A production line scheduling and job sequencing method based on label matching score

Abstract

The application discloses a production line scheduling and job sequencing method based on label matching score, which realizes quantitative evaluation of the adaptation degree between orders and production lines and automatic solution of a globally optimal scheduling scheme by combining a label matching score mechanism and a scheduling optimization model; and realizes continuity and efficiency of production tasks by fragmenting and correcting initial scheduling results and optimizing job sequencing. Thus, the method has the beneficial effects of improving production line utilization, reducing line change loss, guaranteeing on-time delivery of orders, and effectively reducing production management complexity.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing and production scheduling technology, and in particular to a production line scheduling and job sorting method based on label matching and scoring. Background Technology

[0002] In modern discrete manufacturing enterprises, production planning is the core link connecting sales orders and shop floor execution. Line planning requires the rational allocation of numerous order lines to multiple production lines with different process capabilities, and the determination of specific start dates and production quantities. This process must not only meet order delivery dates and customer inspection requirements, but also take into account the daily capacity of the production lines, the earliest start date, and the continuity of production.

[0003] As market order volume expands and product process dimensions increase, the complexity of production scheduling decisions rises exponentially. The key to production scheduling today is no longer simply determining whether a particular production line can produce an order, but rather deciding which production line is optimal for allocating the order.

[0004] Current production scheduling techniques primarily rely on the manual experience of planners. The typical practice is for planners to manually compile order, capacity, and delivery date data, filling in the information day by day and line by line according to the principle of prioritizing delivery dates. When determining the matching relationship between order lines and production lines, priority is determined entirely by the planner's subjective memory and experience regarding the production line's process characteristics, and then operations such as inserting orders, changing lines, or shifting production dates are performed.

[0005] However, this existing technological solution, which relies on human experience, has the following significant drawbacks:

[0006] The results are unstable and rely on personal experience: different planners have different understandings of the same batch of orders and production lines, resulting in huge differences in production scheduling results, and it is difficult to reuse the experience of excellent planners.

[0007] Low resource matching leads to waste: Due to the lack of a quantitative matching mechanism, it is easy for production lines with high matching to be occupied by orders with low matching, while truly suitable orders are assigned to inefficient production lines, resulting in low production efficiency and waste of resources.

[0008] Fragmented production scheduling and low execution efficiency: Manual production scheduling makes it difficult to balance delivery time and capacity globally, which easily generates a large number of fragmented production scheduling tasks, leading to frequent line changes and debugging in the workshop, which seriously affects the continuity and stability of production.

[0009] Rescheduling is costly and slow to respond: When there are emergency orders or production line abnormalities, manual adjustments have far-reaching consequences, and the efficiency of re-compiling plans is low, making it difficult to respond quickly to dynamic changes on the production floor.

[0010] Therefore, there is an urgent need for a technology that can quantify and score the process matching relationship between production lines and orders, and solve it in a unified manner with hard constraints such as delivery time, capacity, and inspection, so as to output a production scheduling plan that takes into account both global optimization and execution continuity. Summary of the Invention

[0011] The purpose of this invention is to propose a method that integrates the process matching relationship between production lines and orders with hard constraints such as delivery time, capacity, and inspection to obtain a production scheduling scheme that balances global optimization and execution continuity.

[0012] To achieve this objective, the present invention adopts the following technical solution:

[0013] A production line scheduling and job sequencing method based on label matching and scoring includes the following steps:

[0014] S101. Obtain the set of order lines, the set of production lines, and the set of planned dates; where each order line contains process attributes and each production line contains process capabilities;

[0015] S102. Based on the preset tag dimensions, map the process attributes of the order line to order tags, and map the process capabilities of the production line to production line tags;

[0016] S103. Calculate the tag matching score between each order line and each production line based on the order tag and the production line tag;

[0017] S104. Construct a production scheduling optimization model, wherein the production scheduling optimization model aims to minimize the total production scheduling cost and includes constraints; wherein the total production scheduling cost includes the matching penalty cost calculated based on the label matching score;

[0018] S105. Solve the production scheduling optimization model to obtain the initial production output of each order line on the corresponding production line and the planned date;

[0019] S106. Based on the initial production output, generate a production line scheduling plan, and control the corresponding production line to carry out production according to the production line scheduling plan.

