Multi-process collaborative steel production order joint batching method based on dynamic feature updating

Abstract

The present application belongs to the technical field of intelligent manufacturing and production scheduling optimization of steel production, and particularly relates to a multi-process collaborative steel production order joint batching method based on dynamic feature updating. The method constructs a feature vector based on the multi-dimensional attributes of orders, introduces a multi-process consistency maintenance capability as a dynamic inhibition factor, screens and forms an initial order cluster. In the expansion process, the expansion is terminated under the condition of a bottleneck process, and a joint batch that meets the collaborative constraints is generated. When an order or device state disturbance occurs, a batch structure stability index is calculated to adaptively select a maintenance, local adjustment or overall reconstruction strategy, and dynamic maintenance of the scheme is realized. Through the "inhibition screening-bottleneck limitation-stability driven update" mechanism, the present application improves the consistency and executability of cross-process batches and reduces the reconstruction frequency.

Description

Technical Field

[0001] This invention belongs to the field of intelligent manufacturing and production scheduling optimization technology in steel production, specifically involving a multi-process collaborative steel production order batching method based on dynamic feature updates. Background Technology

[0002] Steel production is a typical flow-type manufacturing process, characterized by numerous steps, significant variations in cycle time, strong resource constraints, and tight coupling of process paths. It typically includes multiple sequentially linked production processes such as steelmaking, continuous casting, and hot rolling. While each process differs significantly in equipment capacity, production pace, process rules, and logistics organization, coordination in capacity, pace, and logistics is required at the production organization level. Failure to do so can easily lead to problems such as inventory buildup, bottleneck process congestion, decreased energy efficiency, and order delivery delays.

[0003] In existing technologies, steel production orders are typically batched independently within each process stage. Different processes form furnace batches, casting batches, or rolling batches based on their own technological characteristics and local optimization goals. For example, the steelmaking process focuses on steel grade stability and smelting rhythm, the continuous casting process emphasizes casting continuity, and the hot rolling process pays more attention to the continuity of specifications and the costs of roll and width changes. Due to the significant differences in batching goals and constraints among the various processes, the resulting batch structure is difficult to maintain consistency across the multi-process production chain. This often necessitates the splitting or reorganization of upstream batches in downstream processes, leading to reduced efficiency in cross-process production organization.

[0004] To alleviate the aforementioned problems, some existing technologies introduce multi-process collaborative scheduling or cross-process planning optimization methods. Based on a given single-process batch structure, these methods coordinate and optimize the work sequence, process cycle time, or cross-process time arrangements to reduce waiting time and improve production line matching. However, these methods typically rely on an existing batch structure, and their optimization focuses primarily on the time dimension or execution sequence level. They fail to provide a unified model of the batch formation mechanism at the order level, thus making it difficult to fundamentally solve the structural inconsistencies caused by independent batching of processes.

[0005] Meanwhile, as steel production shifts towards smaller batches, more diverse products, and greater flexibility, order insertions, order cancellations, delivery date adjustments, equipment status fluctuations, and changes in material availability are becoming the norm, resulting in a distinctly dynamic production system. Although companies have accumulated substantial data on order attributes, process paths, and equipment status, existing methods generally lack a technical means to uniformly model multi-dimensional attributes at the order level and dynamically maintain batch structure across multiple production processes amidst production disturbances. This makes it difficult to simultaneously ensure the stability of batch solutions and the executability of multiple processes in complex and dynamic environments.

[0006] In summary, the existing technology has at least the following shortcomings: (1) Order batching decisions are mainly made within a single process, lacking a unified batch structure that can be shared and reused across multiple production processes; (2) Information on the multi-dimensional attributes of the order, such as steel type, specifications, process path, delivery time and resource status, has not been systematically used for multi-process joint batching decision-making; (3) There is a lack of order cluster structure maintenance mechanism for production disturbances, and the batch scheme is not stable enough in dynamic environment; (4) It is difficult to simultaneously ensure the coordination of the entire process capacity and the feasibility of the joint batch structure under dynamic production conditions.

[0007] Therefore, in the context of multi-process collaborative production in the steel industry, it is necessary to propose a method and system for order joint batching that takes orders as the basic object, can form a joint batch structure under the constraints of multiple processes, and can be dynamically maintained with production disturbances, so as to improve the coordination, stability and execution efficiency of the entire steel production process. Summary of the Invention

[0008] To address the aforementioned technical problems, this invention provides a method for joint batching of multi-process collaborative steel production orders based on dynamic feature updates.

[0009] To achieve the above-mentioned technical objectives, the present invention provides the following technical solution: A method for joint batching of steel production orders based on multi-process collaboration, the method comprising the following steps: S1: Obtain production order information and production status information, and construct an order feature vector based on the multi-dimensional attributes of the order; S2: Introduce the ability of order combinations to maintain batch structure consistency during multi-process joint execution as a control factor to suppress and screen order combinations that do not have the conditions for stable multi-process execution, and form an initial order cluster structure; S3: During the construction and expansion of order clusters, the executability of the order clusters on each target process is evaluated simultaneously. When any process triggers the constraint first, the corresponding process is identified as the bottleneck process and the order cluster expansion is terminated, generating a joint batch that satisfies the multi-process collaborative constraints.

