A rigging product multi-process collaborative production scheduling optimization method and system

Abstract

The application discloses a rigging product multi-process collaborative production scheduling optimization method and system, and belongs to the technical field of intelligent manufacturing and manufacturing execution system. The method comprises the following steps: determining the production constraint relationship of each process; collecting the real-time running state of each process, determining the current bottleneck process in combination with the production constraint relationship, and calculating the dynamic effective capacity of the bottleneck process; based on the production constraint relationship and the dynamic effective capacity of the bottleneck process, performing compatibility clustering on the work-in-process in the buffer zone before the bottleneck process, generating a candidate production batch; obtaining the batch saturation, work-in-process waiting time and order delivery period of the candidate production batch, and determining the production time of the candidate production batch according to the batch saturation, work-in-process waiting time and order delivery period; in combination with the production constraint relationship and the dynamic effective capacity of the bottleneck process, reversely determining the production rhythm of the previous process from the production time of the candidate production batch, and generating a multi-process collaborative production scheduling scheme.

Description

Technical Field

[0001] This application relates to the field of intelligent manufacturing and manufacturing execution systems, and more specifically, to a method and system for optimizing the collaborative production scheduling of rigging products across multiple processes. Background Technology

[0002] Rigging products (such as wire rope slings and chain slings) are widely used in lifting, port logistics, construction, and ship mooring. Their production process typically involves multiple steps, including material preparation, stranding (braiding), joint pressing, heat treatment, surface treatment, and inspection and packaging. The equipment types, processing capacities, and production cycles for each step vary significantly. Furthermore, orders generally feature a wide variety of products, small batch sizes, and large specification differences. Adjacent steps are usually connected via a buffer zone. These production characteristics place high demands on multi-step collaborative production scheduling, but existing scheduling methods still have the following shortcomings: Firstly, existing production scheduling methods typically presuppose a certain process as a fixed bottleneck process and formulate production plans accordingly. However, the actual effective capacity of each process in rigging products is dynamically affected by factors such as equipment operating status, changeovers, and rework of defective work-in-process. The actual bottleneck process may shift with changes in order structure and fluctuations in equipment status (i.e., "bottleneck drift"). Using a fixed bottleneck assumption is difficult to adapt to the dynamic changes in the production site, which can easily lead to a mismatch between the production scheduling plan and the actual capacity, resulting in waiting between processes or wasted capacity.

[0003] Secondly, existing production scheduling methods typically only group work-in-process (WIP) in the buffer zone before bottleneck processes into batches based on order sequence or single specification, without fully considering the process compatibility between WIP in terms of material specifications, tooling, and fixture sharing. This can easily lead to incompatibility of processes within the same batch, frequent changes in design and mold, and may also cause some WIP to remain in the buffer zone due to the batch not being full for a long time, affecting production efficiency and equipment utilization.

[0004] Third, existing production scheduling methods typically rely solely on whether a batch has reached full capacity when determining the timing of its production launch. This approach fails to consider the waiting time of work-in-process within the buffer zone and the urgency of the corresponding orders. Consequently, it is difficult to strike a balance between ensuring batch saturation and guaranteeing on-time order delivery, which can easily lead to delays in the delivery of some urgent orders or excessive waiting times for some batches in order to reach full capacity.

[0005] Fourth, existing production scheduling methods usually formulate production plans for each process independently according to the process route sequence. They fail to fully consider the reverse pull effect of the actual production rhythm of the bottleneck process on the preceding process, and also fail to consider the impact of the defects and losses of each preceding process on the amount of material transferred. This can easily lead to poor connection between the preceding process and the bottleneck process, resulting in the bottleneck process running out of material due to insufficient material supply, or the preceding process accumulating material due to excessive output.

[0006] Fifth, existing scheduling methods typically rely on manual experience to adjust scheduling plans periodically or irregularly. They lack automatic detection mechanisms for bottleneck process drift and mechanisms for closed-loop iterative correction of scheduling parameters based on production execution feedback data. This makes it difficult for scheduling plans to adapt and update in a timely and accurate manner to changes in actual production conditions. After long-term operation, the parameters used for scheduling may deviate significantly from actual production conditions, thus affecting the effectiveness of the scheduling plan.

[0007] In summary, how to dynamically identify and effectively control bottleneck processes in the multi-process production of rigging products, rationally plan candidate production batches and production timing, and adaptively optimize the production scheduling plan based on changes in actual production status has become an urgent technical problem to be solved. Summary of the Invention

[0008] To overcome a series of shortcomings in the existing technology, the purpose of this application is to provide a method for optimizing the multi-process collaborative production scheduling of rigging products, which includes the following steps: The production constraints of each process are determined based on the process route information, process compatibility rules and work-in-process status data of rigging products. Collect the real-time operating status of each process, and determine the current bottleneck process in combination with production constraints, and calculate the dynamic effective capacity of the bottleneck process; Based on production constraints and the dynamic effective capacity of bottleneck processes, compatibility clustering is performed on the work-in-process in the buffer zone before the bottleneck process to generate candidate production batches. Obtain the batch saturation, work-in-process waiting time, and order delivery date of the candidate production batch, and determine the timing of production of the candidate production batch based on the batch saturation, work-in-process waiting time, and order delivery date; By combining production constraints and the dynamic effective capacity of bottleneck processes, the production schedule of upstream processes is determined in reverse from the production timing of candidate production batches, generating a multi-process collaborative production scheduling scheme.

[0009] In some embodiments, the method for obtaining the process route information of rigging products is as follows: Obtain the product categories and specifications of orders awaiting production; Match the corresponding process route template according to the product category, and modify the process route template based on the specification parameters to generate the process route corresponding to the order to be scheduled for production. Based on the process route, determine the processing sequence, equipment configuration, tooling and fixture configuration, time quota, and material correspondence of each process to obtain the process route information for rigging products.

[0010] In some embodiments, the method for determining the current bottleneck process is as follows: Acquire operational status data for each process within a preset monitoring period; The operational status data is normalized and then fused based on preset weight coefficients to obtain the comprehensive bottleneck score for each process. Based on the comprehensive bottleneck score, the process with the highest score that exceeds the preset bottleneck judgment threshold is selected as the candidate bottleneck process. When there are multiple candidate bottleneck processes, the impact of each candidate bottleneck process on the production cycle time is evaluated based on capacity constraints. The candidate bottleneck process that has the greatest impact on the overall production cycle time is identified as the current bottleneck process.

[0011] In some embodiments, the method for obtaining dynamic effective capacity is as follows: Obtain the theoretical capacity, historical equipment efficiency, and dynamic disturbance information related to capacity changes of the bottleneck process within the current monitoring period; The overall efficiency of historical equipment is updated on a rolling basis based on historical operating data, and the changeover and mold change time is determined based on the differences in work-in-process specifications of candidate production batches. Based on the updated historical equipment overall efficiency, changeover and mold changeover time, and work-in-process defective product rework ratio, calculate the capacity loss corresponding to the bottleneck process. Based on theoretical capacity and capacity loss, determine the dynamic effective capacity of the bottleneck process within the current time window.