[0020] Furthermore, step S103, which calculates the tag matching score between each order line and each production line based on the order tag and production line tag, includes the following sub-steps:

[0021] S1031. Obtain the weight of each of the label dimensions, and the score of the production line for different label values ​​in each of the label dimensions;

[0022] S1032. For any combination of the order line and the production line, query the score of the production line in the corresponding dimension based on the value of the order tag of the order line in each dimension;

[0023] S1033. Multiply the scores of each dimension by their corresponding weights and sum them to obtain the tag matching score between the order line and the production line.

[0024] Furthermore, the constraints in step S104 include:

[0025] Production line capacity constraint: used to limit the cumulative output of each production line within each planned date to not exceed the upper limit of the available capacity of that production line;

[0026] Order completion constraint: Used to ensure that the cumulative production output of each order line over all the planned dates meets the total demand for that order line;

[0027] and at least one of the following constraints:

[0028] Earliest start date constraint for order lines: This constraint restricts the production scheduling date of the order line from being earlier than its earliest start date.

[0029] Manufacturability constraint for order lines: This constraint restricts the production of the order line to only on the permitted production lines.

[0030] Online inspection constraint: Used to ensure that the production output of the order line on the specified inspection date is not lower than the preset minimum batch size;

[0031] Proportional inspection constraint: This is used to ensure that the cumulative production output completed by the order line before the specified inspection date reaches a preset target proportion.

[0032] Furthermore, after step S105 and before step S106, the following steps are also included:

[0033] S201. Identify the order lines where the remaining production to be scheduled is lower than a preset threshold;

[0034] S202. The remaining production to be scheduled in the order line is preferentially allocated to the production line-date combination that already has production records in the order line, so as to reduce the number of production line switching times.

[0035] Furthermore, the following steps are included after step S202:

[0036] S203. Identify fragmented production scheduling segments whose production output is lower than a preset threshold;

[0037] S204. Migrate the fragmented production schedule fragments to production line-date combinations that are closer in time, have sufficient capacity, and match the label of the order line at a higher score.

[0038] Furthermore, following step S106, the following steps are also included:

[0039] S301. Obtain all order lines and their production volumes for the production line to be scheduled within the planned date.

[0040] S302. Calculate the continuity score of each order line to be scheduled for production according to the preset continuity scoring rules; wherein, the continuity scoring rules prioritize arranging the order lines of the same product, the same customer, or the same order adjacent to each other;

[0041] S303. Based on the continuity score from high to low, generate the work execution sequence of the production line within the planned date.

[0042] Furthermore, in step S303, when generating the job execution sequence based on the continuity score, if multiple order lines have the same continuity score, they are sorted according to the urgency of their latest completion date, and the order line with the earlier latest completion date is given priority.

[0043] Furthermore, after step S303, the following steps are also included:

[0044] S304. Check the production scheduling segment at the end of the current planned date and the production scheduling segment at the beginning of the next planned date. When the production output of both production scheduling segments is lower than a preset threshold and belongs to the same order line, merge the two production scheduling segments to reduce the intermittent switching of the production line.

[0045] Furthermore, the production scheduling optimization model in step S104 adopts a mixed integer programming model and is solved by a mathematical programming solver.

[0046] Furthermore, step S106, which generates a production line scheduling plan based on the initial production output, includes the following sub-steps:

[0047] S1061. Aggregate the initial production output according to the production line and the planned date to generate different daily production task lists for the production line;

[0048] S1062. The daily production task list is sent to the corresponding production line control system to drive the corresponding production line to execute the production task.

