Cabinet flexible production line intelligent scheduling and material collaborative optimization MES system

Abstract

The invention belongs to the technical field of intelligent manufacturing and production scheduling, and particularly discloses and provides a cabinet flexible production line intelligent scheduling and material collaborative optimization MES system, which comprises a dynamic process modeling module for constructing a dynamic process route model based on order configuration, BOM and equipment capability; the real-time state sensing module forms a production line operation state data set through multi-source data acquisition; the rolling scheduling optimization module generates an initial production scheduling scheme according to a multi-objective function; the material collaborative distribution module performs material alignment verification based on the schedule and generates a synchronous distribution instruction; the disturbance response rescheduling module triggers local re-optimization for events such as equipment faults. According to the system, a dynamic process route is constructed based on order configuration and equipment capability, multi-target scheduling in a rolling window is driven by fusing multi-source real-time data, accurate cooperation of material delivery and process rhythm is realized, and local rapid rescheduling under disturbance events is supported.

Description

Technical Field

[0001] This invention belongs to the field of intelligent manufacturing and production scheduling technology, and relates to an intelligent scheduling and material collaborative optimization MES system for a flexible cabinet production line. Background Technology

[0002] With the development of the customized home furnishing industry, cabinet manufacturing is transforming from traditional mass standardized production to flexible intelligent manufacturing with multiple varieties and small batches. Manufacturing Execution Systems (MES), as a key hub connecting enterprise resource planning (ERP) and shop floor control, are crucial for improving production efficiency, ensuring delivery, and optimizing resource allocation. Especially in cabinet manufacturing, which is a highly non-standardized, complex, and material-constrained discrete environment, MES not only needs to monitor equipment, personnel, and processes in real time, but also needs to possess the ability to dynamically schedule orders, flexible production lines, and coordinate and intelligently optimize multi-source materials to cope with the dual pressures of rapid market response and cost control.

[0003] Current mainstream MES systems in cabinet manufacturing mostly employ static rules or simple heuristic algorithms for scheduling. Their core revolves around pre-defined process routes and fixed cycle times. They first decompose material requirements based on the product's Bill of Materials (BOM), then set standard working hours based on historical production capacity, and generate production plans through finite or infinite capacity scheduling. Material management relies on warehousing systems or ERP modules to track inventory and trigger replenishment. The MES system passively receives material arrival status, lacking proactive coordination regarding material availability, delivery timing, and production line consumption rhythm. This architecture can maintain efficiency in the early stages when orders are stable and product variations are minimal, achieving production visualization and traceability. However, its core limitation lies in separating scheduling from material coordination, failing to establish dynamic feedback and joint optimization mechanisms.

[0004] Faced with highly customized cabinet orders, customers' diverse needs for materials, hardware, and functional modules result in significant differences in the process, duration, and material combinations for each order. Static scheduling struggles to accurately predict processing load, easily leading to production line stagnation due to bottlenecks or material delays. Material systems rely solely on the initial BOM for matching, neglecting dynamic disturbances such as equipment failures, order insertions, and rework, often resulting in inefficient situations where plans are fully booked but materials are unavailable on-site. Furthermore, existing systems, when handling multi-station parallel operations and resource competition on flexible production lines, often employ local optimization strategies, ignoring the spatiotemporal coupling between material delivery and equipment availability, sacrificing overall output efficiency and resource utilization.

[0005] Therefore, building an MES that integrates intelligent scheduling and material coordination mechanisms, enabling it to generate robust scheduling solutions based on real-time capacity and order priority in the flexible production of cabinets with frequent non-standard and dynamic disturbances, and to drive precise on-demand material delivery to achieve full spatiotemporal alignment of all elements, has become a core challenge and an urgent problem for industry technicians. Summary of the Invention

[0006] In view of this, in order to solve the problems mentioned in the background technology, a smart scheduling and material collaborative optimization MES system for flexible cabinet production lines is proposed.

[0007] The objective of this invention can be achieved through the following technical solution: This invention provides an intelligent scheduling and material collaborative optimization MES system for a flexible cabinet production line, comprising:

[0008] The dynamic process modeling module constructs a dynamic process route model for cabinet orders based on order configuration parameters, product structure BOM, and equipment capacity constraints.

[0009] The real-time status sensing module collects information on workshop equipment status, work-in-process location, personnel work progress, and material inventory in real time, forming a production line operation status dataset.

[0010] The rolling scheduling optimization module generates an initial production scheduling plan that includes process start and end times, equipment allocation, and personnel scheduling based on the dynamic process route model and the production line operation status dataset.

[0011] The material collaborative delivery module performs material completeness verification based on the material consumption sequence and quantity of each process in the initial production scheduling plan, and generates material delivery instructions that are synchronized with the process execution rhythm.

[0012] The disturbance response rescheduling module triggers a scheduling re-optimization mechanism when it detects equipment failure, emergency order insertion, or rework disturbance events. It synchronously updates the dynamic process route model and material delivery instructions, and outputs a collaborative scheduling scheme after the disturbance response.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. The present invention constructs a dynamic process route model based on order configuration and equipment capacity, uses a directed acyclic graph to express multi-station parallel paths, and integrates delivery deadline and bottleneck resource information to perform path scoring and screening, so that the scheduling basis has the adaptive capability of non-standard orders, avoiding resource mismatch and plan distortion caused by fixed process routes.

