Optimization-based nesting method and system based on mixed integer linear programming

Abstract

This application discloses an optimized material cutting method and system, storage medium, and computer equipment based on mixed-integer linear programming. The method includes: acquiring the raw data required for material cutting, including production order information, raw material inventory information, and key process parameters; extracting the net size, required quantity, and delivery date information of the product to be processed from the production order information, and correcting the net size based on the equivalent processing allowance in the key process parameters to obtain the material cutting requirement size for optimization calculation; constructing a multi-objective mixed-integer linear programming model with minimizing raw material cost as the first objective and minimizing the weighted delivery overdue time as the second objective based on the material cutting requirement size, required quantity, delivery date information, raw material inventory information, and processing loss in the key process parameters; solving the multi-objective mixed-integer linear programming model to generate an optimized material cutting scheme, and updating the corresponding raw material inventory status based on the confirmation instruction of the scheme.

Description

Technical Field

[0001] This application relates to the field of optimized material cutting technology, and in particular to an optimized material cutting method and system based on mixed integer linear programming, a storage medium, and a computer device. Background Technology

[0002] In manufacturing, especially discrete manufacturing, the efficient utilization of raw materials is crucial for controlling costs and improving efficiency. The problem of optimizing the cutting of one-dimensional materials such as bars and profiles is widespread in industries like steel structures, machining, and furniture production. Traditional cutting processes heavily rely on worker experience, using manual calculations or simple tools for layout. This is not only inefficient but also makes it difficult to achieve optimal raw material utilization and order delivery times under complex conditions of multiple orders and constraints, leading to significant material waste and inflexible production scheduling.

[0003] Currently, the industry has introduced operations research optimization techniques to address this challenge, such as using linear programming or heuristic algorithms to generate material cutting schemes. However, these existing methods still have significant limitations in practical applications: on the one hand, their mathematical models often fail to fully integrate the complex practical constraints and process experience of the production site, resulting in overly theoretical optimized schemes that may not be executable or may cause hidden waste during actual cutting; on the other hand, existing algorithms struggle to simultaneously consider multiple objectives, leading to a situation where one aspect is neglected when dealing with multiple interdependent factors such as material utilization, and thus failing to generate a truly optimal global solution in terms of overall benefits. Summary of the Invention

[0004] In view of this, this application provides an optimized material cutting method and system, storage medium, and computer equipment based on mixed-integer linear programming. By introducing key process parameters such as equivalent processing allowance and processing loss, the theoretical required size is corrected, making the multi-objective mixed-integer linear programming model accurately fit the complex process reality, ensuring the feasibility of the optimized material cutting scheme and the material utilization rate. At the same time, the multi-objective mixed-integer linear programming model with cost and delivery time as its core achieves synergistic consideration and comprehensive optimization of dual optimization objectives through mathematical weighted coordination. The closed-loop design linking the optimized material cutting scheme confirmation with inventory status realizes automatic synchronization from optimization results to production execution, effectively avoiding the risk of raw material reuse. The embodiments of this application solve the pain points of existing technologies where the model is out of touch with reality and the objectives are not balanced. It can automatically and quickly generate a globally optimal material cutting plan that takes into account both material utilization rate and order delivery timeliness, freeing personnel from heavy manual calculations and realizing the automation, intelligence, and reliability of production scheduling optimization.

[0005] According to one aspect of this application, an optimized material cutting method based on mixed-integer linear programming is provided, comprising: Obtain the raw data required for material cutting, wherein the raw data includes production order information, raw material inventory information and key process parameters from external systems; Extract the net dimensions, required quantity, and delivery date information of the products to be processed from the production order information. Correct the net dimensions based on the equivalent processing allowance in the key process parameters to obtain the blanking requirement dimensions for optimization calculation. Based on the required material dimensions, required quantity, delivery date information, raw material inventory information, and processing loss in the key process parameters, a multi-objective mixed integer linear programming model is constructed with the first objective of minimizing raw material cost and the second objective of minimizing weighted delivery overdue time. The multi-objective mixed integer linear programming model is solved to generate an optimized material cutting scheme, and the corresponding raw material inventory status is updated based on the confirmation instruction of the optimized material cutting scheme.

[0006] According to another aspect of this application, an optimized material feeding system based on mixed-integer linear programming is provided, comprising: The data acquisition module is used to acquire the raw data required for material cutting, wherein the raw data includes production order information, raw material inventory information and key process parameters from external systems; The size correction module is used to extract the net size, required quantity and delivery date information of the product to be processed from the production order information, and correct the net size based on the equivalent processing allowance in the key process parameters to obtain the blanking requirement size for optimization calculation. The model building module is used to construct a multi-objective mixed integer linear programming model based on the material cutting requirement size, the required quantity, the delivery date information, the raw material inventory information, and the processing loss in the key process parameters. The model aims to minimize the raw material cost as the first objective and minimize the weighted delivery overdue time as the second objective. The solution module is used to solve the multi-objective mixed integer linear programming model, generate an optimized material cutting scheme, and update the corresponding raw material inventory status based on the confirmation instruction of the optimized material cutting scheme.

[0007] According to another aspect of this application, a storage medium is provided that stores a computer program thereon, which, when executed by a processor, implements the above-described optimized material cutting method based on mixed-integer linear programming.

[0008] According to another aspect of this application, a computer device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor executes the program to implement the above-described optimized material cutting method based on mixed integer linear programming.