[0049] The technical solution provided by this invention may include the following beneficial effects:

[0050] A production line scheduling and job sequencing method based on tag matching and scoring is proposed. This method combines a tag matching and scoring mechanism with a scheduling optimization model to achieve quantitative assessment of the fit between orders and production lines and automatic solution of the globally optimal scheduling plan. By performing fragmented correction and job sequencing optimization on the initial scheduling results, the continuity and efficiency of production tasks are achieved. This results in improved production line utilization, reduced changeover losses, guaranteed on-time order delivery, and effectively reduced production management complexity. Attached Figure Description

[0051] Figure 1 This is a flowchart of the steps of the production line scheduling and job sorting method based on tag matching and scoring proposed in the embodiments of the present invention. Detailed Implementation

[0052] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that the specific embodiments described herein are for illustrative and explanatory purposes only and are not intended to limit the present invention.

[0053] For details, please refer to Figure 1 , Figure 1 This is a flowchart illustrating the steps of the production line scheduling and job sequencing method based on tag matching scores proposed in this invention. The production line scheduling and job sequencing method based on tag matching scores specifically includes the following steps:

[0054] S101. Obtain the set of order lines, the set of production lines, and the set of planned dates; where each order line contains process attributes and each production line contains process capabilities.

[0055] Specifically, during implementation, step S101 retrieves the set of order lines to be scheduled from the Enterprise Resource Planning (ERP) or Manufacturing Execution System (MES). Each order line contains its process attributes, such as product model, color, size, fabric, and special process requirements. Simultaneously, it retrieves the set of all available production lines within the factory, each containing its process capabilities, such as equipment type, supported processes, maximum capacity, and changeover time. Furthermore, it also retrieves the set of planned dates covered by the scheduling, such as every working day for the next two weeks.

[0056] For example, to facilitate modeling and understanding, the set of order lines, the set of production lines, and the set of planned dates are defined as follows in this embodiment of the invention:

[0057] Order line collection, index ;

[0058] Order collection, index ;

[0059] Production line set, index ;

[0060] : Set of planned dates, index ;

[0061] : A collection of tag dimensions.

[0062] S102. Based on the preset tag dimensions, map the process attributes of the order line to order tags, and map the process capabilities of the production line to production line tags.

[0063] In this embodiment of the invention, a set of standardized label dimensions is predefined, such as "product type," "process complexity," and "color." Then, the specific process attribute values ​​of each order line (such as "T-shirt," "printed," and "red") are categorized under the corresponding label dimensions to form structured order labels. Similarly, the process capabilities of each production line (such as "proficient in knitting" and "supports digital printing") are also mapped to production line labels, thereby achieving the standardization of heterogeneous data.

[0064] S103. Calculate the tag matching score between each order line and each production line based on the order tag and the production line tag.

[0065] For any given order line and production line, step S103 compares their matching degree across various label dimensions. For example, if the order requires "digital printing" and the production line has "digital printing" capabilities, it will receive a high score in that dimension; if the production line only supports "screen printing," it will receive a lower score. By weighting and calculating the scores across all dimensions, a quantified label matching score is obtained, which directly reflects the suitability of the production line for executing the order.

[0066] Furthermore, the calculation process of the label matching score in S103 includes the following sub-steps:

[0067] S1031. Obtain the weight of each of the label dimensions, and the score of the production line for different label values ​​in each of the label dimensions.

[0068] During implementation, a weighting table can be designed to define the importance of different label dimensions. For example, the weight of "process type" might be 0.5, and the weight of "color" might be 0.2. Simultaneously, a production line capability score table can be designed to record the proficiency score (e.g., 0-10 points) of each production line for different labels across different dimensions. For example, production line A might have a score of 10 points for the "knitted" label and a score of 2 points for the "woven" label in the "process type" dimension.

[0069] S1032. For any combination of the order line and the production line, query the score of the production line in the corresponding dimension based on the value of the order tag of the order line in each dimension.