[0014] 2. This invention drives global resource coordination with a multi-objective function within each time window, and quickly responds to dynamic events such as equipment failures or order insertions by freezing undisturbed processes and local re-optimization, thus ensuring scheduling stability and improving disturbance recovery efficiency.

[0015] 3. This invention introduces a time window calculation mechanism that is strictly synchronized with the process rhythm in the material distribution process, and embeds a replacement material verification and BOM dynamic update process, so that the material completeness judgment is transformed from static BOM verification to dynamic collaboration based on real-time scheduling and inventory status, effectively eliminating the disconnect between production planning and material supply.

[0016] 4. This invention uses IoT sensors, RFID positioning, and workstation terminals to form a multi-source data acquisition architecture, which enables millisecond-level synchronization of equipment, work-in-process, personnel, and material status. This provides high-fidelity input for scheduling and material collaboration, supports precise alignment of all elements (human, machine, material, method, and environment) in the spatiotemporal dimension, and significantly improves the resource utilization rate and order fulfillment capability of the flexible production line for cabinets. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of 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.

[0018] Figure 1 This is a schematic diagram of the system module connections of the present invention.

[0019] Figure 2 This is a schematic diagram of the analysis process for the collaborative material delivery module. Detailed Implementation

[0020] 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.

[0021] Please see Figure 1 As shown, this invention provides an intelligent scheduling and material collaborative optimization MES system for a flexible cabinet production line, including a dynamic process modeling module, a real-time status perception module, a rolling scheduling optimization module, a material collaborative delivery module, and a disturbance response rescheduling module. The dynamic process modeling module is connected to the real-time status perception module, the real-time status perception module is connected to the rolling scheduling optimization module, the rolling scheduling optimization module is connected to the material collaborative delivery module, the material collaborative delivery module is connected to the disturbance response rescheduling module, and the disturbance response rescheduling module is connected to the dynamic process modeling module, the rolling scheduling optimization module, and the material collaborative delivery module, respectively.

[0022] The dynamic process modeling module constructs a dynamic process route model for cabinet orders based on order configuration parameters, product structure BOM, and equipment capacity constraints.

[0023] In a preferred embodiment of the present invention, the construction of a dynamic process route model for a cabinet order includes: parsing the customized configuration items in the customer order and extracting the board type, hardware specifications, functional module combinations, and surface treatment process parameters.

[0024] For example, the board type includes, but is not limited to, solid wood particleboard, plywood, and OSB; the hardware specifications include, but are not limited to, hinge brand, slide length, and handle model; the functional module combination includes, but is not limited to, the number of drawers, pull-out basket configuration, and lighting system; and the surface treatment process parameters include, but are not limited to, edge banding color, paint gloss, and coating texture.

[0025] Based on a preset process rule library, the set of optional processing steps corresponding to the configuration item is matched, and logical dependencies between the steps are established.

[0026] It should be noted that the process rule base stores the mapping relationship between various configuration items and optional processing steps in a structured form. Preferably, when the order specifies the use of paint-free board, invisible hinges, and embedded handles, the system automatically matches the corresponding set of processes, including material cutting, edge banding, drilling, trial assembly, and packaging, and excludes unsuitable painting or sanding processes.

[0027] It should be noted that the process rule base is constructed as follows: historical successful order process execution records are collected, the mapping relationship between configuration items and processes is extracted, and data such as processing cycle time and pass rate of various configuration item combinations on different equipment are statistically analyzed. Equipment adaptation scores are assigned to optional processes and activation thresholds are set. For new configuration item combinations without historical data, the expert rule reasoning unit is activated to generate temporary process paths based on material properties, etc. Subsequently, the actual execution performance of temporary paths is periodically included in the base to complete the closed-loop iterative update of the rule base, providing support for matching configuration items with optional processes and establishing process logical dependencies.

[0028] It should be noted that the expert rule reasoning unit employs a case-based reasoning (CBR) mechanism. Specifically, the system extracts feature vectors for newly added configuration items, such as sheet thickness, number of openings, and edge banding material, and calculates their Euclidean distance to the feature vectors of historical cases in the process rule base to determine similarity. The historical case with the highest similarity is selected as the basis, and combined with preset material property correction coefficients and dimensional tolerance rules, the historical process parameters are fine-tuned to generate a temporary process path. This ensures that the generated process path still has high executability even without direct historical data.

[0029] Based on the processing capacity of currently available equipment, fixture compatibility, and historical processing cycle data, a list of candidate equipment is assigned to each optional processing step.

[0030] It should be noted that processing capacity is the core foundation for equipment to meet the process requirements, covering the processing range, accuracy, process type, and key operating parameters, which must be fully matched with the order process parameters; fixture compatibility refers to whether the existing fixtures of the equipment can stably fix the workpiece in terms of specifications and type, and can be used without additional adjustments; historical processing cycle data is the average time spent per piece in the past on similar equipment for this process, reflecting processing efficiency and stability. The three together provide key basis for the selection of candidate equipment, ensuring equipment compatibility and production efficiency.