[0012] In some embodiments, the method for compatibility clustering of work-in-process in the buffer zone before the bottleneck process is as follows: Obtain multidimensional attribute data of each work-in-process within the buffer zone before the bottleneck process. The multidimensional attribute data should include at least material specification attributes, order priority attributes, and tooling shared attributes. Based on multidimensional attribute data, construct clustering feature vectors for each work-in-process; A compatibility constraint space is constructed based on clustering feature vectors, and clustering is performed within the compatibility constraint space to obtain candidate production batches; Based on preset batch constraint rules, candidate production batches are verified and adjusted to obtain candidate production batches that meet the batch constraints.

[0013] In some embodiments, the method for obtaining the batch saturation, work-in-process waiting time, and order delivery date of a candidate production batch is as follows: Obtain the clustered work-in-process quantity and the single-batch optimal processing batch of the bottleneck process in the candidate production batch, and determine the batch saturation based on the clustered work-in-process quantity and the single-batch optimal processing batch. The batch saturation is compared with a preset saturation threshold, and the basic production conditions for the candidate production batch are determined based on the comparison results. When the batch saturation does not reach the preset saturation threshold, the waiting time and corresponding order delivery date of each work-in-process in the candidate production batch are obtained, and the timing of production is further evaluated based on the waiting time and the order delivery date. The waiting time for the work-in-process is calculated based on the time difference between the timestamp of the work-in-process arriving in the buffer and the current time, and the waiting time is compared with a preset waiting time upper limit. When the waiting time of any work-in-process exceeds the upper limit of the waiting time, a forced merging mechanism is triggered to merge the work-in-process into the nearest candidate production batch that is process-compatible. The urgency coefficient of order delivery date is determined based on the time difference between the order delivery date and the current date, historical delay records, and customer level, and the production priority of candidate production batches is adjusted based on the urgency coefficient.

[0014] In some embodiments, the method for determining the timing of production of a candidate production batch is as follows: Obtain batch saturation, work-in-process waiting time, and order delivery date information for candidate production batches, and construct a set of production evaluation indicators; The indicators in the production evaluation index set are normalized and weighted and integrated to obtain the comprehensive production priority score for each candidate production batch. The production order of candidate production batches is determined based on the overall production priority score; By combining the current time with the preset production window, the production start time is determined, and production is triggered at the production start time.

[0015] In some embodiments, the method for determining the production start-up schedule of the preceding process in reverse is as follows: Determine the target commissioning time and target production cycle for the bottleneck process based on the timing of commissioning and dynamic effective capacity. Determine the unit time quota for each preceding process and the safety stock requirements between adjacent processes based on the process route information. Based on the target production time, and combined with the unit time quota and safety stock requirements, the production time and quantity of each preceding process are determined step by step using the cycle time backwards method. When multiple parallel upstream processes converge into the bottleneck process, the cycle time is calculated backward for each parallel upstream process. The feasibility of the cycle time calculation results is verified based on the capacity constraints of each upstream process, and the commissioning time and quantity of the corresponding upstream process are adjusted when the cycle time constraints are not met.

[0016] In some embodiments, it also includes: Continuously collect the real-time operating status of each process, and detect whether the bottleneck process has drifted based on the real-time operating status of the process and the production constraint relationship; If a bottleneck process drift is detected, the current bottleneck process is redefined, and the dynamic effective capacity calculation of the bottleneck process, the generation of candidate production batches, and the determination of the production rhythm of the preceding process are re-executed in sequence, and the multi-process collaborative production scheduling plan is updated. If no bottleneck process drift is detected, the current multi-process collaborative production scheduling scheme will be maintained. Obtain production execution feedback data, and iteratively correct the process operation parameters, bottleneck process dynamic effective capacity calculation parameters, and candidate production batch commissioning parameters in the production constraint relationship based on the production execution feedback data to obtain an optimized multi-process collaborative production scheduling scheme.

[0017] This application also aims to provide a multi-process collaborative production scheduling optimization system for rigging products, used to implement the above-mentioned multi-process collaborative production scheduling optimization method for rigging products, including: The constraint relationship determination module is used to determine the production constraint relationships of each process based on the process route information, process compatibility rules and work-in-process status data of the rigging product. The bottleneck capacity calculation module is used to collect the real-time operating status of each process, determine the current bottleneck process in combination with production constraints, and calculate its dynamic effective capacity. The batch scheduling module is used to perform compatibility clustering of work-in-process in the buffer zone before the bottleneck process to generate candidate production batches based on production constraints and the dynamic effective capacity of the bottleneck process. It determines the timing of production launch by combining batch saturation, work-in-process waiting time and order delivery date, and reversely determines the production rhythm of the preceding process to generate a multi-process collaborative production scheduling plan.

[0018] Compared with the prior art, this application has the following beneficial effects: This application achieves coordinated scheduling of production resources across multiple processes in rigging products by dynamically identifying production bottlenecks, optimizing batch production timing, and coordinating the cycle times of upstream and downstream processes. This reduces waiting time and capacity waste in bottleneck processes, and improves the flexibility, stability, and overall production efficiency of production scheduling. Attached Figure Description

[0019] Figure 1This is a schematic diagram of the overall process of a multi-process collaborative production scheduling optimization method for rigging products provided in an embodiment of the present invention; Figure 2 This is a flowchart illustrating the method for determining the multi-process production constraint relationship of rigging products in an embodiment of the present invention; Figure 3 This is a flowchart illustrating the bottleneck process identification and dynamic effective capacity calculation method in an embodiment of the present invention. Figure 4 This is a flowchart illustrating the method for generating candidate production batches and determining the timing of production in an embodiment of the present invention. Figure 5 This is a flowchart illustrating the bottleneck process drift detection and production execution feedback data iterative optimization method in an embodiment of the present invention. Detailed Implementation

[0020] To make the technical solution, implementation logic, and technical effects of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and specific application scenarios. This embodiment takes a multi-process collaborative production workshop for wire rope slings and chain slings of a rigging manufacturing enterprise as a specific application scenario. The typical process route of this workshop includes blanking, stranding (braiding), joint pressing, heat treatment, surface treatment, and inspection and packaging. Adjacent processes are set up in product buffer zones. It has the production characteristics of many product varieties, small batches, large differences in processing capacity of each process, and bottleneck processes that are prone to drift with changes in order structure. It is suitable to use the multi-process collaborative production scheduling optimization method and system described in this embodiment for dynamic scheduling and rolling optimization.