[0070] Specifically, for an order line to be calculated (such as a knitted T-shirt) and a production line (such as production line A), the corresponding score (10 points) in the "process type" dimension can be found in the capability score table of production line A based on the label ("knitted") of the order line.

[0071] S1033. Multiply the scores of each dimension by their corresponding weights and sum them to obtain the tag matching score between the order line and the production line.

[0072] According to the example in the above embodiment, during implementation, by repeating step S1032, the scores of the order line and production line A on all label dimensions are obtained, and then multiplied by the preset weights respectively, and finally all products are added together. Specifically, according to steps S1031 to S1033, the scoring rules are as follows:

[0073] Each dimension is worth a maximum of 10 points;

[0074] Weights are allocated as percentages, with a total weight of 100%.

[0075] Single-dimensional score = Dimension score × Dimension weight;

[0076] Total tag matching score = sum of individual scores for each dimension.

[0077] S104. Construct a production scheduling optimization model, wherein the production scheduling optimization model aims to minimize the total production scheduling cost and includes constraints; wherein the total production scheduling cost includes the matching penalty cost calculated based on the tag matching score.

[0078] Specifically, in this embodiment of the invention, an optimization model is constructed in the form of mathematical programming. The objective function of the model is to minimize the total production scheduling cost, where the core cost term is the matching penalty cost calculated based on step S103 (i.e., the lower the matching score, the higher the penalty cost). Simultaneously, the model must include a series of hard constraints to ensure the feasibility of the production scheduling plan.

[0079] Based on the relevant parameters concerning production lines and order lines involved in the above embodiments, the relevant parameters and decision variables are defined as follows in this embodiment of the invention:

[0080] Order line Remaining production to be scheduled;

[0081] Production line On the date Effective production capacity;

[0082] Order line exist Is it feasible to produce?

[0083] :Order Earliest start date;

[0084] Order line Latest completion date;

[0085] Order line Total demand;

[0086] Order line Historical completed work;

[0087] Target ratio for proportional inspection;

[0088] Inspection date;

[0089] Minimum order quantity for online inspection;

[0090] Tag Dimension Weight (percentage);

[0091] Production line In dimensions Get the value of the label The score (out of 10 points);

[0092] Order line In dimensions The value of the tag;

[0093] Order line exist Production output;

[0094] Order line Unfinished amount;

[0095] : Whether it is in production;

[0096] :Order Whether to use production line ;

[0097] Order line Whether to use production line ;

[0098] Deviation from the baseline plan;

[0099] Inspection slack variable.

[0100] For order line With production line The tag matching score between them is defined as:

[0101] .

[0102] Based on the relevant parameters and decision variables defined in the above embodiments, the production scheduling optimization model constructed in step S104 can be expressed as:

[0103] .

[0104] Furthermore, the constraints included in the production scheduling optimization model in step S104 include:

[0105] Production line capacity constraint: This constraint limits the cumulative output of each production line within each planned date to no more than the line's available capacity limit. For example, if production line A has a daily capacity of 1000 units, then the sum of the output allocated to all order lines for production line A on a given day must be less than or equal to 1000 units. This constraint prevents the production line from operating at overcapacity, ensuring the physical feasibility of the production plan. It can be expressed as:

[0106] .

[0107] Order completion constraint: This constraint ensures that the cumulative production output for each order line across all planned dates meets the total demand for that order line. For example, if order line B requires 5000 units, then the sum of all production outputs allocated to order line B throughout the entire production cycle must equal 5000 units. This constraint ensures that all orders are produced completely, preventing missed orders. It can be represented as:

[0108] .

[0109] Production line capacity constraints and order fulfillment constraints are necessary constraints for the production scheduling optimization model. During implementation, the constraints may also include at least one of the following constraints:

[0110] Earliest Start Date Constraint for Order Lines: This constraint restricts the production scheduling date of an order line from being earlier than the earliest start date when its materials are readily available or its technical preparations are complete. It can be represented as:

[0111] .