[0031] It should be noted that the specific process for allocating the candidate equipment list is as follows: Analyze the process requirements such as machining accuracy and material compatibility of the order's operations, as well as the basic conditions such as the current operating status of the equipment and fixture compatibility; then calculate the process matching degree and quantitative indicators such as comprehensive efficiency (OEE) and average machining cycle time; compare the hard indicator fit, efficiency data, and subsequent task saturation of each piece of equipment to form an equipment tier; prioritize the equipment with a perfect process matching degree, OEE greater than a preset threshold, and the shortest machining cycle time. If multiple equipment have similar efficiency, combine order delivery time and load balancing decisions to finally determine the optimal allocation result, providing a precise range of equipment selection for scheduling optimization.

[0032] For process nodes with multiple path choices, a process network structure in the form of a directed acyclic graph is constructed, where each path in the directed acyclic graph corresponds to a feasible processing sequence.

[0033] It's important to note that in flexible cabinet manufacturing, some process nodes offer multiple feasible processing options, forming multi-path processing solutions. A directed acyclic graph (DAG) represents specific processes with nodes and uses directed edges to illustrate the logical dependencies between processes, ensuring no cyclic dependencies. Each complete path from the starting process node to the ending process node corresponds to a coherent and feasible processing sequence, clearly presenting the optional execution paths for each process. This provides structured support for subsequently selecting the optimal process route based on capacity and delivery time.

[0034] Preferably, for the edge banding process, if there are three edge banding machines on the production line, two of which support PVC edge banding and have automatic edge trimming functions, and the other only supports ABS edge banding, the system will automatically select the appropriate equipment based on the edge banding material selected in the order. For process nodes with multiple path options, such as drilling processes that can be completed in a CNC machining center or on a dedicated six-sided drilling machine, the system will create multi-path branch points in the directed acyclic graph structure and connect the upstream sheet metal processing process node with the downstream hardware assembly process node through process constraint edges.

[0035] It should be noted that the directed acyclic graph not only includes the logical dependencies between processes, but also accesses the delivery date information of orders through the delivery date coupling interface, and marks the key equipment with limited current capacity through bottleneck resource labeling, thereby providing a path scoring basis for subsequent scheduling.

[0036] The process network structure is coupled with order delivery deadlines, priority weights, and capacity bottleneck information to generate a dynamic process route model that supports scheduling engine calls.

[0037] It should be noted that priority weight is a quantitative indicator set up to adapt to the non-standard flexible production of cabinets and support the generation of dynamic process route models, used to reflect the importance of orders. Its value is calculated based on a comprehensive analysis of customer contract terms, historical performance records, and order amounts, and can dynamically reflect the priority differences of different orders. Coupled with process network structure, order delivery deadlines, and capacity bottleneck information, it can serve as a key scoring criterion in multi-path process screening, helping the system prioritize processing paths that meet the needs of high-priority orders and reduce the risk of delays, providing accurate decision support for the scheduling engine.

[0038] It should be noted that, firstly, by combining the expedited clauses in the contract, historical performance records, and order amounts, the order priority weight is quantified. Then, based on the correlation between the estimated total occupancy time of each path in the process network on bottleneck equipment and the order delivery deadline, a delay risk index is determined. Subsequently, reasonable weightings for the two types of indicators are set, and a comprehensive score for each feasible path is calculated by positively incorporating the priority weight and negatively incorporating the delay risk index. After traversing and sorting, the optimal path and several high-scoring paths are selected and integrated to form a structured dynamic process route model for use by the scheduling engine.

[0039] In a preferred embodiment of the present invention, the construction and maintenance of the process rule base includes: collecting process execution records of historical successful orders and extracting the mapping relationship between configuration items and actual adopted processes.

[0040] For each configuration combination, the average processing cycle time, pass rate, and number of fixture changes on different equipment are statistically analyzed.

[0041] It should be noted that different combinations of customized configuration items correspond to different processing requirements. Statistical average processing cycle time can clarify the equipment's processing efficiency, the pass rate can reflect the processing quality adaptability, and the number of fixture changes is related to production changeover costs. These three factors together constitute the core evaluation dimensions of equipment adaptability. Through this data, optional processes can be assigned accurate adaptability scores, reasonable activation thresholds can be set, and temporary process paths for newly added configuration item combinations can be provided as a reference. This ensures that the processes and equipment matched to the process rule library are scientifically feasible, supports the accurate construction of dynamic process route models, and improves scheduling and production efficiency.

[0042] Based on the statistical results, each selectable processing step is assigned an equipment compatibility score, and a step activation threshold is set.

[0043] It should be noted that a comprehensive score is assigned based on core screening dimensions such as processing capacity, fixture adaptability, historical processing cycle time, and pass rate. Processing capacity is the basic hard indicator; if it fully matches the corresponding process requirements, a base score is awarded, and if it does not match, it is directly eliminated. Then, a tiered scoring system is applied to fixture adaptability, including whether there are additional adjustment requirements, the efficiency of historical processing cycle time, and the quality of the pass rate. Finally, a comprehensive calculation is performed to determine the adaptability score of each candidate equipment for the corresponding selectable process. The higher the score, the better the adaptability of the equipment to the process and the better the overall production efficiency.