[0021] In a broad embodiment of the present invention, a method for optimizing the scheduling of multi-process collaborative production of rigging products includes the following steps: The production constraints of each process are determined based on the process route information, process compatibility rules and work-in-process status data of rigging products. Collect the real-time operating status of each process, and determine the current bottleneck process in combination with production constraints, and calculate the dynamic effective capacity of the bottleneck process; Based on production constraints and the dynamic effective capacity of bottleneck processes, compatibility clustering is performed on the work-in-process in the buffer zone before the bottleneck process to generate candidate production batches. Obtain the batch saturation, work-in-process waiting time, and order delivery date of the candidate production batch, and determine the timing of production of the candidate production batch based on the batch saturation, work-in-process waiting time, and order delivery date; By combining production constraints and the dynamic effective capacity of bottleneck processes, the production schedule of upstream processes is determined in reverse from the production timing of candidate production batches, thus generating a multi-process collaborative production scheduling scheme. Continuously collect the real-time operating status of each process, and detect whether the bottleneck process has drifted based on the real-time operating status of the process and the production constraint relationship; If a bottleneck process drift is detected, the current bottleneck process is redefined, and the dynamic effective capacity calculation of the bottleneck process, the generation of candidate production batches, and the determination of the production rhythm of the preceding process are re-executed in sequence, and the multi-process collaborative production scheduling plan is updated. If no bottleneck process drift is detected, the current multi-process collaborative production scheduling scheme will be maintained. Obtain production execution feedback data, and iteratively correct the process operation parameters, bottleneck process dynamic effective capacity calculation parameters, and candidate production batch commissioning parameters in the production constraint relationship based on the production execution feedback data to obtain an optimized multi-process collaborative production scheduling scheme.

[0022] Furthermore, the method for obtaining process route information for rigging products is as follows: Obtain the product categories and specifications of orders awaiting production; Match the corresponding process route template according to the product category, and modify the process route template based on the specification parameters to generate the process route corresponding to the order to be scheduled for production. Based on the process route, determine the processing sequence, equipment configuration, tooling and fixture configuration, time quota, and material correspondence of each process to obtain the process route information for rigging products.

[0023] Furthermore, the method for obtaining the process compatibility rules for rigging products is as follows: Obtain the equipment processing parameters, material specification parameters, and process parameters corresponding to each process; Based on the equipment processing parameters, material specifications, and process parameters, establish the compatibility parameter range for each process. Determine the compatibility relationship between processes based on the range of compatibility parameters, and generate corresponding compatibility judgment conditions; Based on compatibility relationships and compatibility determination criteria, generate process compatibility rules for rigging products.

[0024] Furthermore, the method for obtaining the work-in-process status data of rigging products is as follows: Obtain the production identification information and production execution information corresponding to each work-in-process; Establish the association between work-in-process and order information based on production identification information, and determine the corresponding status parameters; Status parameters are updated in real time based on production execution information; Generate work-in-process status data for rigging products based on the updated status parameters.

[0025] Furthermore, the method for determining production constraints is as follows: Obtain the constraint parameters corresponding to capacity constraints, equipment mutual exclusion constraints, material sharing constraints, and quality cross-contamination constraints; Establish a corresponding constraint rule library based on the constraint parameters; The constraint weights of each constraint rule are determined based on the product type, order priority, delivery cycle, and production scale of the current production order. Generate a constraint association matrix based on each constraint rule and its corresponding constraint weight, and determine the production constraint relationship based on the constraint association matrix.

[0026] Furthermore, the method for collecting the real-time operating status of each process is as follows: Collect status information of equipment during operation in each process based on a distributed process sensing system; Based on the dynamic changes in the processing cycle of each process, the corresponding data sampling period is adaptively adjusted. The status information is sampled in stages according to the adjusted data sampling period to obtain real-time operating status data. Standardize and uniformly encode real-time operating status data to generate process operating status data; The data on the operation status of each process are collected and stored centrally to form a dataset of the operation status of the entire process.

[0027] Furthermore, the method for identifying the current bottleneck process is as follows: Acquire operational status data for each process within a preset monitoring period; The operational status data is normalized and then fused based on preset weight coefficients to obtain the comprehensive bottleneck score for each process. Based on the comprehensive bottleneck score, the process with the highest score that exceeds the preset bottleneck judgment threshold is selected as the candidate bottleneck process. When there are multiple candidate bottleneck processes, the impact of each candidate bottleneck process on the production cycle time is evaluated based on capacity constraints. The candidate bottleneck process that has the greatest impact on the overall production cycle time is identified as the current bottleneck process.

[0028] Furthermore, the method for obtaining dynamic effective production capacity is as follows: Obtain the theoretical capacity, historical equipment efficiency, and dynamic disturbance information related to capacity changes of the bottleneck process within the current monitoring period; The overall efficiency of historical equipment is updated on a rolling basis based on historical operating data, and the changeover and mold change time is determined based on the differences in work-in-process specifications of candidate production batches. Based on the updated historical equipment overall efficiency, changeover and mold changeover time, and work-in-process defective product rework ratio, calculate the capacity loss corresponding to the bottleneck process. Based on theoretical capacity and capacity loss, determine the dynamic effective capacity of the bottleneck process within the current time window.

[0029] Furthermore, the method for compatibility clustering of work-in-process in the buffer zone before the bottleneck process is as follows: Obtain multidimensional attribute data of each work-in-process within the buffer zone before the bottleneck process. The multidimensional attribute data should include at least material specification attributes, order priority attributes, and tooling shared attributes. Based on multidimensional attribute data, construct clustering feature vectors for each work-in-process; A compatibility constraint space is constructed based on clustering feature vectors, and clustering is performed within the compatibility constraint space to obtain candidate production batches; Based on preset batch constraint rules, candidate production batches are verified and adjusted to obtain candidate production batches that meet the batch constraints.

[0030] Furthermore, the method for obtaining the batch saturation, work-in-process waiting time, and order delivery date of the candidate production batch is as follows: Obtain the clustered work-in-process quantity and the single-batch optimal processing batch of the bottleneck process in the candidate production batch, and determine the batch saturation based on the clustered work-in-process quantity and the single-batch optimal processing batch. The batch saturation is compared with a preset saturation threshold, and the basic production conditions for the candidate production batch are determined based on the comparison results. When the batch saturation does not reach the preset saturation threshold, the waiting time and corresponding order delivery date of each work-in-process in the candidate production batch are obtained, and the timing of production is further evaluated based on the waiting time and the order delivery date. The waiting time for the work-in-process is calculated based on the time difference between the timestamp of the work-in-process arriving in the buffer and the current time, and the waiting time is compared with a preset waiting time upper limit. When the waiting time of any work-in-process exceeds the upper limit of the waiting time, a forced merging mechanism is triggered to merge the work-in-process into the nearest candidate production batch that is process-compatible. The urgency coefficient of order delivery date is determined based on the time difference between the order delivery date and the current date, historical delay records, and customer level, and the production priority of candidate production batches is adjusted based on the urgency coefficient.