[0112] Order line manufacturability constraint: This restricts the order line to be scheduled only on the permitted production lines. It can be represented as:

[0113] .

[0114] Online inspection constraint: This ensures that the completed production volume of the order line is not less than a preset minimum batch size on the customer-specified inspection date, guaranteeing sufficient samples for customer inspection. It can be represented as:

[0115] .

[0116] Proportional inspection constraint: This ensures that the cumulative production output completed by an order line reaches a preset target proportion of the total demand before the specified inspection date. It can be expressed as:

[0117] .

[0118] During implementation, order-production line and order line-production line can also be linked:

[0119] ;

[0120] And set threshold constraints for label matching scores:

[0121] .

[0122] S105. Solve the production scheduling optimization model to obtain the initial production schedule for each order line on the corresponding production line and the planned date.

[0123] Specifically, the production scheduling optimization model constructed in step S104 is mathematically a mixed integer programming (MIP) model. The decision variables in the production scheduling optimization model can be integers, and the objective function and constraints are all linear expressions. By calling mature commercial or open-source mathematical programming solvers for efficient solving, it is possible to find the optimal solution or a high-quality approximate optimal solution for large-scale production scheduling problems within an acceptable timeframe.

[0124] In this embodiment of the invention, the constructed model is input into a mathematical programming solver (such as COPT, CPLEX, Gurobi, etc.) for solving. The solver, while satisfying all constraints, seeks the solution that minimizes the total production scheduling cost, and outputs the specific production output for each order line on each day and each production line, thus forming an initial production scheduling plan.

[0125] In this embodiment of the invention, after step S105, there is also an optimization and correction step for the production scheduling results, specifically:

[0126] S201. Identify the order lines where the remaining production to be scheduled is lower than a preset threshold.

[0127] Step S201 scans the production scheduling results of all order lines to identify those order lines with remaining unscheduled quantities (or odds and ends generated by model solving) less than a small threshold (e.g., 50 units). If these odds and ends are scheduled for production separately, it will lead to frequent production line changes and low efficiency.

[0128] S202. The remaining production to be scheduled in the order line is preferentially allocated to the production line-date combination that already has production records in the order line, so as to reduce the number of production line switching times.

[0129] Specifically, for the small orders identified in S201, the system first checks whether the order has already been scheduled for production on a certain production line on a certain day. If so, these small quantities are merged into the production task for that production line-date combination. This design eliminates the need for an extra production line switch for a small number of products, effectively reducing the number of production line switches and improving overall production efficiency.

[0130] S203. Identify fragmented production segments whose production output is lower than a preset threshold.

[0131] Step S203 examines the initial production schedule and identifies isolated production task segments with very small production volumes (e.g., less than 100 units). These fragmented tasks also lead to a waste of production efficiency.

[0132] S204. Migrate the fragmented production schedule fragments to production line-date combinations that are closer in time, have sufficient capacity, and match the label of the order line at a higher score.

[0133] For the fragmented tasks identified in step S203, an attempt will be made to migrate them to other production line-date combinations that are closer in time, have sufficient capacity, and match the order line label more closely. For example, a small order of 10 units produced on Wednesday on production line A may be merged into a larger order of the same type produced on Tuesday on production line B. This step aims to further consolidate scattered tasks, reduce production interruptions, and select more suitable production lines for production whenever possible.

[0134] S106. Based on the initial production output, generate a production line scheduling plan, and control the corresponding production line to carry out production according to the production line scheduling plan.

[0135] Finally, the initial production scheduling data output by the solver is formatted and organized to generate a visual, executable production line scheduling plan. This plan is then distributed to the corresponding production line control system or MES terminal to guide the production line to produce according to the plan.

[0136] Step S106, which generates a production line scheduling plan based on the initial production output, includes the following sub-steps:

[0137] S1061. Aggregate the initial production output according to the production line and the planned date to generate different daily production task lists for the production line.