[0044] When there is no historical data to support the newly added configuration item combination, the expert rule reasoning unit is activated. Based on the similar case retrieval and parameter correction algorithm, the feature similarity between the new configuration item and the historical case is calculated. The historical process path with the highest similarity is called and a temporary process path is generated according to the material properties, geometric dimensions and functional requirements.

[0045] The actual performance of temporary process paths is regularly incorporated into the process rule base to complete the closed-loop iterative update of the rule base.

[0046] In a preferred embodiment of the present invention, generating a dynamic process route model that supports scheduling engine calls includes: assigning a dynamic priority weight value to each order based on customer contract terms, historical performance records, and order amount.

[0047] Preferably, the dynamic priority weight value is set in the following way: ,in Indicates the dynamic priority weight value. Indicates the contract grade coefficient. This indicates the average delay rate of the corresponding customer's historical orders. This represents the normalized order amount coefficient. This represents the preset weighting factor.

[0048] Identify equipment units in the current production line that are operating at high load and whose theoretical processing time exceeds the conventional threshold, and mark them as capacity bottleneck resources.

[0049] It should be noted that the specific method for identifying high-load equipment units on the production line is as follows: The real-time workload of each piece of equipment and the proportion of pending processes are statistically analyzed. Combined with the equipment's rated processing capacity, the theoretical completion time of existing tasks is calculated. Simultaneously, the continuous running time and real-time load status of the equipment are captured and compared with the equipment's normal operating load benchmark for comprehensive judgment. Equipment with saturated workload and theoretical processing time exceeding the normal threshold is recorded as high-load equipment units on the production line.

[0050] Furthermore, determining whether the workload is saturated requires statistical analysis of the ratio of work-in-process tasks currently being handled by the equipment to the number of tasks awaiting scheduling. Combined with the equipment's rated processing capacity, the theoretical completion time of all existing tasks is calculated. If this theoretical completion time reaches the equipment's maximum available working time within the current scheduling cycle, then the workload is determined to be saturated.

[0051] The setting of conventional thresholds is based on the historical operating data of the equipment. The average processing cycle time of the equipment under normal production conditions and the range of task processing time when running at full load are statistically analyzed. Combined with the overall production line capacity planning, order delivery deadline requirements and equipment maintenance cycle, the conventional threshold of theoretical processing time is comprehensively determined to ensure that the threshold not only meets the actual operating capacity of the equipment, but also adapts to the reasonable needs of production scheduling.

[0052] For each feasible path in the process network, the estimated time occupied by each process on the bottleneck equipment is accumulated, and the delay risk index of the path is calculated in conjunction with the order delivery deadline.

[0053] It should be noted that the formula for calculating the delay risk index is as follows: ,in This indicates the risk index of delay. This indicates the estimated total time spent by all processes in the process path using the bottleneck equipment. This indicates the remaining delivery deadline for the order.

[0054] Multiple feasible paths are ranked and filtered based on a weighted score of priority and delay risk index.

[0055] It should be noted that, combining the aforementioned two core indicators, priority weight and delay risk index, a corresponding weighted comprehensive score is calculated for each feasible processing path in the process network. The score is calculated by positively weighting the order priority weight and negatively weighting the delay risk index. Based on the calculated comprehensive score, all feasible processing paths are sorted in descending order, and the path with the best comprehensive score is selected first. At the same time, several high-scoring high-quality paths are screened out to complete the path sorting and screening work, thereby determining the optimal process route that meets the order requirements.

[0056] It should be noted that the above weighting is based on historical data statistics or dynamic adjustments using machine learning, and is grounded in the following principles: prioritizing the core needs of production scheduling, ensuring the delivery of high-priority orders while strictly controlling production delay risks, and balancing the dual dimensions of order importance and delivery feasibility. The priority weight is set according to the core impact of order attributes, aligning with the priority differences in order contract agreements, cooperation levels, and output value; the delay risk index is set according to the key constraints of capacity bottlenecks on delivery, aligning with actual production line capacity constraints and delivery fulfillment requirements. The weighting of these two factors is comprehensively determined in conjunction with actual production scheduling needs, ensuring that the weighted score scientifically reflects the path suitability value and supports the rationality of path ranking and selection.

[0057] Finally, the single path with the best score is selected and the top N high-scoring paths are retained as the dynamic process route model for this order, so that the scheduling engine can make path selection and scheduling decisions when there is resource competition.

[0058] Preferably, the system ultimately selects a score. The longest path is used as the main process route, in which This indicates the preset weight.

[0059] The real-time status sensing module collects information on workshop equipment status, work-in-process location, personnel work progress, and material inventory in real time, forming a production line operation status dataset.

[0060] In a preferred embodiment of the present invention, the formation of the production line operation status dataset includes: collecting equipment operation mode, spindle load, alarm codes and energy consumption data at a preset frequency through IoT sensors and PLC controllers deployed on the production equipment.

[0061] RFID positioning tags are used to track the real-time coordinates of work-in-process pallets or tooling fixtures in the workshop, and the coordinates are mapped to specific workstation areas in a pre-calibrated digital workshop coordinate system.

[0062] The system receives operator confirmations for starting, pausing, completing, and reporting abnormalities via mobile or workstation terminals, generating a workflow event stream for personnel tasks.