[0031] Furthermore, the method for determining the timing of production for candidate production batches is as follows: Obtain batch saturation, work-in-process waiting time, and order delivery date information for candidate production batches, and construct a set of production evaluation indicators; The indicators in the production evaluation index set are normalized and weighted and integrated to obtain the comprehensive production priority score for each candidate production batch. The production order of candidate production batches is determined based on the overall production priority score; By combining the current time with the preset production window, the production start time is determined, and production is triggered at the production start time.

[0032] Furthermore, the method for determining the production schedule of the preceding process in reverse is as follows: Determine the target commissioning time and target production cycle for the bottleneck process based on the timing of commissioning and dynamic effective capacity. Determine the unit time quota for each preceding process and the safety stock requirements between adjacent processes based on the process route information. Based on the target production time, and combined with the unit time quota and safety stock requirements, the production time and quantity of each preceding process are determined step by step using the cycle time backwards method. When multiple parallel upstream processes converge into the bottleneck process, the cycle time is calculated backward for each parallel upstream process. The feasibility of the cycle time calculation results is verified based on the capacity constraints of each upstream process, and the commissioning time and quantity of the corresponding upstream process are adjusted when the cycle time constraints are not met.

[0033] Furthermore, the method for generating multi-process collaborative production scheduling schemes is as follows: Determine the resource allocation parameters for each process. The resource allocation parameters should include at least the equipment allocation information, production time window, planned processing quantity, tooling changeover sequence, and shift handover plan. A preliminary collaborative scheduling scheme for each process is generated based on process constraints. Based on the spatial layout information of adjacent processes and historical transfer time data, the material transfer time reserve is determined, and the material transfer time reserve is embedded into the preliminary collaborative scheduling scheme. Based on the allowance for material transfer time, the preliminary collaborative scheduling plan is supplemented with time dimension to form a complete multi-process collaborative production scheduling plan that includes process processing time and inter-process connection time.

[0034] Furthermore, the method for determining whether a bottleneck process has drifted is as follows: Acquire the operational status data of each process within two adjacent monitoring cycles; Calculate the bottleneck score for each process based on the operational status data, and determine the change in bottleneck score for each process between adjacent monitoring cycles. The bottleneck score change is compared with a preset drift threshold to identify bottleneck risk processes. The current bottleneck process is re-determined based on the bottleneck scores of each process within the current monitoring period; When the current bottleneck process is different from the bottleneck process in the previous monitoring cycle, and the difference in bottleneck scores between the two exceeds the preset switching threshold, it is determined that the bottleneck process has drifted.

[0035] Furthermore, the method for iterative correction based on production execution feedback data is as follows: The production execution feedback data from multiple historical monitoring periods are weighted and smoothed to generate corrected feedback data. The weight of historical data decreases over time and the data closer to the current moment has a higher weight. Update the process operation parameters in the production constraint relationship based on the corrected feedback data; Update the dynamic effective capacity calculation parameters for bottleneck processes based on the corrected feedback data; Update the production parameters for candidate production batches based on the corrected feedback data; The production execution feedback data is repeatedly acquired and the parameters are updated according to the preset correction cycle to complete the iterative correction of the production scheduling parameters.

[0036] A multi-process collaborative production scheduling optimization system for rigging products, characterized in that it includes: The constraint relationship determination module is used to determine the production constraint relationships of each process based on the process route information, process compatibility rules and work-in-process status data of the rigging product. The bottleneck capacity calculation module is used to collect the real-time operating status of each process, determine the current bottleneck process in combination with production constraints, and calculate its dynamic effective capacity. The batch scheduling module is used to perform compatibility clustering of work-in-process in the buffer zone before the bottleneck process to generate candidate production batches based on production constraints and the dynamic effective capacity of the bottleneck process. It determines the timing of production launch by combining batch saturation, work-in-process waiting time and order delivery date, and reversely determines the production rhythm of the preceding process to generate a multi-process collaborative production scheduling plan. The drift detection and update module is used to continuously detect whether the bottleneck process has drifted. If drift occurs, the batch scheduling module is triggered to re-execute the corresponding processing to update the scheduling plan. If no drift occurs, the current scheduling plan is maintained. The parameter correction module is used to iteratively correct production constraints, dynamic effective capacity calculation parameters of bottleneck processes, and production batch commissioning parameters based on production execution feedback data, so as to obtain an optimized scheduling scheme.

[0037] The preferred embodiments of the present invention are described below with reference to the accompanying drawings, and the present invention will be further described in detail.

[0038] like Figure 1 As shown, the multi-process collaborative production scheduling optimization method for rigging products provided in this embodiment includes the following steps: Step S101: Based on the process route information, process compatibility rules and work-in-process status data of the rigging product, determine the production constraint relationship of each process; Step S102: Collect the real-time operating status of each process, determine the current bottleneck process in combination with the production constraints, and calculate the dynamic effective capacity of the bottleneck process; Step S103: Based on production constraints and the dynamic effective capacity of bottleneck processes, perform compatibility clustering on the work-in-process in the buffer zone before the bottleneck processes to generate candidate production batches; Step S104: Obtain the batch saturation, work-in-process waiting time, and order delivery date of the candidate production batch to determine the timing for putting the candidate production batch into production; Step S105: Determine the production rhythm of the preceding process (including yield correction of the production quantity) in reverse from the production timing of the candidate production batch, and generate a multi-process collaborative production scheduling plan; Step S106: Continuously monitor whether the bottleneck process has drifted. If it has drifted, update the scheduling plan. If it has not drifted, maintain the current scheduling plan. Step S107: Obtain production execution feedback data, iteratively correct the production constraint relationship, the dynamic effective capacity calculation parameters of the bottleneck process and the production batch commissioning parameters, and output the optimized multi-process collaborative production scheduling scheme.

[0039] The following is combined Figures 2 to 5 The document provides specific calculation examples to illustrate the detailed implementation logic of data acquisition, processing and analysis, evaluation and decision-making, and feedback optimization involved in each of the above steps.

[0040] like Figure 2 As shown, this embodiment first obtains the process route information, process compatibility rules, and work-in-process status data of the rigging products, and determines the production constraints of each process accordingly. Specifically, the process route information is obtained by acquiring the product category (such as wire rope slings and chain rigging) and specification parameters (such as rope diameter, length, breaking strength, and joint type) of the orders to be scheduled for production. The corresponding process route template is matched according to the product category, and the equipment configuration, tooling and fixture configuration, and time quota in the template are corrected according to the specification parameters. The process compatibility rules are obtained by acquiring the equipment processing parameters (such as press tonnage and stranding machine speed range), material specification parameters (such as wire diameter range and rope core type), and process parameters (such as pressing pressure and holding time) corresponding to each process, establishing the compatibility parameter range for each process, and generating compatibility judgment conditions. The work-in-process status data is obtained by acquiring the production identification information (such as batch number and order number) and production execution information (such as the current process, completed quantity, and quality inspection results) of each work-in-process, establishing the association between work-in-process and order information, and updating it in real time.

