize the production plan based on constraints. The task scheduler is used to configure and trigger scheduled machine scheduling tasks and periodic planned health check tasks based on the optimized production plan. The plan correction module is used to assess the impact of events on the current plan based on the dynamic event data, and to make plan correction decisions based on preset rules. The planning and synchronization module is used to publish the process plan generated after scheduling or correction as an executable instruction in the MES and synchronize it to the production terminal. The manual control module is used to force insertion and detect conflicts based on intelligent machine tool scheduling, so as to achieve human-machine collaboration.

2. The MES-based automated production scheduling, machine scheduling, and dynamic correction system according to claim 1, characterized in that, The static basic data includes: product process route, equipment capacity and calendar, material issuance time, and standard working hours; The dynamic event data includes: new orders / order changes, work order reporting progress, equipment status, and programming status.

3. The MES-based automated production scheduling, machine scheduling, and dynamic correction system according to claim 1, characterized in that, The production scheduling engine module includes: The production scheduling strategy management unit is used to set up various production scheduling strategies and implement them. The intelligent capacity synchronization unit is used to collect equipment abnormal events in real time through IoT and MES, predict capacity loss with AI and dynamically correct theoretical capacity deviation, trigger the production scheduling engine to recalculate, and ensure that the deviation rate between planned and actual capacity is less than the preset value. The intelligent pre-scheduling unit integrates forward scheduling simulation and backward scheduling verification algorithms. It conducts scenario testing through Monte Carlo simulation, intelligently predicts the risk of capacity conflict, and outputs pre-scheduling solutions with early warnings and bottleneck optimization suggestions.

4. The MES-based automated production scheduling, machine scheduling, and dynamic correction system according to claim 3, characterized in that, The production scheduling strategy management unit sets various production scheduling strategies, including: Forward scheduling mode: It deduces forward from the current time, dynamically constructs a resource constraint matrix, uses a dual-engine drive to generate time / strategy constraints, iteratively searches for the optimal node through ant colony algorithm, uses a hybrid algorithm for intelligent screening, and outputs a global scheduling solution that balances efficiency and fairness through closed-loop recursion. Inverted ordering mode: Starting from the end of the delivery date, reverse the process to build an inverted resource constraint network; use dual engines to generate an inverted time model and policy matrix; use ant colony inverted search combined with a hybrid algorithm to select the best nodes; and output a closed-loop solution that meets the delivery date and has the best resource utilization. Insertion mode: Emergency orders are dynamically inserted to build multi-dimensional resource constraints; the insertion strategy is configured using a rule engine, the optimal node is located iteratively using the ant colony algorithm, overload conflicts are intelligently detected and reordered, and the final plan that balances efficiency and stability is output.

5. The MES-based automated production scheduling, machine scheduling, and dynamic correction system according to claim 1, characterized in that, The scheduling engine module has a built-in scheduling algorithm that optimizes the production plan based on the input constraints. The scheduling algorithm employs a hybrid algorithm combining constraint programming and heuristic algorithms. First, it utilizes the hard and soft constraints defined by constraint programming. Then, it uses a heuristic algorithm to perform global optimization within the solution space that satisfies the hard constraints, aiming to minimize the total delay time, maximize equipment utilization, and minimize setup time. The hard constraints include: equipment uniqueness, process sequence, and material delivery time. The soft constraints include: delivery date and equipment changeover time.

6. The MES-based automated production scheduling, machine scheduling, and dynamic correction system according to claim 1, characterized in that, The task scheduler includes: The timed scheduling unit is used to configure one or more customizable timed tasks. When a task is triggered, the scheduling engine module is automatically invoked to generate a new round of master production plan using the latest order data and basic data. The scheduled work reporting unit is used to process the work reporting cache in batches on a regular basis, update the work order status, and drive the scheduling of subsequent processes. The timed correction unit is used to trigger periodic planned health check tasks.

7. The MES-based automated production scheduling, machine scheduling, and dynamic correction system according to claim 6, characterized in that, In the scheduled work unit: The work order is updated by periodically consuming the reported work records in the cache through a scheduled task; the actual number of completed work orders, status, and process timestamp are updated. The status includes: started, partially completed, and closed. Based on real-time progress reporting, the pre-scheduling management of subsequent processes is dynamically driven: combining process routes, resource load and delivery constraints, the estimated start / completion time of downstream tasks is automatically deduced, realizing rolling advance planning.

8. The MES-based automated production scheduling, machine scheduling, and dynamic correction system according to claim 1, characterized in that, The planned correction module includes: The work reporting correction unit is used to automatically compare and dynamically adjust the original scheduling plan based on the actual work reporting information in the workshop; when an abnormal work reporting is detected, an abnormal work order is automatically generated. The automatic correction unit is used to periodically check the deviation of the plan execution and automatically trigger the recalculation correction.

9. The MES-based automated production scheduling, machine scheduling, and dynamic correction system according to claim 1, characterized in that, In the manual control module: Intelligent machine tool switching scheduling: Analyzes the current load, available time period, process capacity and priority rules of the target machine tool, and automatically calculates the optimal insertion position, that is, finds the earliest feasible time window without affecting high-priority tasks; when the machine tool switching is executed, the relevant work orders are immediately locked to prevent interference from other operations during the rescheduling period, and ensure data consistency and scheduling atomicity. Forced insertion and conflict detection: For all affected work orders that are at risk of conflict or overdue due to insertion, an abnormal work order list will be automatically generated, the impact type will be marked, and the list will be pushed to the planner's or scheduler's workbench; users can then make manual decisions based on this: accept the adjustment, further optimize, split the work order, or coordinate resources.

Images