[0063] The real-time inventory quantity, storage location distribution, and quality inspection status of each material code are synchronized periodically from the warehouse management system interface.

[0064] After aligning the aforementioned multi-source heterogeneous data with timestamps, cleaning the data, and fusion the semantics, a unified production line operation status dataset is constructed, which serves as the real-time input basis for scheduling and material collaborative optimization.

[0065] It should be noted that in the MES system, data cleaning first involves filtering outliers, filling in missing items, removing duplicate data, and correcting data formats and errors from heterogeneous data collected from multiple sources such as IoT sensors, RFID positioning, workstation terminals, and warehouse system interfaces, including equipment, work-in-process, personnel, and materials. Semantic fusion, on the other hand, is based on preset semantic rules for the production domain. It aligns the cleaned data by timestamp, transforms data from different sources and formats into a unified semantic standard, and realizes logical connections and semantic consistency between data. Ultimately, it forms a production line operation status dataset with standardized structure and consistent semantics.

[0066] The rolling scheduling optimization module generates an initial production scheduling plan that includes process start and end times, equipment allocation, and personnel scheduling based on the dynamic process route model and the production line operation status dataset.

[0067] In a preferred embodiment of the present invention, the generation of the initial production scheduling scheme specifically includes: dividing the scheduling period into multiple consecutive rolling time windows, each time window covering a set of schedulable orders within a preset future duration.

[0068] Within the current time window, a mixed integer programming model is constructed with the goal of minimizing order weighted delay, equipment idle time, and personnel switching costs as multiple objective functions.

[0069] It should be noted that the construction method of the mixed integer programming model is as follows: A decision variable system is defined, comprising 0-1 variables including process-equipment / personnel allocation decisions and continuous variables including process start and end times and delay durations; then, with the goal of minimizing order weighted delays, equipment idle time, and personnel switching costs, a single objective function is constructed through weighted summation; subsequently, hard constraints such as process logical dependencies, equipment candidate list constraints, equipment capacity limitations, and personnel skill adaptation are embedded to form a complete linear programming expression; finally, combining the process parameters provided by the dynamic process modeling module and the production line data collected by the real-time status perception module, the mathematical programming solution engine solves the problem within a limited time within a rolling time window, outputting the optimal scheduling decision adapted to the current production scenario.

[0070] Preferably, the multi-objective function is: ,in Indicates order The completion time of the last process, For delivery time, For equipment The total amount of free time, For personnel Switching costs incurred by moving between workstations This represents a pre-set weighting factor.

[0071] It should be noted that the personnel switching cost is specifically defined as: the theoretical movement time required for the same operator to move from the current workstation to the next assigned workstation, plus the skill adaptation or preparation time due to differences in workstation equipment. By minimizing this total time, the wasted time of personnel on non-value-adding activities can be effectively reduced. The criteria for determining capacity bottleneck resources are: real-time monitoring of the task queues of each piece of equipment; when the estimated total processing time of a certain piece of equipment exceeds 80% of the available time of that equipment in the current scheduling cycle, it is determined to be a capacity bottleneck resource.

[0072] The process dependencies, equipment candidate list, and capacity constraints in the dynamic process route model are embedded as hard constraints in the solver.

[0073] Preferably, hard constraints are strictly embedded in the process logic dependencies of the dynamic process route input, such as edge banding must begin after material cutting. Equipment candidate list restrictions limit a process to pre-selected equipment. Equipment capacity limits include a maximum daily operating time of 16 hours or less for a single machine, and personnel skill matching ensures only certified personnel can operate CNC equipment.

[0074] The mathematical programming engine is invoked to solve for the optimal process allocation scheme within the current window within a limited computation time.

[0075] Preferably, in order to meet the real-time requirements, a rolling time-domain control strategy is adopted, which limits the solution window to N processes, and a genetic algorithm is used to approximate the solution of the MIP model.

[0076] The current window scheduling result is solidified and pushed to the execution layer, while rolling forward to the next time window, repeating the above modeling and solving process to achieve dynamic rolling updates of the scheduling scheme.

[0077] For the material collaborative delivery module, please refer to [link / reference]. Figure 2 As shown, based on the material consumption sequence and quantity of each process in the initial production scheduling plan, material completeness verification is performed, and material delivery instructions synchronized with the process execution rhythm are generated.

[0078] In a preferred embodiment of the present invention, the generation of material delivery instructions synchronized with the process execution rhythm includes: calculating the theoretical consumption time window of each type of material at each workstation based on the planned start time and duration of each process in the initial production scheduling scheme.

[0079] By combining current material inventory levels, in-transit purchase orders, and warehouse location information, the availability status of each material within the corresponding time window can be determined.

[0080] It should be noted that determining the availability status of each material within a corresponding time window involves a comprehensive assessment of three core pieces of information: first, the current actual inventory level of the material, clarifying the amount of material readily available; second, the delivery cycle and quantity of inbound purchase orders, confirming the amount of material that can be replenished within the specified time window; and third, the existing storage location and transfer time of the material, clarifying the ease of retrieval and arrival time of the inventory material. After integrating these three types of information, the material demand and timeliness requirements of the orders within the corresponding time window are compared to ultimately determine the availability status of the material: whether it is sufficient, available on time, or requires replenishment.