[0041] Based on this, this embodiment obtains the constraint parameters corresponding to four types of constraint rules: capacity constraints, equipment mutual exclusion constraints, material sharing constraints, and quality cross-contamination constraints, and establishes a constraint rule library; according to the product type, order priority, delivery cycle, and production scale of the current production order, it determines the constraint weights corresponding to each type of constraint rule; and generates a constraint association matrix based on each constraint rule and its corresponding weight, with the elements in the matrix representing the process. With process The correlation strength between them under the combined effect of the four types of constraints is calculated as shown in Equation (1).

[0042] (1) In the above formula, Indicate process With process The constraint strength value between them is used to characterize the degree of mutual constraint between them in production scheduling. The larger the value, the more they need to be coordinated and the less they should be arbitrarily split for production scheduling. This represents the constraint weight corresponding to the c-th type of constraint rule (c takes values ​​from 1 to 4, corresponding to capacity constraints, equipment mutual exclusion constraints, material sharing constraints, and quality cross-contamination constraints, respectively). The weight is determined based on factors such as order priority and delivery cycle, and the sum of the four weights is 1. This represents the original correlation coefficient between process r and process s under the c-th type of constraint rule. Its value ranges from [0,1] and is obtained by comparing the corresponding constraint parameters in the constraint rule base. A larger value indicates a stronger constraint in that dimension. Based on the values ​​of each element in the constraint association matrix, the production constraint relationships between each pair of processes are determined, serving as the basic input for subsequent bottleneck process identification, batch clustering, and scheduling rhythm calculation.

[0043] like Figure 3 As shown, this embodiment is based on a distributed process perception system. It adaptively adjusts the data sampling period according to the dynamic changes in the processing cycle of each process, and collects the status information of each process equipment during operation (such as equipment load rate, amount of work-in-process, average processing cycle, and downtime due to fault). After standardization and unified coding, a full process operation status dataset is formed.

[0044] For the operational status data within the preset monitoring period, normalization is first performed according to formula (2) (when all processes under the same indicator have the same value, i.e.) and When they are equal, the normalized value of this indicator is set to 0 and it is not included in the subsequent weighted fusion. (2) In the above formula, Indicate process The original collected values ​​under the kth bottleneck evaluation index (such as equipment load rate, work-in-process inventory, average processing cycle time); express The normalized index value ranges from [0,1]. , This represents the maximum and minimum values ​​of all processes under the k-th index, used to eliminate dimensional differences.

[0045] Based on normalization, each indicator is weighted and integrated according to a preset weight coefficient to calculate the comprehensive bottleneck score corresponding to each process, as shown in equation (3): (3) In the above formula, This represents the overall bottleneck score of process i, with a value range of [0,1]. The higher the score, the greater the constraint that process has on the overall production cycle time. The weight coefficient corresponding to the k-th bottleneck evaluation indicator is pre-calibrated by process experts based on historical production data, and the sum of all weights is 1; n represents the total number of indicators participating in the bottleneck evaluation.

[0046] Select the bottleneck with the highest overall bottleneck score that exceeds the preset bottleneck determination threshold. The process is selected as a candidate bottleneck process. When there are multiple candidate bottleneck processes, the impact of each candidate bottleneck process on the production cycle time is evaluated based on the capacity constraints. The candidate bottleneck process with the greatest impact on the overall production cycle time is determined as the current bottleneck process.

[0047] Example: Taking the monitoring data of a certain period in the workshop of this embodiment as an example, the equipment load rate, work-in-process accumulation, and average processing cycle of the three processes of stranding, joint pressing, and heat treatment are normalized and weighted as follows: =0.4、 =0.3、 After weighted fusion (=0.3), the overall bottleneck score B for the joint pressing process is 0.86, for the twisted strand process it is 0.71, and for the heat treatment process it is 0.63. (Preset bottleneck determination threshold) =0.75, therefore the joint pressing process is identified as the current bottleneck process.

[0048] After identifying the current bottleneck process, further obtain the theoretical capacity, historical equipment comprehensive efficiency, and dynamic disturbance information related to capacity changes (changeover, defective product rework) of the process within the current monitoring period. It should be noted that, in order to avoid repeatedly deducting the same part of capacity loss, the historical equipment comprehensive efficiency η in this embodiment specifically represents the performance loss caused by fluctuation factors such as short-term equipment shutdown and reduced speed operation, other than changeover and defective product rework. It is obtained by rolling updates of historical operating data. The impact of changeover and defective product rework on capacity is calculated separately by changeover time and defective product rework ratio. The three do not overlap and together constitute the capacity loss of the bottleneck process, as shown in equation (4): (4) In the above formula, This indicates the capacity loss of the bottleneck process during the current monitoring period, expressed in units per shift. This indicates the theoretical capacity of the bottleneck process, expressed in pieces per shift. This represents the performance efficiency coefficient after rolling updates of historical operating data (excluding the impact of model change, mold change and defect rework), and its value range is (0,1]. This indicates the cumulative time spent on changeover and mold replacement within the current cycle, determined based on the differences in work-in-process specifications between candidate production batches, in minutes. This indicates the total duration of a single production cycle (shift) for the bottleneck process, in minutes; This represents the percentage of defective work-in-process products that need to be reworked in the bottleneck process, with a value range of [0,1].

[0049] Based on theoretical production capacity and capacity loss The dynamic effective capacity of the bottleneck process within the current time window is determined according to formula (5): (5) In the above formula, This represents the dynamic effective capacity of the bottleneck process within the current time window, measured in units per shift. It is a key input parameter for determining the timing of subsequent candidate production batches and for retrospectively calculating the pace of preceding processes. Example: Theoretical production capacity of the joint pressing process =200 pieces / shift, performance efficiency coefficient after rolling updates =0.95, total duration of a single shift =480 minutes, total time spent on changeover and mold replacement in this cycle. =25 minutes, defective product return rate =4%. From equation (4), we get =200×[1-0.95×(480-25) / 480×(1-0.04)]=200×(1-0.95×0.948×0.96)≈200×0.135≈27 pieces; From equation (5), the dynamic effective production capacity is obtained. =200-27=173 pieces / shift, which serves as the upper limit of the actual production capacity of the bottleneck process during the current monitoring period.