[0009] By employing the aforementioned technical solutions, this application provides an optimized material cutting method and system, storage medium, and computer equipment based on mixed-integer linear programming. By introducing key process parameters such as equivalent processing allowance and processing loss to correct the theoretical required dimensions, the multi-objective mixed-integer linear programming model accurately fits the complex process reality, ensuring the feasibility of the optimized material cutting scheme and material utilization rate. Simultaneously, the multi-objective mixed-integer linear programming model, with cost and delivery time as its core, achieves synergistic consideration and comprehensive optimization of dual optimization objectives through mathematical weighted coordination. The closed-loop design linking the optimized material cutting scheme confirmation with inventory status enables automatic synchronization from optimization results to production execution, effectively avoiding the risk of raw material reuse. The embodiments of this application solve the pain points of existing technologies where models are detached from reality and objectives are not balanced. It can automatically and quickly generate a globally optimal material cutting plan that balances material utilization and order delivery timeliness, freeing personnel from heavy manual calculations and achieving automation, intelligence, and reliability in production scheduling optimization.

[0010] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0011] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 A flowchart illustrating an optimized material cutting method based on mixed integer linear programming provided in an embodiment of this application is shown. Figure 2 A flowchart illustrating another optimized material cutting method based on mixed integer linear programming provided in an embodiment of this application is shown. Figure 3 This illustration shows a schematic diagram of an optimized feeding system based on mixed integer linear programming provided in an embodiment of this application; Figure 4 A schematic diagram of the device structure of a computer device provided in an embodiment of this application is shown. Detailed Implementation

[0012] The present application will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of the present application can be combined with each other.

[0013] This embodiment provides an optimized material cutting method based on mixed-integer linear programming, such as... Figure 1 As shown, the method includes: Step 101: Obtain the raw data required for material cutting, wherein the raw data includes production order information, raw material inventory information and key process parameters from external systems.

[0014] Step 102: Extract the net dimensions, required quantity, and delivery date information of the products to be processed from the production order information. Correct the net dimensions based on the equivalent processing allowance in the key process parameters to obtain the blanking requirement dimensions for optimization calculation.

[0015] Step 103: Based on the required material dimensions, required quantity, delivery date information, raw material inventory information, and processing loss in the key process parameters, construct a multi-objective mixed integer linear programming model with the first objective of minimizing raw material cost and the second objective of minimizing weighted delivery delay time.

[0016] Step 104: Solve the multi-objective mixed integer linear programming model to generate an optimized material cutting scheme, and update the corresponding raw material inventory status based on the confirmation instruction of the optimized material cutting scheme.

[0017] This application provides an optimized material cutting method based on mixed-integer linear programming. First, it automatically acquires the core raw data required for optimized material cutting from an external system such as an existing enterprise production management system, including ERP (Enterprise Resource Planning), WMS (Warehouse Management System), or MES (Manufacturing Execution System). This data mainly includes three parts: production order information that specifically describes "what to produce," raw material inventory information reflecting "what materials are available," and key process parameters defining "how to process."

[0018] Next, the acquired production order information is analyzed to extract three key attributes for each product to be processed: the final net size required by the customer, the required quantity, and the required delivery time (reflected in the delivery date information). The net size cannot be directly used for cutting planning because material must be reserved for subsequent processes such as grinding and welding in actual production. Therefore, the net size can be precisely adjusted based on the equivalent machining allowance defined by experience in the key process parameters. The adjusted result is called the blanking requirement size, which is the theoretical cutting size that conforms to actual production conditions and is required by the optimization algorithm.

[0019] Subsequently, based on the processed data, a complex mathematical optimization model is constructed: a multi-objective mixed-integer linear programming model. The model's inputs include the required dimensions, quantity, delivery date, and raw material inventory information (e.g., the actual dimensions and inventory quantity of raw materials), as well as processing losses (e.g., kerf width) from key process parameters. The core of this model is to simultaneously pursue two optimization objectives: the first objective is to minimize raw material costs, i.e., to save materials as much as possible; the second objective is to minimize the weighted delivery delay time, i.e., to prioritize the delivery of urgent orders and minimize delays and their penalties. Through mathematical weighting, these two sometimes conflicting objectives can be intelligently balanced. It is important to note that the first objective can be either the primary or secondary objective, depending on the actual needs.

[0020] Subsequently, the constructed multi-objective mixed-integer linear programming model is solved to automatically calculate a detailed optimized material cutting plan. This plan can clearly specify how each raw material should be cut, as well as the cutting order for different orders. Once the operator confirms the optimized material cutting plan, it can not only output the optimized material cutting plan to guide production, but also automatically send instructions to the inventory system (such as the aforementioned warehouse management system) to update the inventory status of the raw materials occupied in the optimized material cutting plan in real time (e.g., marked as "locked"), thus forming a complete closed loop from intelligent calculation to production execution, and then to real-time feedback of resource status.

[0021] By applying the technical solution of this embodiment, the theoretical required dimensions are corrected by introducing key process parameters such as equivalent processing allowance and processing loss, enabling the multi-objective mixed-integer linear programming model to accurately fit the complex process reality, ensuring the feasibility of the optimized cutting plan and material utilization rate. Simultaneously, the multi-objective mixed-integer linear programming model constructed with cost and delivery time as its core achieves synergistic consideration and comprehensive optimization of dual optimization objectives through mathematical weighted coordination. The closed-loop design linking the optimized cutting plan confirmation with inventory status realizes automatic synchronization from optimization results to production execution, effectively avoiding the risk of raw material reuse. This embodiment solves the pain points of existing technologies where models are detached from reality and objectives are neglected. It can automatically and quickly generate a globally optimal cutting plan that balances material utilization rate and order delivery timeliness, freeing personnel from heavy manual calculations and achieving automation, intelligence, and reliability in production scheduling optimization.