[0081] For unavailable materials, initiate a material shortage warning and trigger an alternative material verification process. If an alternative solution exists, update the BOM mapping relationship.

[0082] It needs to be explained that unavailable materials refer to materials that cannot meet the process's requisition requirements on time and in sufficient quantity within the corresponding time window of the production order. These materials may be due to insufficient current inventory, the arrival cycle of purchased materials exceeding the production demand timeline, or the warehouse transfer timeline not matching the process material delivery node. In all these cases, the materials cannot be requisitioned and put into production within the specified production period. This is a core negative category in determining the availability of materials and a type of material that needs to be avoided and adjusted in production scheduling.

[0083] Based on the material completeness verification results, a material delivery instruction set is generated, which includes material code, delivery quantity, target workstation, expected delivery time, and delivery priority.

[0084] The material distribution instruction set is sent to the warehouse logistics control system, and a linkage update mechanism is established with the scheduling plan changes.

[0085] In a preferred embodiment of the present invention, the step of initiating a material shortage warning and triggering a substitute material verification process includes: when the available inventory of a certain type of material within the corresponding process consumption time window is lower than the quantity required by the process, the system automatically marks the material as being in a material shortage state and records the material shortage level.

[0086] It should be noted that the material shortage level is a classification of the degree, scope, and urgency of the shortage of unavailable materials within the corresponding time window of a production order. It is a core dimension for quantitatively assessing material shortage risk. It is comprehensively determined based on the proportion of the material shortage quantity to the total process demand, the criticality of the process corresponding to the shortage material, the duration of production downtime caused by the shortage, and the impact on order delivery. A higher level indicates a more severe material shortage and a greater negative impact on production line scheduling and order delivery, providing a clear classification basis for subsequent material shortage warnings, replenishment priority ranking, and production plan adjustments.

[0087] Based on a pre-defined material substitution rule base, a set of alternative material codes that have the same functional attributes, size compatibility, and surface treatment matching degree as the material in short supply are retrieved.

[0088] It should be noted that the material substitution rule base is a standardized set of rules that integrates various material substitution-related information. Its core includes a list of compliant substitute materials for the main materials required for production, the appropriate process requirements for each substitute material, the matching degree of its physicochemical properties, the standards for adjusting processing parameters, and the basis for prioritizing substitutions. It also clearly defines the activation conditions and limiting scenarios for substitute materials. This rule base provides a compliant basis for rapid material replacement in material shortage scenarios, directly matching order process requirements to determine substitution feasibility, ensuring smooth production process continuity during material shortages, avoiding production stoppages, and serving as the core rule system supporting dynamic scheduling and flexible production.

[0089] For each candidate alternative material, verify its current inventory level, quality status, and whether it is already bound to other high-priority orders.

[0090] If a material that meets the substitution conditions exists, generate a material substitution suggestion and send it to the management terminal for confirmation. Upon receiving the confirmation instruction, update the mapping relationship of the material in the current order BOM and recalculate the material availability status of the process.

[0091] Specifically, given the stringent requirements for material consistency in customized cabinet production, when the system finds a substitute material, such as using hinges of the same specifications from brand B to replace a stock-out item from brand A, it does not directly modify the BOM. Instead, it sends a pop-up notification or message to the terminal in the production control center. Only after a planner or authorized manager confirms the substitution on the terminal will the system update the BOM and recalculate the schedule. This human-computer interaction mechanism effectively avoids the risk of customer order cancellations due to automatic substitution.

[0092] The alternative decision results, along with the original material shortage information, are recorded in the material collaboration log for subsequent scheduling re-optimization and supply chain replenishment strategy adjustments.

[0093] The disturbance response rescheduling module triggers a scheduling re-optimization mechanism when it detects equipment failure, emergency order insertion, or rework disturbance events. It synchronously updates the dynamic process route model and material delivery instructions, and outputs a collaborative scheduling scheme after the disturbance response.

[0094] In a preferred embodiment of the present invention, the triggering scheduling re-optimization mechanism includes: monitoring the real-time event flow in the workshop and identifying three types of disturbance events: equipment downtime exceeding the threshold, new order insertion, or work-in-process rework.

[0095] It should be noted that the scheduling re-optimization trigger mechanism is to continuously monitor the real-time production event flow in the workshop and accurately identify three types of core production disturbance events: first, abnormal downtime events where the equipment downtime exceeds the set threshold; second, scheduling change events where new orders are urgently inserted during the production process; and third, process abnormal events where work-in-process is defective and needs to be reworked, thus triggering scheduling re-optimization.

[0096] For each type of disturbance event, extract the set of affected processes and their upstream and downstream related processes, and define the scope of rescheduling.

[0097] It should be noted that, using the core affected process triggered by the production disturbance as the benchmark anchor point, the upstream and downstream related processes are determined using the process logic adjacency determination method. Upstream related processes are all processes that must be completed before the core process and are directly preceding processes in terms of technology, including only direct precursor processes without other process intervals. Downstream related processes are all processes that immediately follow the core process after its completion and are directly following processes in terms of technology, including only direct successor processes without other process intervals. Upstream and downstream processes defined in this way are all adjacent processes strongly related to the core process, with no redundant processes included, accurately defining the process impact boundaries of the disturbance.