[0050] like Figure 4 As shown, multidimensional attribute data of each work-in-process within the buffer zone before the bottleneck process is obtained. The multidimensional attribute data includes at least material specification attributes, order priority attributes, and tooling shared attributes. Based on the multidimensional attribute data, a clustering feature vector is constructed for each work-in-process. To measure the process compatibility between any two work-in-process p and q, the compatibility distance between them in the compatibility constraint space is calculated according to equation (6): (6) In the above formula, Indicating work in progress With work in progress The compatibility distance between them, the smaller the value, the higher the degree of compatibility between the two processes; , Indicating work in progress , The standardized value under the h-th clustering feature (such as specification attribute, tooling shared attribute); This represents the weight coefficient corresponding to the h-th cluster feature; This represents the total number of feature dimensions involved in the clustering.

[0051] when Not exceeding the preset compatibility threshold At that time, determine the work in progress. and Processes are compatible and can be classified into the same candidate production batch. After clustering within the compatibility constraint space, the candidate production batches are verified and adjusted based on preset batch constraint rules (i.e., the number of work-in-process in a single candidate batch is limited to not exceeding the optimal processing batch of the bottleneck process), thus obtaining candidate production batches that meet the batch constraints.

[0052] For each candidate production batch Obtain the number of in-process items that have been clustered. and the optimal batch size for bottleneck processes Calculate the batch saturation according to formula (7): (7) In the above formula, Indicates candidate production batch Batch saturation reflects the degree to which the quantity of work-in-process in that batch is filled relative to the optimal batch size for a single processing run at the bottleneck process; due to Batch constraint rules limit the number of batches to no more than , The value range is [0,1]; Candidate production batch The number of work-in-process items that have been clustered, in units of pieces; This represents the optimal batch size for a single bottleneck process, determined by equipment process parameters and historical production data, and is expressed in pieces.

[0053] Batch saturation With preset saturation threshold Comparison: When Not less than When the candidate production batch is deemed to have met the basic conditions for production, it is directly placed into the production sequencing queue; when Not achieved At the same time, further obtain the waiting time of each work-in-process within the batch and the corresponding order delivery date, and conduct a supplementary assessment on whether to start production ahead of schedule (that is, allow batches that have not yet reached the saturation standard but have been waiting for a long time or have urgent delivery dates to start production ahead of schedule, so as to avoid rigid waiting leading to delivery delays). The waiting time of work-in-process is calculated according to formula (8): (8) In the above formula, This indicates the waiting time of work-in-process within the buffer zone; Indicates the current moment; This indicates the timestamp when the work-in-process arrives in the buffer.

[0054] When batch Waiting time for any work-in-process When the preset waiting time limit is exceeded, a forced merging mechanism is triggered to merge the work-in-process into a process that is compatible and has a compatibility distance. The smallest candidate production batch is selected to avoid individual work-in-process items becoming stuck due to prolonged unsaturation of the batch. Simultaneously, the batch size is determined by considering the time difference between the order delivery date and the current date, historical delay records, and customer rating. Each order included Calculate the order delivery urgency coefficient according to formula (9): (9) In the above formula, Indicates order The corresponding delivery urgency coefficient ranges from [0,1], with a larger value indicating a more urgent order. Indicates order The remaining delivery time is the time difference between the order delivery date and the current date. , This represents the maximum and minimum remaining delivery times for each order in the current pending production order pool; Indicates order The normalized historical delay record score ranges from [0,1], with a higher value for more historical delays. Indicates order The normalized customer level coefficient has a value range of [0,1], with the value increasing as the customer level increases. , , This represents the weighting coefficients for three factors: remaining delivery time, historical delay records, and customer level. The sum of these three factors is 1.

[0055] When the work-in-process contained in candidate production batch j comes from multiple orders, the batch number is taken. The longest waiting time for each work-in-process item is used as the representative waiting time for that batch. Take the urgency coefficient of each order's delivery date. The maximum value in the range is used as the representative urgency coefficient for that batch. This reflects the production reality that "the urgency of a batch depends on the most urgent work-in-process or order within it." After obtaining the batch saturation, representative waiting time, and representative urgency coefficient of candidate production batches, the three indicators are normalized respectively. Since it already takes values ​​in the range [0,1], it can be directly used as... use; Normalized based on the preset waiting time upper limit to obtain ; Since the value is already in the range [0,1], it can be directly used as... Using the formula (10), a weighted fusion is performed to obtain the comprehensive production priority score for the candidate production batch, which is used to determine the priority of production for the candidate production batch. Not achieved Batch sorting and determining whether it is worthwhile to start production ahead of schedule: (10) In the above formula, This represents the overall production priority score for candidate production batch j; the higher the score, the higher the priority for production. , , Indicates candidate production batch The normalized values ​​of batch saturation, waiting time, and delivery urgency coefficient; , , This represents the weighting coefficients for the three indicators: batch saturation, waiting time, and delivery urgency. The sum of these three is 1.

[0056] For candidate batches that have reached the saturation threshold, they are directly added to the production queue in the order of their formation; for candidate batches that have not reached the saturation threshold, they are added based on their overall production priority score. Sort from high to low, when If the preset early production threshold is exceeded, early production is allowed. The final production time for each candidate production batch is determined by combining the current time with the preset production window, and production is triggered at the production time to form the production plan for the bottleneck process.

[0057] Example: Candidate batch A has been clustered into work-in-process. =38 pieces, optimal batch size for bottleneck process in a single operation =45 pieces, from equation (7) we get ≈0.84, not reaching the preset saturation threshold. =0.85; Representative work-in-process waiting time within this batch. =24 hours, remaining delivery time for the order =5 days, current order pool =10 days =1 day, historical delay record rating =0.3, Customer level coefficient =0.8, take =0.5、 =0.2、 =0.3, from equation (9) we get ≈0.58, which is the urgency coefficient representing this batch. Take weight =0.4、 =0.3、 =0.3, normalized to the preset waiting time limit of 36 hours. =24 / 36≈0.67, from equation (10) we can obtain the overall production priority score for this batch. =0.4×0.84+0.3×0.67+0.3×0.58≈0.71, which exceeds the preset early production threshold of 0.65. Therefore, although this batch is not saturated, it is still allowed to start production ahead of schedule.

[0058] Based on the production timing of candidate production batches and the dynamic effective capacity of bottleneck processes, determine the target production time and target production quantity for the bottleneck processes. Since each preceding process also involves defects and losses during processing, if the preceding process directly follows... After production begins, the number of qualified work-in-process products reaching the bottleneck process will be insufficient due to losses. Therefore, the production quantity needs to be increased and corrected according to the historical pass rate of each preceding process, as shown in equation (11): (11) In the above formula, This indicates the planned production quantity of the preceding process m after yield correction; This indicates the target production quantity of the bottleneck process, which is the work-in-process quantity of the corresponding candidate batch after batch constraint verification. This represents the historical pass rate of the preceding process m, with a value range of (0,1], obtained from historical production data statistics.