[0022] Optionally, in this embodiment, step 104, "solving the multi-objective mixed-integer linear programming model to generate an optimized material cutting scheme," includes: using a hybrid optimization strategy combining a genetic algorithm and a Lagrange relaxation method to iteratively solve the multi-objective mixed-integer linear programming model and generate an optimized material cutting scheme. Specifically, in the iterative loop of the genetic algorithm, the Lagrange relaxation method is called to relax the multi-objective mixed-integer linear programming model to calculate the real-time cost lower bound of the first objective. The fitness function of the genetic algorithm is then adaptively corrected using the real-time cost lower bound until a preset condition is met, and the final optimized material cutting scheme is output.

[0023] In this embodiment, an innovative hybrid optimization strategy is employed to solve the constructed multi-objective mixed-integer linear programming model. The core of this strategy is the organic combination of genetic algorithms and the Lagrange relaxation method. Genetic algorithms are global search algorithms that simulate biological evolution, maintaining a population representing potential solutions and simulating natural evolutionary mechanisms such as selection, crossover, and mutation to iteratively improve the quality of solutions. Lagrange relaxation, on the other hand, is a classic mathematical optimization technique that relaxes complex constraints in the model, transforming the original problem into a more easily solvable relaxation problem, thus allowing for the rapid calculation of a theoretical lower bound on real-time costs. Combining the two aims to leverage the global exploration capabilities of genetic algorithms while utilizing the theoretical guidance provided by Lagrange relaxation to improve search efficiency.

[0024] In the specific iterative solution process, the genetic algorithm serves as the main framework driving the evolution of the population. In each evolutionary cycle, the Lagrange relaxation method can be invoked to simplify the complex original model, i.e., relaxation, with the aim of quickly calculating a real-time cost lower bound for the primary objective. This lower bound is a crucial theoretical reference value, representing the theoretical limit of the optimal cost achievable under current conditions. Subsequently, this real-time cost lower bound is used to adaptively adjust the fitness function of the genetic algorithm. The fitness function is a scoring criterion used in the genetic algorithm to evaluate the quality of each solution (individual). By incorporating the theoretical lower bound information provided by the Lagrange relaxation method into this scoring criterion, the search direction can be dynamically adjusted, guiding the population to evolve towards a theoretically better region that balances cost and delivery time, thereby greatly improving the targeting and efficiency of the search.

[0025] The above process can be repeated until preset conditions are met, such as reaching the maximum number of iterations or the quality of the solution no longer significantly improving in consecutive iterations. When the iteration terminates, the individual with the best fitness is selected from the final generation population and decoded into a specific, executable optimized material cutting plan, which specifies the cutting method for each raw material and the cutting sequence for different production orders.

[0026] This application's embodiments combine the advantages of two methods, achieving a synergy between empirical search and theoretical guidance. The genetic algorithm provides powerful global exploration capabilities, avoiding getting trapped in local optima; while the Lagrange relaxation method provides a theoretical lower bound as feedback at each iteration step, dynamically correcting the search path. This synergy significantly accelerates the overall convergence speed, making it possible to solve large-scale, complex multi-objective material handling problems within an acceptable timeframe, while ensuring the quality of the solution. This ensures that the final solution is not only theoretically efficient but also achieves a balance between cost and delivery time in a real-world production environment.

[0027] Optionally, in this embodiment, the step of "iteratively solving the multi-objective mixed-integer linear programming model using a hybrid optimization strategy combining genetic algorithm and Lagrange relaxation method to generate an optimized cutting scheme" includes: randomly generating an initial population of individuals representing candidate cutting combinations based on constraints such as the required cutting size, the required quantity, the raw material inventory information, and the processing loss; constructing an initial fitness function based on the first and second objectives of the multi-objective mixed-integer linear programming model, wherein the second objective includes an overdue penalty determined by the delivery date information; and in each iteration, using the Lagrange relaxation method to iteratively solve the multi-objective mixed-integer linear programming model to generate an optimized cutting scheme. The linear programming model is relaxed to obtain a lower bound for the first objective. Based on this lower bound, the weighting coefficients in the fitness function used to balance the first and second objectives are dynamically adjusted. Based on the dynamically adjusted fitness function, the fitness value of each individual in the current population is calculated. Selection, crossover, and mutation operations are performed based on these fitness values ​​to generate a new population. Iterative operations are then performed again based on the new population until a preset condition is met. The individual with the best fitness value in the last population group is decoded into a specific cutting combination scheme, which is output as the optimized cutting scheme. This cutting combination scheme includes a cutting method and a cutting order.

[0028] In this embodiment, the solution process begins with initialization. Based on known constraints such as required material dimensions, quantity, raw material inventory information (including material dimensions and quantity), and processing losses, an initial population of individuals is randomly generated. Here, each individual represents a complete, randomly arranged set of candidate cutting combinations, defining the initial combination of which parts are cut from which raw materials. Although the schemes are randomly generated, they must strictly adhere to physical constraints (e.g., the total length of parts cut from a single raw material plus losses cannot exceed the length of the raw material itself), thereby ensuring that each individual in the population represents a theoretically feasible starting point.