[0138] Specifically, during implementation, the production scheduling data output by the solver in step S105, which is centered on order behavior, is reorganized into a view centered on production lines and dates. By traversing all production lines and dates, all tasks that each production line needs to produce each day are summarized, generating a clear daily production task list.

[0139] S1062. The daily production task list is sent to the corresponding production line control system to drive the corresponding production line to execute the production task.

[0140] The generated daily production task list is automatically sent to the Manufacturing Execution System (MES) in the workshop or the corresponding industrial control computer on the production line via application programming interface (API) or file transfer. After receiving the task list, the production line operators or automation system can drive the production line to execute production tasks according to the plan, realizing closed-loop management from planning to execution.

[0141] In this embodiment of the invention, after step S106, a further step of sorting multiple production tasks on the same production line on the same day is included, specifically:

[0142] S301. Obtain all order lines and their production volumes for the production line to be scheduled within the planned date.

[0143] Specifically, taking "Production Line A" on "Monday" as an example, obtain all order lines and their corresponding quantities that are planned to be produced on Production Line A on Monday.

[0144] S302. Calculate the continuity score of each order line to be scheduled for production according to the preset continuity scoring rules; wherein the continuity scoring rules prioritize arranging the order lines of the same product, the same customer, or the same order adjacent to each other.

[0145] In this embodiment of the invention, a set of continuity scoring rules can be pre-designed. The core idea of ​​this rule is similarity priority, that is, arranging order lines of the same product, customer, or order adjacently will obtain a higher continuity score, because this can reduce the time spent on changing products or materials. In this way, a total score can be calculated for each possible sorting combination.

[0146] S303. Based on the continuity score from high to low, generate the work execution sequence of the production line within the planned date.

[0147] Finally, the sorting scheme with the highest continuity score is selected to generate the final job execution sequence. For example, the sorting result might be: first produce all the red T-shirts for customer X, then produce all the blue T-shirts for customer X, and finally produce the hoodies for customer Y.

[0148] S304. Check the production scheduling segment at the end of the current planned date and the production scheduling segment at the beginning of the next planned date. When the production output of both production scheduling segments is lower than a preset threshold and belongs to the same order line, merge the two production scheduling segments to reduce the intermittent switching of the production line.

[0149] Finally, examine production tasks such as the last production task on "Monday" and the first production task on "Tuesday". If these two tasks belong to the same order line and their respective production volumes are below a preset threshold (indicating they are both small-batch tasks), determine whether merging them for production would be more efficient.

[0150] If the judgment result is yes, then Tuesday's tasks will be moved to the end of Monday, or Monday's tasks will be postponed to the beginning of Tuesday, merging the two small-batch tasks into a single larger-batch task for continuous production. This measure can effectively reduce production line interruptions caused by lunch breaks, off-duty hours, etc., and further improve production efficiency.

[0151] The technical solution provided by this invention may include the following beneficial effects:

[0152] A production line scheduling and job sequencing method based on tag matching and scoring is proposed. This method combines a tag matching and scoring mechanism with a scheduling optimization model to achieve quantitative assessment of the fit between orders and production lines and automatic solution of the globally optimal scheduling plan. By performing fragmented correction and job sequencing optimization on the initial scheduling results, the continuity and efficiency of production tasks are achieved. This results in improved production line utilization, reduced changeover losses, guaranteed on-time order delivery, and effectively reduced production management complexity.

[0153] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0154] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0155] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory / random access memory, magnetic disk, optical disk) and includes several instructions to cause a multimedia terminal device (which may be a mobile phone, computer, television receiver, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0156] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A production line scheduling and job sequencing method based on label matching and scoring, characterized in that, Includes the following steps: S101. Obtain the set of order lines, the set of production lines, and the set of planned dates; where each order line contains process attributes and each production line contains process capabilities; S102. Based on the preset tag dimensions, map the process attributes of the order line to order tags, and map the process capabilities of the production line to production line tags; S103. Calculate the tag matching score between each order line and each production line based on the order tag and the production line tag; S104. Construct a production scheduling optimization model, wherein the production scheduling optimization model aims to minimize the total production scheduling cost and includes constraints; wherein the total production scheduling cost includes the matching penalty cost calculated based on the label matching score; S105. Solve the production scheduling optimization model to obtain the initial production output of each order line on the corresponding production line and the planned date; S106. Based on the initial production output, generate a production line scheduling plan, and control the corresponding production line to carry out production according to the production line scheduling plan.