[0059] Based on the process route information, determine the unit time quota for each preceding process and the safety stock requirements between adjacent processes; take the target production time of the bottleneck process as the benchmark, and combine it with the production quantity determined by formula (11). The unit time quota and safety stock requirements are determined step by step by using the cycle time backwards method to determine the production start time of each preceding process, as shown in equation (12): (12) In the above formula, This indicates the production start-up time corresponding to the preceding process m; This indicates the start-up time of the next process immediately preceding process m (closer to the bottleneck process). When process m is the direct preceding process of the bottleneck process, This refers to the target production start-up time for the bottleneck process; Indicate process The unit work hour quota; This represents the buffer time corresponding to the safety stock between process m and its downstream adjacent processes.

[0060] When multiple parallel upstream processes converge into the bottleneck process, the production quantity and cycle time are calculated for each parallel upstream process according to formulas (11) and (12). The cycle time calculation results are verified based on the capacity constraints of each upstream process. If a upstream process does not meet the cycle time constraints during the verification, the production time and production quantity of that process are adjusted to avoid the bottleneck process from running out of material due to insufficient capacity of the upstream process.

[0061] Example: Target production quantity for this batch of joint pressing process (bottleneck process). =45 pieces, with a target production start time of 14:00 on the same day. Its historical pass rate for the direct upstream process, the twisted strand process. =96%, from equation (11) we get =45 / 0.96≈47 pieces (rounded up), meaning that the twisting process needs to produce 47 pieces to compensate for approximately 4% process loss and ensure that 45 qualified work-in-process pieces reach the bottleneck process buffer. The single-piece time quota for the twisting process. =6 minutes, safety stock buffer time =20 minutes, from equation (12) we get =14:00-(47×6+20) minutes=14:00-302 minutes=08:58, that is, the twisting process should start production at 08:58 on the same day.

[0062] Determine the resource allocation parameters for each process (including at least equipment allocation information, production time window, planned processing quantity, tooling switchover sequence, and shift handover plan), and generate a preliminary collaborative scheduling plan for each process based on production constraints. Combine the spatial layout information of adjacent processes and historical transfer time data to determine the material transfer time reserve, and embed it into the preliminary collaborative scheduling plan. Complete the plan with a time dimension, and finally form a complete multi-process collaborative production scheduling plan that includes the processing time of each process and the connection time between processes.

[0063] like Figure 5 As shown, this embodiment continuously collects the real-time operating status of each process, obtains the operating status data of each process within two adjacent monitoring cycles, calculates the comprehensive bottleneck score of each process in adjacent monitoring cycles according to equations (2) and (3), and determines the change in bottleneck score of each process between adjacent monitoring cycles according to equation (13): (13) In the above formula, Indicate process The change in bottleneck score between two adjacent monitoring periods, a positive value indicates an increase in the bottleneck risk of the process. , Indicate process The comprehensive bottleneck score corresponding to the current monitoring period t and the previous monitoring period (t-1).

[0064] Will With preset drift threshold Compare and filter Greater than The process is designated as a bottleneck risk process; the current bottleneck process is re-determined based on the bottleneck scores of each process within the current monitoring period. If the re-determined current bottleneck process differs from the bottleneck process corresponding to the previous monitoring period, and the difference in their corresponding bottleneck scores exceeds a preset switching threshold... If a bottleneck process drift is detected, the dynamic effective capacity calculation for the bottleneck process, the generation of candidate production batches, and the determination of the production schedule of the preceding processes are re-executed sequentially to update the multi-process collaborative production scheduling plan. If no bottleneck process drift is detected, the current multi-process collaborative production scheduling plan is maintained to avoid frequent scheduling adjustments that could disrupt on-site production.

[0065] Example: The bottleneck score for the joint pressing process in the current monitoring cycle is 0.86, while the bottleneck score for the same process in the previous monitoring cycle was 0.78. According to equation (13), the change in bottleneck score is 0.08, which is greater than the preset drift threshold. =0.05; meanwhile, the bottleneck process identified in the previous monitoring cycle was the heat treatment process (score 0.79), which is different from the bottleneck process in this cycle (joint pressing). The score difference between the two is 0.86-0.79=0.07, which is greater than the preset switching threshold. =0.05, indicating that the bottleneck process has shifted from the heat treatment process to the joint pressing process, and the system immediately triggers a scheduling update.

[0066] To ensure that the scheduling scheme can be continuously optimized according to the actual situation on the production site, this embodiment obtains production execution feedback data (such as actual processing time, actual performance efficiency, actual defective product rework ratio, and actual material transfer time), and uses an exponential weighted recursive method to calculate the process operation parameters and bottleneck process dynamic effective capacity parameters in the production constraint relationship (such as in equation (4)). The parameters for production of candidate batches (such as γ_m in equation (11)) are iteratively corrected as shown in equation (14): (14) In the above formula, , Indicates the parameter to be corrected (such as the performance efficiency coefficient). Single-piece work time quota Historical pass rate Repair rate etc.) in the The first and second ( The value corresponding to each correction cycle; X(t) represents the representative value of the parameter collected by the production execution feedback system in the t-th correction cycle, which is obtained by taking the arithmetic average of multiple original feedback records in this cycle, in order to eliminate random noise of a single record; This represents the correction factor (learning rate), with a value range of (0,1). The larger the value, the faster the parameter converges to the representative value of the current cycle. Expanding equation (14) over time, we can see that the feedback data of each historical cycle affects the current parameter. Contribution weights are based on ( It exhibits exponential decay, thus naturally achieving a weighted smoothing effect where "feedback data closer to the current time has a higher weight," without requiring additional smoothing steps. The production execution feedback data is repeatedly acquired according to the preset correction cycle and the parameters are updated according to formula (14) to complete the iterative correction of the production scheduling parameters, forming a closed-loop optimization mechanism of "data acquisition - evaluation decision - execution feedback - parameter correction".

[0067] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for optimizing the collaborative production scheduling of rigging products across multiple processes, characterized in that, Includes the following steps: The production constraints of each process are determined based on the process route information, process compatibility rules and work-in-process status data of rigging products. Collect the real-time operating status of each process, and determine the current bottleneck process in combination with production constraints, and calculate the dynamic effective capacity of the bottleneck process; Based on production constraints and the dynamic effective capacity of bottleneck processes, compatibility clustering is performed on the work-in-process in the buffer zone before the bottleneck process to generate candidate production batches. Obtain the batch saturation, work-in-process waiting time, and order delivery date of the candidate production batch, and determine the timing of production of the candidate production batch based on the batch saturation, work-in-process waiting time, and order delivery date; By combining production constraints and the dynamic effective capacity of bottleneck processes, the production schedule of upstream processes is determined in reverse from the production timing of candidate production batches, generating a multi-process collaborative production scheduling scheme.