[0029] Next, a fitness function is constructed. This function is directly derived from the two optimization objectives set by the multi-objective mixed-integer linear programming model: the first objective is to minimize raw material costs, and the second objective is to minimize weighted delivery delay time. The specific calculation of the second objective can incorporate a delay penalty determined by delivery information; that is, if an optimized material cutting plan causes a delay in the delivery of an order, a quantified penalty value can be generated based on the delay time and the urgency of the order. The fitness function combines these two objectives with a certain weight ratio to form a single score used for initial ranking of individuals in the population.

[0030] In each iteration, the Lagrange relaxation method can be introduced for theoretical guidance. This method relaxes the complex and coupled constraints in the original model, thereby quickly solving a simplified problem. The optimal solution to this simplified problem provides crucial information: the theoretical minimum value that the first objective can achieve under the current population distribution, i.e., the relaxation lower bound. This lower bound is not an actual solution, but a theoretical benchmark. Subsequently, this dynamically updated theoretical benchmark can be used to dynamically adjust the weight coefficients in the fitness function used to balance the first and second objectives. For example, when the calculated cost lower bound is very low, it indicates a significant potential for cost reduction, and the algorithm can automatically increase the weight of the first objective in the fitness function, guiding the search to focus more on saving materials; conversely, it can focus more on delivery time. This makes the optimization search no longer blind, but provides real-time theoretical feedback.

[0031] Then, the classic evolutionary stage of the genetic algorithm begins. Based on the dynamically adjusted fitness function, the fitness value of each individual in the current population is recalculated. Next, selection is performed according to the principle of survival of the fittest to select superior individuals. Then, a new population is generated through crossover (combining some characteristics of different individuals) and mutation (randomly changing some characteristics of individuals). This process simulates biological evolution, aiming to inherit the superior traits of the parent generation and explore new possibilities, thereby generating a higher-quality next-generation solution.

[0032] The above iterative process is repeated until preset conditions are met, such as reaching the maximum number of iterations or the quality of the solution remaining stable. Finally, the individual with the best fitness value is selected from the last generation of the population and encoded as a specific, executable cutting combination scheme. This scheme can clearly list the cutting method (which parts to cut) and the cutting sequence (the production order among multiple orders) for each raw material, thereby directly guiding workshop production.

[0033] Optionally, before "randomly generating a set of initial population individuals representing candidate cutting combination schemes", the method further includes: constructing a first order feature corresponding to the production order information based on the material attributes, net size, and required quantity contained in the production order information, and performing feature similarity matching between the first order feature and the second order feature of historical orders with confirmed optimized cutting schemes; if a historical order with a similarity exceeding a preset threshold is matched, the confirmed optimized cutting scheme corresponding to the historical order is encoded as a seed chromosome individual that meets the requirements of the genetic algorithm; correspondingly, "randomly generating a set of initial population individuals representing candidate cutting combination schemes according to the constraints of the cutting requirement size, the required quantity, the raw material inventory information, and the processing loss" includes: generating other random individuals besides the seed chromosome individuals according to the constraints of the cutting requirement size, the required quantity, the raw material inventory information, and the processing loss, and combining the seed chromosome individuals and the other random individuals together into a complete initial population individual.

[0034] In this embodiment, before officially starting the random generation of the initial population, a structured first order feature is first constructed based on the material attributes, net size distribution patterns, and required quantities in the production order information. Then, the first order feature is compared with a database storing previously successfully implemented optimized material cutting schemes and their corresponding second order features. Feature similarity is calculated to identify whether there are orders with highly similar processing requirements among the historical orders.

[0035] If one or more historical orders with a feature similarity exceeding a preset threshold are successfully matched, it indicates that valuable reference experience has been found. Next, the validated and confirmed optimized material cutting schemes corresponding to these historical orders are encoded from their storage format into an internal expression form—a seed chromosome individual—that can be directly recognized and processed by the genetic algorithm. In the genetic algorithm, one seed chromosome individual represents a complete optimized material cutting scheme encoding.

[0036] When constructing the initial population, a batch of random individuals is generated based on standard constraints (material cutting requirements, quantity requirements, raw material inventory information, and processing losses) to ensure population diversity and the ability to explore the unknown solution space. Simultaneously, seed chromosome individuals carrying historical experience are directly injected into the newly generated random individuals. Finally, the seed chromosome individuals and random individuals combine to form a complete initial population. From the first generation, this initial population possesses both the guidance of high-quality solutions and the potential for broad-coverage search.

[0037] This application embodiment transforms the optimal solutions obtained from similar orders in the past into seeds to guide the algorithm's search. This effectively converts past successful experiences into prior knowledge that the algorithm can inherit, thereby significantly accelerating the initial convergence speed of the genetic algorithm. It helps the algorithm focus on promising search areas more quickly and also helps improve the overall quality of the final solution. This avoids the problems of blind initial search and slow convergence that may be caused by completely random initialization, and realizes the continuous accumulation and intelligent reuse of historical experience value.

[0038] Optionally, in this embodiment of the application, step 102, "correcting the net size based on the equivalent machining allowance in the key process parameters to obtain the blanking requirement size for optimization calculation," includes: determining a quantitative correction relationship between the equivalent machining allowance and the net size based on the key process parameters; performing numerical calculations on each net size extracted from the production order information according to the quantitative correction relationship to generate a corresponding theoretical blanking size containing the process allowance; and determining all the calculated theoretical blanking sizes as the blanking requirement size for optimization calculation.