[0098] Freeze the allocated resources and time windows of undisturbed processes, and re-execute scheduling optimization calculations only for processes within the scope.

[0099] During the re-optimization process, the delivery time window for relevant materials is adjusted simultaneously, and new material requirements caused by scheduling changes are verified.

[0100] The output includes a disturbance response collaborative scheduling scheme that includes the updated process schedule, equipment reallocation results, and revised material delivery instructions, and is pushed to the execution layer control system.

[0101] In a preferred embodiment of the present invention, freezing the allocated resources and time windows of the unaffected processes and re-performing the scheduling optimization calculation only for processes within the scope includes: identifying the directly affected process nodes by the disturbance event, and determining the set of all logically related affected processes by traversing the dynamic process route model in both forward and backward directions.

[0102] Processes not included in this set are marked as frozen, while their original equipment allocation, personnel assignments, and time windows remain unchanged.

[0103] For the set of affected processes, reload its dynamic process subnetwork and construct a local optimization problem by combining the current real-time equipment availability, personnel on-duty status and material availability.

[0104] During local optimization, the end time of the frozen process must be restricted to be no earlier than its originally planned start time in order to avoid resource conflicts.

[0105] After solving the local scheduling model, only the new scheduling parameters of the affected processes are output, and the relevant material delivery instructions and personnel task queues are updated simultaneously.

[0106] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, and all such modifications or additions should fall within the protection scope of the present invention.

Claims

1. A smart scheduling and material collaborative optimization MES system for a flexible cabinet production line, characterized in that: include: The dynamic process modeling module constructs a dynamic process route model for cabinet orders based on order configuration parameters, product structure BOM, and equipment capacity constraints. The real-time status perception module collects information on workshop equipment status, work-in-process location, personnel work progress, and material inventory in real time, forming a production line operation status dataset. The rolling scheduling optimization module generates an initial production scheduling plan that includes process start and end times, equipment allocation, and personnel scheduling based on the dynamic process route model and the production line operation status dataset. The material collaborative delivery module performs material completeness verification based on the material consumption sequence and quantity of each process in the initial production scheduling plan, and generates material delivery instructions that are synchronized with the process execution rhythm. The disturbance response rescheduling module triggers a scheduling re-optimization mechanism when it detects equipment failure, emergency order insertion, or rework disturbance events. It synchronously updates the dynamic process route model and material delivery instructions, and outputs a collaborative scheduling scheme after the disturbance response.

2. The intelligent scheduling and material collaborative optimization MES system for a flexible cabinet production line according to claim 1, characterized in that: The dynamic process route model for constructing cabinet orders includes: Analyze the customized configuration items in customer orders to extract sheet material type, hardware specifications, functional module combinations, and surface treatment process parameters; Based on a preset process rule library, the set of optional processing steps corresponding to the configuration item is matched, and logical dependencies between the steps are established. Based on the processing capacity of currently available equipment, fixture compatibility, and historical processing cycle data, a list of candidate equipment is assigned to each optional processing step. For process nodes with multiple path choices, a process network structure in the form of a directed acyclic graph is constructed, wherein each path in the directed acyclic graph corresponds to a feasible processing sequence; The process network structure is coupled with order delivery deadlines, priority weights, and capacity bottleneck information to generate a dynamic process route model that supports scheduling engine calls.

3. The intelligent scheduling and material collaborative optimization MES system for a flexible cabinet production line according to claim 2, characterized in that: The construction and maintenance of the process rule base includes: Collect process execution records from historical successful orders and extract the mapping relationship between configuration items and actual processes used; For each type of configuration item combination, the average processing cycle time, pass rate and fixture changeover times on different equipment are statistically analyzed; Based on the statistical results, each optional processing step is assigned an equipment compatibility score, and a step activation threshold is set. When there is no historical data to support the newly added configuration item combination, the expert rule reasoning unit is activated. Based on the similar case retrieval and parameter correction algorithm, the feature similarity between the new configuration item and the historical case is calculated. The historical process path with the highest similarity is called and a temporary process path is generated based on material properties, geometric dimensions and functional requirements. The actual performance of temporary process paths is regularly incorporated into the process rule base to complete the closed-loop iterative update of the rule base.

4. The intelligent scheduling and material collaborative optimization MES system for a flexible cabinet production line according to claim 2, characterized in that: The generation of a dynamic process route model that supports scheduling engine calls includes: Each order is assigned a dynamic priority weight value calculated based on customer contract terms, historical performance records, and order amount; Identify equipment units in the current production line that are operating at high load and whose theoretical processing time exceeds the conventional threshold, and mark them as capacity bottleneck resources. On each feasible path of the process network, the estimated time occupied by each process on the bottleneck equipment is accumulated, and the delay risk index of the path is calculated in combination with the order delivery deadline. Multiple feasible paths are ranked and filtered based on a weighted score of priority weight and delay risk index; Finally, the single path with the best score is selected and the top N high-scoring paths are retained as the dynamic process route model for this order, so that the scheduling engine can make path selection and scheduling decisions when there is resource competition.