2. The method for optimizing multi-process collaborative production scheduling of rigging products according to claim 1, characterized in that, The method for obtaining process route information for rigging products is as follows: Obtain the product categories and specifications of orders awaiting production; Match the corresponding process route template according to the product category, and modify the process route template based on the specification parameters to generate the process route corresponding to the order to be scheduled for production. Based on the process route, determine the processing sequence, equipment configuration, tooling and fixture configuration, time quota, and material correspondence of each process to obtain the process route information for rigging products.

3. The method for optimizing multi-process collaborative production scheduling of rigging products according to claim 1, characterized in that, The method for identifying the current bottleneck process is as follows: Acquire operational status data for each process within a preset monitoring period; The operational status data is normalized and then fused based on preset weight coefficients to obtain the comprehensive bottleneck score for each process. Based on the comprehensive bottleneck score, the process with the highest score that exceeds the preset bottleneck judgment threshold is selected as the candidate bottleneck process. When there are multiple candidate bottleneck processes, the impact of each candidate bottleneck process on the production cycle time is evaluated based on capacity constraints. The candidate bottleneck process that has the greatest impact on the overall production cycle time is identified as the current bottleneck process.

4. The method for optimizing multi-process collaborative production scheduling of rigging products according to claim 1, characterized in that, The method for obtaining dynamic effective production capacity is as follows: Obtain the theoretical capacity, historical equipment efficiency, and dynamic disturbance information related to capacity changes of the bottleneck process within the current monitoring period; The overall efficiency of historical equipment is updated on a rolling basis based on historical operating data, and the changeover and mold change time is determined based on the differences in work-in-process specifications of candidate production batches. Based on the updated historical equipment overall efficiency, changeover and mold changeover time, and work-in-process defective product rework ratio, calculate the capacity loss corresponding to the bottleneck process. Based on theoretical capacity and capacity loss, determine the dynamic effective capacity of the bottleneck process within the current time window.

5. The method for optimizing multi-process collaborative production scheduling of rigging products according to claim 1, characterized in that, The method for compatibility clustering of work-in-process in the buffer zone before the bottleneck process is as follows: Obtain multidimensional attribute data of each work-in-process within the buffer zone before the bottleneck process. The multidimensional attribute data should include at least material specification attributes, order priority attributes, and tooling shared attributes. Based on multidimensional attribute data, construct clustering feature vectors for each work-in-process; A compatibility constraint space is constructed based on clustering feature vectors, and clustering is performed within the compatibility constraint space to obtain candidate production batches; Based on preset batch constraint rules, candidate production batches are verified and adjusted to obtain candidate production batches that meet the batch constraints.

6. The method for optimizing multi-process collaborative production scheduling of rigging products according to claim 1, characterized in that, The method for obtaining the batch saturation, work-in-process waiting time, and order delivery date of candidate production batches is as follows: Obtain the clustered work-in-process quantity and the single-batch optimal processing batch of the bottleneck process in the candidate production batch, and determine the batch saturation based on the clustered work-in-process quantity and the single-batch optimal processing batch. The batch saturation is compared with a preset saturation threshold, and the basic production conditions for the candidate production batch are determined based on the comparison results. When the batch saturation does not reach the preset saturation threshold, the waiting time and corresponding order delivery date of each work-in-process in the candidate production batch are obtained, and the timing of production is further evaluated based on the waiting time and the order delivery date. The waiting time for the work-in-process is calculated based on the time difference between the timestamp of the work-in-process arriving in the buffer and the current time, and the waiting time is compared with a preset waiting time upper limit. When the waiting time of any work-in-process exceeds the upper limit of the waiting time, a forced merging mechanism is triggered to merge the work-in-process into the nearest candidate production batch that is process-compatible. The urgency coefficient of order delivery date is determined based on the time difference between the order delivery date and the current date, historical delay records, and customer level, and the production priority of candidate production batches is adjusted based on the urgency coefficient.

7. The method for optimizing multi-process collaborative production scheduling of rigging products according to claim 1, characterized in that, The method for determining the timing of production launch for candidate production batches is as follows: Obtain batch saturation, work-in-process waiting time, and order delivery date information for candidate production batches, and construct a set of production evaluation indicators; The indicators in the production evaluation index set are normalized and weighted and integrated to obtain the comprehensive production priority score for each candidate production batch. The production order of candidate production batches is determined based on the overall production priority score; By combining the current time with the preset production window, the production start time is determined, and production is triggered at the production start time.

8. The method for optimizing the multi-process collaborative production scheduling of rigging products according to claim 1, characterized in that, The method for determining the production schedule of the preceding process in reverse is as follows: Determine the target commissioning time and target production cycle for the bottleneck process based on the timing of commissioning and dynamic effective capacity. Determine the unit time quota for each preceding process and the safety stock requirements between adjacent processes based on the process route information. Based on the target production time, and combined with the unit time quota and safety stock requirements, the production time and quantity of each preceding process are determined step by step using the cycle time backwards method. When multiple parallel upstream processes converge into the bottleneck process, the cycle time is calculated backward for each parallel upstream process. The feasibility of the cycle time calculation results is verified based on the capacity constraints of each upstream process, and the commissioning time and quantity of the corresponding upstream process are adjusted when the cycle time constraints are not met.

9. The method for optimizing the multi-process collaborative production scheduling of rigging products according to any one of claims 1-8, characterized in that, Also includes: Continuously collect the real-time operating status of each process, and detect whether the bottleneck process has drifted based on the real-time operating status of the process and the production constraint relationship; If a bottleneck process drift is detected, the current bottleneck process is redefined, and the dynamic effective capacity calculation of the bottleneck process, the generation of candidate production batches, and the determination of the production rhythm of the preceding process are re-executed in sequence, and the multi-process collaborative production scheduling plan is updated. If no bottleneck process drift is detected, the current multi-process collaborative production scheduling scheme will be maintained. Obtain production execution feedback data, and iteratively correct the process operation parameters, bottleneck process dynamic effective capacity calculation parameters, and candidate production batch commissioning parameters in the production constraint relationship based on the production execution feedback data to obtain an optimized multi-process collaborative production scheduling scheme.

10. A multi-process collaborative production scheduling optimization system for rigging products, used to implement the multi-process collaborative production scheduling optimization method for rigging products as described in claim 9, characterized in that, include: The constraint relationship determination module is used to determine the production constraint relationships of each process based on the process route information, process compatibility rules and work-in-process status data of the rigging product. The bottleneck capacity calculation module is used to collect the real-time operating status of each process, determine the current bottleneck process in combination with production constraints, and calculate its dynamic effective capacity. The batch scheduling module is used to perform compatibility clustering of work-in-process in the buffer zone before the bottleneck process to generate candidate production batches based on production constraints and the dynamic effective capacity of the bottleneck process. It determines the timing of production launch by combining batch saturation, work-in-process waiting time and order delivery date, and reversely determines the production rhythm of the preceding process to generate a multi-process collaborative production scheduling plan.

Images