[0039] In this embodiment, firstly, based on key process parameters, a specific quantitative correction relationship between the equivalent machining allowance and the net dimension is determined. Here, key process parameters typically include empirical rules or data tables developed by the process department for different materials and processing types. The equivalent machining allowance refers to the additional dimension beyond the theoretical net dimension needed to compensate for unavoidable material loss or deformation during subsequent processes such as cutting, heat treatment, and grinding. Determining the quantitative correction relationship involves transforming this empirical process knowledge into a clear mathematical rule, for example, specifying it as adding a fixed value, increasing it as a percentage of the net dimension, or querying a specific allowance value based on complex conditions.

[0040] Next, applying the quantitative correction relationship determined in the previous step, automated numerical calculations are performed on each net dimension extracted from the production order information. The net dimension is the final product size required by the customer, while the calculated size is the theoretical blanking size including equivalent machining allowances. This size is prepared for the actual cutting process, ensuring that the cut raw material segments, after all subsequent processing, still precisely meet the customer's dimensional requirements for the final product. Specifically, all requirement items in the production order can be traversed, generating the corresponding theoretical blanking size for each item, thereby completing the precise conversion from design dimensions to manufacturing dimensions.

[0041] Finally, all the calculated individual theoretical blanking dimensions are compiled into a complete dataset, which is then formally designated as the blanking requirement dimensions used in subsequent optimization calculations. All subsequent solutions to the multi-objective mixed-integer linear programming model are based on these blanking requirement dimensions, which already include equivalent machining allowances.

[0042] This application embodiment transforms the equivalent machining allowance into a precise, computer-processable quantitative correction relationship and automatically applies it to all net dimensions. This fundamentally eliminates material waste or part scrap caused by human estimation errors or omissions. As a result, the multi-objective mixed integer linear programming model is no longer an ideal model divorced from the actual workshop, but an executable model. This significantly improves the success rate and material utilization rate of the final optimized cutting scheme in real production.

[0043] In this embodiment of the application, optionally, the step 104 of "updating the corresponding raw material inventory status based on the confirmation instruction for the optimized cutting scheme" includes: saving the optimized cutting scheme to the database and initializing the status of the optimized cutting scheme to an inactive state; when a confirmation instruction for the optimized cutting scheme is received, updating the status of the optimized cutting scheme to an active state, and simultaneously sending an inventory lock instruction to the inventory system, so that the inventory system marks all raw materials planned to be occupied in the active optimized cutting scheme as occupied and locked based on the inventory lock instruction.

[0044] In this embodiment, the optimized material cutting plan is first saved to the database for persistent storage. Simultaneously, the optimized material cutting plan is assigned an initial inactive state. This state is a key management identifier, indicating that the optimized material cutting plan currently exists only as a preliminary plan; although it has undergone optimization calculations, it has not yet received final production authorization. In this state, the raw material occupancy information planned in the optimized material cutting plan is not transmitted to any execution system, and the raw materials retain their availability attribute in the inventory. This allows planners to review, compare, and even generate multiple alternative plans simultaneously without causing resource conflicts.

[0045] Once the planner approves and issues a confirmation instruction, the optimized material cutting plan can be updated in the database from "not yet effective" to "effective." This status change indicates that the optimized material cutting plan has been granted execution permission and has moved from the planning phase to the scheduling phase. As a linked action to this status change, an inventory lock instruction can be sent synchronously to the inventory system (such as WMS). This instruction is a key signal for cross-system collaboration, carrying the specific raw material numbers and quantities planned for use by the effective plan, driving the inventory system to reserve actual physical resources.

[0046] Upon receiving an inventory lock command, the inventory system can immediately perform a core operation in its database: marking the inventory status of all raw materials specified in the command as occupied and locked. This status is a global state visible to all subsequent production scheduling systems. In this way, any new material allocation optimization calculations or production dispatching can automatically exclude these locked raw materials when retrieving available inventory, thereby fundamentally preventing the duplication of planning for the same physical raw material by different production tasks.

[0047] This application's embodiments ensure the reversibility and flexibility of the optimization calculation (the optimized material cutting scheme can be arbitrarily adjusted before confirmation) through inactive / active state management and automated instruction synchronization. At the same time, it also guarantees the exclusivity and exclusivity of resource allocation after the optimized material cutting scheme is confirmed, solving the problem of "one material, multiple allocations" caused by information asynchrony in the traditional mode.

[0048] Optionally, after "obtaining the raw data required for material cutting" in step 101, the method further includes: identifying the target thread performing optimized material cutting calculation in a pre-created thread pool, determining the first inventory corresponding to the target thread, and determining whether the first inventory overlaps with the second inventory required by the production order information; when there is an overlap, starting a timer, and when the timer reaches a preset time threshold, obtaining an idle thread from the thread pool, and performing the step of extracting the net size, required quantity, and delivery date information of the product to be processed from the production order information based on the idle thread; when there is no overlap, directly obtaining an idle thread from the thread pool, and performing the step of extracting the net size, required quantity, and delivery date information of the product to be processed from the production order information based on the idle thread.