5. The intelligent scheduling and material collaborative optimization MES system for a flexible cabinet production line according to claim 1, characterized in that: The data set for forming the production line operation status includes: By deploying IoT sensors and PLC controllers on production equipment, the system collects equipment operating modes, spindle loads, alarm codes, and energy consumption data at preset frequencies. RFID positioning tags are used to track the real-time coordinates of work-in-process pallets or tooling fixtures in the workshop, and the coordinates are mapped to specific workstation areas in a pre-calibrated digital workshop coordinate system. The system receives operator confirmations for starting, pausing, completing, and reporting abnormalities via mobile or workstation terminals, and generates a workflow event stream for personnel work progress. The real-time inventory quantity, storage location distribution, and quality inspection status of each material code are synchronized periodically from the warehouse management system interface. After aligning the aforementioned multi-source heterogeneous data with timestamps, cleaning the data, and fusion the semantics, a unified production line operation status dataset is constructed, which serves as the real-time input basis for scheduling and material collaborative optimization.

6. The intelligent scheduling and material collaborative optimization MES system for a flexible cabinet production line according to claim 1, characterized in that: The generation of the initial production scheduling plan specifically includes: The scheduling period is divided into multiple consecutive rolling time windows, each time window covering a set of schedulable orders within a preset future duration; Within the current time window, a mixed integer programming model is constructed with the multiple objective functions of minimizing order weighted delay, equipment idle time, and personnel switching costs. The process dependencies, equipment candidate list and capacity constraints in the dynamic process route model are embedded as hard constraints in the solver. Call the mathematical programming engine to find the optimal process allocation scheme within the current window within a limited computation time; The current window scheduling result is solidified and pushed to the execution layer, while rolling forward to the next time window, repeating the above modeling and solving process to achieve dynamic rolling updates of the scheduling scheme.

7. The intelligent scheduling and material collaborative optimization MES system for a flexible cabinet production line according to claim 1, characterized in that: The generation of material delivery instructions synchronized with the process execution rhythm includes: Based on the planned start time and duration of each process in the initial production scheduling scheme, calculate the theoretical consumption time window for each type of material at each workstation; Based on the current material inventory level, in-transit purchase orders, and warehouse location information, determine the availability status of each material within the corresponding time window; For unavailable materials, initiate a material shortage warning and trigger the alternative material verification process. If an alternative solution exists, update the BOM mapping relationship. Based on the material completeness verification results, a material delivery instruction set is generated, which includes material code, delivery quantity, target workstation, expected delivery time and delivery priority. The material distribution instruction set is sent to the warehouse logistics control system, and a linkage update mechanism is established with the scheduling plan changes.

8. The intelligent scheduling and material collaborative optimization MES system for a flexible cabinet production line according to claim 7, characterized in that: The process of initiating a material shortage warning and triggering a substitute material verification includes: When the available inventory of a certain type of material is lower than the quantity required by the corresponding process within the consumption time window, the system automatically marks the material as being in a shortage state and records the shortage level. Based on a pre-defined material substitution rule base, retrieve a set of alternative material codes that have the same functional attributes, size compatibility, and surface treatment matching degree as the material in short supply. For each candidate alternative material, verify its current inventory level, quality status, and whether it is already bound to other high-priority orders; If there are materials that meet the substitution conditions, generate material substitution suggestion information and send it to the management terminal for confirmation. After receiving the confirmation instruction, update the mapping relationship of the material in the current order BOM and recalculate the material availability status of the process. The alternative decision results, along with the original material shortage information, are recorded in the material collaboration log for subsequent scheduling re-optimization and supply chain replenishment strategy adjustments.

9. The intelligent scheduling and material collaborative optimization MES system for a flexible cabinet production line according to claim 1, characterized in that: The triggering schedule re-optimization mechanism includes: Monitor the real-time event flow in the workshop and identify three types of disturbance events: equipment downtime exceeding threshold, new order insertion, or work-in-process repair. For each type of disturbance event, extract the set of affected processes and their upstream and downstream related processes, and define the scope of rescheduling; Freeze the allocated resources and time windows of unaffected processes, and re-execute scheduling optimization calculations only for processes within the scope; During the re-optimization process, the delivery time window for relevant materials is adjusted simultaneously, and the new material requirements caused by scheduling changes are verified. The output includes a disturbance response collaborative scheduling scheme that includes the updated process schedule, equipment reallocation results, and revised material delivery instructions, and is pushed to the execution layer control system.

10. The intelligent scheduling and material collaborative optimization MES system for a flexible cabinet production line according to claim 9, characterized in that: The freezing of allocated resources and time windows for undisturbed processes involves re-performing scheduling optimization calculations only for processes within the scope, including: Identify the directly affected process nodes by disturbance events, and determine the set of all logically related affected processes by traversing the dynamic process route model in both forward and backward directions. Mark processes not included in this set as frozen, while retaining their original equipment allocation, personnel assignment, and time window unchanged; For the set of affected processes, reload its dynamic process subnetwork and construct a local optimization problem by combining the current real-time equipment availability, personnel on-duty status and material arrival status; During local optimization, the end time of the frozen process must be restricted to be no earlier than its originally planned start time in order to avoid resource conflicts. After solving the local scheduling model, only the new scheduling parameters of the affected processes are output, and the relevant material delivery instructions and personnel task queues are updated simultaneously.

Images