2. The method according to claim 1, characterized in that, Step S103, calculating the tag matching score between each order line and each production line based on the order tag and production line tag, includes the following sub-steps: S1031. Obtain the weight of each of the label dimensions, and the score of the production line for different label values ​​in each of the label dimensions; S1032. For any combination of the order line and the production line, query the score of the production line in the corresponding dimension based on the value of the order tag of the order line in each dimension; S1033. Multiply the scores of each dimension by their corresponding weights and sum them to obtain the tag matching score between the order line and the production line.

3. The method according to claim 1, characterized in that, The constraints in step S104 include: Production line capacity constraint: used to limit the cumulative output of each production line within each planned date to not exceed the upper limit of the available capacity of that production line; Order completion constraint: Used to ensure that the cumulative production output of each order line over all the planned dates meets the total demand for that order line; and at least one of the following constraints: Earliest start date constraint for order lines: This constraint restricts the production scheduling date of the order line from being earlier than its earliest start date. Manufacturability constraint for order lines: This constraint restricts the production of the order line to only on the permitted production lines. Online inspection constraint: Used to ensure that the production output of the order line on the specified inspection date is not lower than the preset minimum batch size; Proportional inspection constraint: This is used to ensure that the cumulative production output completed by the order line before the specified inspection date reaches a preset target proportion.

4. The method according to claim 1, characterized in that, After step S105 and before step S106, the following steps are also included: S201. Identify the order lines where the remaining production to be scheduled is lower than a preset threshold; S202. The remaining production to be scheduled in the order line is preferentially allocated to the production line-date combination that already has production records in the order line, so as to reduce the number of production line switching times.

5. The method according to claim 4, characterized in that, Step S202 is followed by the following steps: S203. Identify fragmented production scheduling segments whose production output is lower than a preset threshold; S204. Migrate the fragmented production schedule fragments to production line-date combinations that are closer in time, have sufficient capacity, and match the label of the order line at a higher score.

6. The method according to claim 1, characterized in that, Following step S106, the following steps are also included: S301. Obtain all order lines and their production volumes for the production line to be scheduled within the planned date. S302. Calculate the continuity score of each order line to be scheduled for production according to the preset continuity scoring rules; wherein, the continuity scoring rules prioritize arranging the order lines of the same product, the same customer, or the same order adjacent to each other; S303. Based on the continuity score from high to low, generate the work execution sequence of the production line within the planned date.

7. The method according to claim 6, characterized in that, In step S303, when generating the job execution sequence based on the continuity score, if multiple order lines have the same continuity score, they are sorted according to the urgency of their latest completion date, and the order line with the earlier latest completion date is given priority.

8. The method according to claim 6, characterized in that, Step S303 is followed by the following steps: S304. Check the production scheduling segment at the end of the current planned date and the production scheduling segment at the beginning of the next planned date. When the production output of both production scheduling segments is lower than a preset threshold and belongs to the same order line, merge the two production scheduling segments to reduce the intermittent switching of the production line.

9. The method according to claim 1, characterized in that, The production scheduling optimization model in step S104 adopts a mixed integer programming model and is solved by a mathematical programming solver.

10. The method according to claim 1, characterized in that, Step S106, which involves generating a production line scheduling plan based on the initial production output, includes the following sub-steps: S1061. Aggregate the initial production output according to the production line and the planned date to generate different daily production task lists for the production line; S1062. The daily production task list is sent to the corresponding production line control system to drive the corresponding production line to execute the production task.

Images