[0049] In this embodiment, the system maintains a pre-created thread pool, a set of reusable computing resources used to process multiple optimized material cutting scheme calculation tasks in parallel. After acquiring the raw data, firstly, all target threads running in the thread pool are identified, and the first inventory corresponding to each target thread is analyzed, i.e., the set of raw materials that the optimized material cutting scheme being calculated is attempting to plan to use. Then, it is determined whether this first inventory overlaps with the second inventory required by the production order information, i.e., whether the production order information will compete with the optimized material cutting scheme being calculated for the same physical raw material.

[0050] When inventory overlap is detected, it indicates that immediately calculating the optimized material cutting plan corresponding to the production order information may lead to resource allocation conflicts. At this point, a timer is started, initiating a short delay period. This is because the currently running optimized material cutting plan calculation task may soon complete and release the virtual inventory resources it occupies. When the timer reaches a preset time threshold, another attempt is made, retrieving an idle thread from the thread pool. If the conflict resolves naturally due to the completion of the previous optimized material cutting plan calculation task, subsequent data extraction and optimization steps can be safely executed using this idle thread.

[0051] When no inventory overlap is detected, it indicates that the raw materials required for the optimized cutting plan corresponding to the production order information are completely independent and do not compete for resources with any ongoing optimized cutting plan calculation tasks. In this case, an idle thread can be directly obtained from the thread pool, and the optimized cutting plan calculation task corresponding to the production order information can be immediately assigned to it to begin executing a series of subsequent calculation steps, such as extracting net dimensions, required quantities, and delivery date information from the production order information, without any waiting. This ensures that the system's throughput and response speed are optimal when inventory resources are sufficient.

[0052] This application's embodiments effectively prevent infeasibility and resource waste caused by multiple optimization threads simultaneously planning the same inventory raw materials by proactively detecting inventory conflicts and supplementing them with intelligent delays, ensuring the effectiveness of each optimized material cutting plan. Simultaneously, its lightweight conflict resolution strategy minimizes the impact on overall concurrency performance, enabling the system to efficiently and stably handle optimized material cutting plan calculation requests under high concurrency, thus improving the system's robustness and practicality in multi-tasking and dynamic environments.

[0053] Furthermore, as a refinement and extension of the specific implementation of the above embodiments, and to fully illustrate the specific implementation process of this embodiment, another optimized material cutting method based on mixed-integer linear programming is provided, such as... Figure 2 As shown, the method includes: Production order information, including customer product requirements, quantities, and delivery deadlines, is automatically retrieved from the ERP (Enterprise Resource Planning) system. Simultaneously, raw material inventory status is obtained in real-time from the WMS (Warehouse Management System), and data processing is performed to obtain raw material inventory information, such as available material specifications, quantities, and storage locations. Furthermore, key process parameters (such as the aforementioned equivalent processing allowance and cutting loss) are also input as core knowledge. These three types of data form the computational basis for the optimized material cutting scheme and are input into the optimized material cutting algorithm. This algorithm, based on a multi-objective mixed-integer linear programming model, aims to minimize raw material costs and delivery delays. It comprehensively considers all order requirements, inventory constraints, and process rules, performing rapid calculations and global optimization to produce an optimized material cutting scheme. This scheme specifies in detail how each raw material should be cut to fulfill which orders, and the cutting order of each order.

[0054] After generating the optimized material cutting plan, it is not immediately put into production. Instead, it enters a "confirmation-feedback" human-machine interaction and decision-making closed loop. Operators can review the optimized material cutting plan. If they are satisfied, they confirm the plan; if they believe that adjustments are necessary (such as based on urgent orders on site), they can choose to modify the plan requirements or parameters.

[0055] Once the optimized material cutting plan is finalized, two key actions can be executed immediately: First, a command is sent to the WMS to lock all raw material inventory involved in the optimized plan, marking its status as occupied and locked to prevent it from being repeatedly allocated by other tasks, thus physically ensuring the uniqueness and executability of the plan; second, the confirmed optimized material cutting plan is sent to the MES (Manufacturing Execution System) for production execution, guiding shop floor equipment to perform precise cutting. Thus, a complete closed loop from intelligent calculation and human decision-making to resource locking and production execution is achieved.

[0056] Furthermore, as Figure 1 In terms of specific implementation, this application provides an optimized material feeding system based on mixed-integer linear programming, such as... Figure 3 As shown, the system includes: The data acquisition module is used to acquire the raw data required for material cutting, wherein the raw data includes production order information, raw material inventory information and key process parameters from external systems; The size correction module is used to extract the net size, required quantity and delivery date information of the product to be processed from the production order information, and correct the net size based on the equivalent processing allowance in the key process parameters to obtain the blanking requirement size for optimization calculation. The model building module is used to construct a multi-objective mixed integer linear programming model based on the material cutting requirement size, the required quantity, the delivery date information, the raw material inventory information, and the processing loss in the key process parameters. The model aims to minimize the raw material cost as the first objective and minimize the weighted delivery overdue time as the second objective. The solution module is used to solve the multi-objective mixed integer linear programming model, generate an optimized material cutting scheme, and update the corresponding raw material inventory status based on the confirmation instruction of the optimized material cutting scheme.

[0057] Optionally, the solution module is used for: A hybrid optimization strategy combining genetic algorithm and Lagrange relaxation method is used to iteratively solve the multi-objective mixed-integer linear programming model to generate an optimized material cutting scheme. Specifically, in the iterative loop of the genetic algorithm, the Lagrange relaxation method is called to relax the multi-objective mixed-integer linear programming model to calculate the real-time cost lower bound of the first objective. The fitness function of the genetic algorithm is then adaptively adjusted using the real-time cost lower bound until a preset condition is met, and the final optimized material cutting scheme is output.

[0058] Optionally, the solution module is further configured to: Based on the constraints of the required material size, the required quantity, the raw material inventory information, and the processing loss, an initial population of individuals representing candidate cutting combination schemes is randomly generated; Based on the first and second objectives of the multi-objective mixed integer linear programming model, an initial fitness function is constructed, wherein the second objective includes an overdue penalty determined by the delivery information; In each iteration, the Lagrange relaxation method is used to relax and solve the multi-objective mixed integer linear programming model to obtain the relaxation lower bound of the first objective, and the weight coefficients used to balance the first objective and the second objective in the fitness function are dynamically modified according to the relaxation lower bound. Based on the dynamically corrected fitness function, the fitness value of each individual in the current population is calculated, and selection, crossover, and mutation operations are performed according to the fitness value to generate a new population of individuals; The iteration operation is performed again based on the new population individuals until the preset conditions are met. The population individual with the best fitness value in the last group of population individuals is decoded into a specific cutting combination scheme, which is output as the optimized cutting scheme. The cutting combination scheme includes cutting method and cutting order.

[0059] Optionally, the system further includes a matching module; the matching module is used for: Before randomly generating an initial population of individuals representing candidate cutting combination schemes, a first order feature corresponding to the production order information is constructed based on the material attributes, net dimensions, and required quantity contained in the production order information, and the first order feature is matched with the second order feature of historical orders with confirmed optimized cutting schemes based on feature similarity. If a historical order with a similarity exceeding a preset threshold is matched, the confirmed optimized cutting scheme corresponding to the historical order is encoded as a seed chromosome individual that meets the requirements of the genetic algorithm; Accordingly, the solution module is also used for: Based on the constraints of the required material size, the required quantity, the raw material inventory information, and the processing loss, random individuals other than the seed chromosome individual are generated, and the seed chromosome individual and the remaining random individuals are combined to form a complete initial population.

[0060] Optionally, the size correction module is used for: Based on the key process parameters, determine the quantitative correction relationship between the equivalent machining allowance and the net size; Based on the quantitative correction relationship, numerical calculations are performed on each net dimension extracted from the production order information to generate the corresponding theoretical blanking dimension that includes process allowances. All the theoretical blanking dimensions obtained from the calculation are determined as the blanking requirement dimensions for optimization calculation.

[0061] Optionally, the solution module is further configured to: Save the optimized material cutting scheme to the database and initialize the status of the optimized material cutting scheme to an inactive state. Upon receiving a confirmation instruction for the optimized material cutting scheme, the status of the optimized material cutting scheme is updated to "effective," and an inventory lock instruction is simultaneously sent to the inventory system. This enables the inventory system to mark all raw materials planned to be occupied in the effective optimized material cutting scheme as "occupied and locked," based on the inventory lock instruction.

[0062] Optionally, the system further includes a judgment module; The judgment module is used to identify the target thread performing optimized material cutting calculation in the pre-created thread pool after obtaining the raw data required for material cutting, determine the first inventory corresponding to the target thread, and determine whether the first inventory overlaps with the second inventory required by the production order information. The size correction module is also used to start a timer when there is overlap, and when the timer reaches a preset time threshold, obtain an idle thread from the thread pool, and perform the step of extracting the net size, required quantity and delivery date information of the product to be processed from the production order information based on the idle thread; The size correction module is also used to directly obtain an idle thread from the thread pool when there is no overlap, and to perform the step of extracting the net size, required quantity and delivery date information of the product to be processed from the production order information based on the idle thread.

[0063] It should be noted that other corresponding descriptions of the functional units involved in the optimized material feeding system based on mixed-integer linear programming provided in this application embodiment can be found in the following references. Figures 1 to 2 The corresponding descriptions in the method will not be repeated here.

[0064] This application also provides a computer device, which may specifically be a personal computer, a server, a network device, etc. Figure 4 As shown, the computer device includes a bus, a processor, memory, and a communication interface, and may also include an input / output interface and a display device. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores location information. The network interface allows communication with external terminals via a network connection. When the computer program is executed by the processor, it implements the steps in the various method embodiments.

[0065] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0066] In one embodiment, a computer-readable storage medium is provided, which may be non-volatile or volatile, having stored thereon a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0067] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0068] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0069] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0070] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0071] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An optimized material cutting method based on mixed-integer linear programming, characterized in that, include: Obtain the raw data required for material cutting, wherein the raw data includes production order information, raw material inventory information and key process parameters from external systems; Extract the net dimensions, required quantity, and delivery date information of the products to be processed from the production order information. Correct the net dimensions based on the equivalent processing allowance in the key process parameters to obtain the blanking requirement dimensions for optimization calculation. Based on the required material dimensions, required quantity, delivery date information, raw material inventory information, and processing loss in the key process parameters, a multi-objective mixed integer linear programming model is constructed with the first objective of minimizing raw material cost and the second objective of minimizing weighted delivery overdue time. The multi-objective mixed integer linear programming model is solved to generate an optimized material cutting scheme, and the corresponding raw material inventory status is updated based on the confirmation instruction of the optimized material cutting scheme.

2. The method according to claim 1, characterized in that, Solving the multi-objective mixed-integer linear programming model to generate an optimized material cutting scheme includes: A hybrid optimization strategy combining genetic algorithm and Lagrange relaxation method is used to iteratively solve the multi-objective mixed-integer linear programming model to generate an optimized material cutting scheme. Specifically, in the iterative loop of the genetic algorithm, the Lagrange relaxation method is called to relax the multi-objective mixed-integer linear programming model to calculate the real-time cost lower bound of the first objective. The fitness function of the genetic algorithm is then adaptively adjusted using the real-time cost lower bound until a preset condition is met, and the final optimized material cutting scheme is output.

3. The method according to claim 2, characterized in that, The hybrid optimization strategy, combining genetic algorithm and Lagrange relaxation method, iteratively solves the multi-objective mixed-integer linear programming model to generate an optimized material cutting scheme, including: Based on the constraints of the required material size, the required quantity, the raw material inventory information, and the processing loss, an initial population of individuals representing candidate cutting combination schemes is randomly generated; Based on the first and second objectives of the multi-objective mixed integer linear programming model, an initial fitness function is constructed, wherein the second objective includes an overdue penalty determined by the delivery information; In each iteration, the Lagrange relaxation method is used to relax and solve the multi-objective mixed integer linear programming model to obtain the relaxation lower bound of the first objective, and the weight coefficients used to balance the first objective and the second objective in the fitness function are dynamically modified according to the relaxation lower bound. Based on the dynamically corrected fitness function, the fitness value of each individual in the current population is calculated, and selection, crossover, and mutation operations are performed according to the fitness value to generate a new population of individuals; The iteration operation is performed again based on the new population individuals until the preset conditions are met. The population individual with the best fitness value in the last group of population individuals is decoded into a specific cutting combination scheme, which is output as the optimized cutting scheme. The cutting combination scheme includes cutting method and cutting order.

4. The method according to claim 2 or 3, characterized in that, Before randomly generating an initial population of individuals representing candidate cutting combinations, the method further includes: Based on the material attributes, net dimensions, and required quantity contained in the production order information, a first order feature corresponding to the production order information is constructed, and the first order feature is matched with the second order feature of historical orders with confirmed optimized cutting schemes based on feature similarity. If a historical order with a similarity exceeding a preset threshold is matched, the confirmed optimized cutting scheme corresponding to the historical order is encoded as a seed chromosome individual that meets the requirements of the genetic algorithm; Accordingly, based on the constraints of the required material size, the required quantity, the raw material inventory information, and the processing loss, an initial population of individuals representing candidate cutting combinations is randomly generated, including: Based on the constraints of the required material size, the required quantity, the raw material inventory information, and the processing loss, random individuals other than the seed chromosome individual are generated, and the seed chromosome individual and the remaining random individuals are combined to form a complete initial population.

5. The method according to claim 1, characterized in that, The step of correcting the net size based on the equivalent machining allowance in the key process parameters to obtain the blanking requirement size for optimization calculation includes: Based on the key process parameters, determine the quantitative correction relationship between the equivalent machining allowance and the net size; Based on the quantitative correction relationship, numerical calculations are performed on each net dimension extracted from the production order information to generate the corresponding theoretical blanking dimension that includes process allowances. All the calculated theoretical blanking dimensions are determined as the blanking requirement dimensions for optimization calculations.

6. The method according to claim 1, characterized in that, The step of updating the corresponding raw material inventory status based on the confirmation instruction for the optimized feeding scheme includes: Save the optimized material cutting scheme to the database and initialize the status of the optimized material cutting scheme to an inactive state. Upon receiving a confirmation instruction for the optimized material cutting scheme, the status of the optimized material cutting scheme is updated to "effective," and an inventory lock instruction is simultaneously sent to the inventory system. This enables the inventory system to mark all raw materials planned to be occupied in the effective optimized material cutting scheme as "occupied and locked," based on the inventory lock instruction.

7. The method according to claim 1, characterized in that, After obtaining the raw data required for material cutting, the method further includes: Identify the target thread performing optimized material cutting calculation in the pre-created thread pool, determine the first inventory corresponding to the target thread, and determine whether the first inventory overlaps with the second inventory required by the production order information; When there is overlap, a timer is started, and when the timer reaches a preset time threshold, an idle thread is obtained from the thread pool, and the step of extracting the net size, required quantity and delivery date information of the product to be processed from the production order information is performed based on the idle thread. When there is no overlap, an idle thread is directly obtained from the thread pool, and the steps of extracting the net size, required quantity and delivery date information of the product to be processed from the production order information are performed based on the idle thread.

8. An optimized material feeding system based on mixed-integer linear programming, characterized in that, include: The data acquisition module is used to acquire the raw data required for material cutting, wherein the raw data includes production order information, raw material inventory information and key process parameters from external systems; The size correction module is used to extract the net size, required quantity and delivery date information of the product to be processed from the production order information, and correct the net size based on the equivalent processing allowance in the key process parameters to obtain the blanking requirement size for optimization calculation. The model building module is used to construct a multi-objective mixed integer linear programming model based on the material cutting requirement size, the required quantity, the delivery date information, the raw material inventory information, and the processing loss in the key process parameters. The model aims to minimize the raw material cost as the first objective and minimize the weighted delivery overdue time as the second objective. The solution module is used to solve the multi-objective mixed integer linear programming model, generate an optimized material cutting scheme, and update the corresponding raw material inventory status based on the confirmation instruction of the optimized material cutting scheme.

9. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

10. A computer device, comprising a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 7.

Images