[0010] Furthermore, in step S1, the production order information includes: steel type information, specification parameter information, process path information, delivery date information, and order quantity information; the production status information includes: equipment availability status, material availability status, and production resource constraint information for processes such as steelmaking, continuous casting, and hot rolling.

[0011] Furthermore, in step S1: The multidimensional attributes of orders are quantified to construct a multidimensional feature representation of orders that reflects the batching requirements of orders under multi-process conditions. The multidimensional feature representation of orders includes steel type code features, specification parameter features, process path features, delivery window features, and order quantity features. The features of different dimensions are then normalized to form an order feature vector, which is expressed as follows: in, For the first in the production order set One order, x i For order feature vectors; Indicates the steel type characteristics of the order. Indicates specification parameters and characteristics, Indicates process path characteristics, Indicates delivery window characteristics, This indicates the order quantity characteristic.

[0012] Furthermore, in step S2, the ability of order combinations to maintain batch structure consistency during the joint execution of multiple processes is used as a dynamic inhibition factor in the joint batch similarity calculation process; The formula for calculating the joint batch similarity is as follows: in, Indicates the similarity of combined batches; Indicates order With orders Similarity in basic attributes; This represents a coefficient indicating the ability of an order to maintain batch structure consistency during the joint execution of multiple processes. A preset threshold for the ability to maintain consistency across multiple processes; When the consistency retention coefficient of the multi-process structure of an order combination meets the preset threshold condition, the corresponding order combination is allowed to participate in the construction of the order cluster; otherwise, the corresponding order combination is suppressed so that it does not participate in the joint batch calculation.

[0013] Furthermore, in the process of forming the initial order cluster structure, seed orders are used as initial cluster elements, and the clusters are formed according to the similarity of the joint batches. The candidate orders are merged in descending order of quality. The expansion process can be represented as follows: in, Indicates the first The order cluster after the next iteration (or update). Indicates the first Order clusters in the next iteration; This indicates candidate orders to be merged; Obtain the initial order cluster set It represents a group of orders that are similar in physical properties and are preliminarily determined to have the potential for joint execution across multiple processes, including steelmaking, continuous casting, and hot rolling.

[0014] The initial order cluster set is used to characterize order combinations that have the potential for multi-process joint execution under current production conditions.

[0015] Furthermore, in step S3, during the order cluster expansion process, a multi-process executability assessment is performed simultaneously; When a target process exists satisfy At that time, the process This process was identified as a bottleneck, and the order cluster was terminated. Further expansion; When order cluster All processes in the target process set P satisfy the following: At that time, determine the order cluster The conditions for forming a joint batch are met, and corresponding joint batch candidates are generated; in, Indicates order cluster In the process Available capabilities on This indicates the production requirements corresponding to the order cluster; , For the target process set.

[0016] Furthermore, the method also includes: S4: During production operations, when order or production status disturbances occur, the system adaptively selects an update strategy of maintaining, partially adjusting, or reconstructing the entire batching scheme based on the impact of the disturbance on the stability of the combined batch structure and the executability of multiple processes, thereby achieving dynamic maintenance of the combined batching scheme. S5: Output the updated joint batching scheme to the production planning system or production execution system. The joint batching scheme is used to describe the batch structure across processes and its executability characteristics in the multi-process production chain, and serves as a structured input for multi-process production organization and scheduling decisions.

[0017] Furthermore, in step S4, the disturbance event includes one or more of the following: order insertion, order cancellation, delivery date adjustment, equipment status change, and material availability status change; After detecting a disturbance event, the joint batch structural stability index is first calculated. , represented as: in, This indicates the change in the center position of the order cluster; Indicates the percentage change in the composition of a combined batch; This indicates the satisfaction status of multi-process executability constraints; Set structural stability threshold and According to the joint batch structural stability index The relationship with the preset threshold range is used to adaptively select a joint batch update strategy, which is expressed as follows: in, and This is the structural stability threshold, and ; This indicates that no structural update will be performed, and the existing joint batch structure will remain unchanged; This indicates that a partial adjustment or partial reorganization will be performed on the order cluster containing the affected orders; This indicates that a joint batch structure reconstruction has been triggered, and an overall reconstruction of the order cluster set and the joint batch structure has been performed.

[0018] This invention also provides a steel production order joint batching system based on multi-process collaboration. The system includes an order information acquisition module, an order status modeling module, an order grouping module, a joint batching generation module, a disturbance identification and dynamic update decision module, and a result output module. Order Information Acquisition Module: Responsible for collecting original production orders and real-time production status information, providing basic data input for the entire system.

[0019] Order status modeling module: Quantifies and normalizes the acquired multidimensional attributes, transforming complex production requirements into mathematically computable order feature vectors.

[0020] Order grouping module: Utilizing a joint batching similarity evaluation mechanism, and taking into account the ability to maintain consistency across multiple processes, an initial order cluster structure is formed through iterative expansion.

[0021] Joint batch generation module: Performs multi-process executability assessment, identifies bottleneck processes and limits order cluster expansion, and finally generates joint batch solutions that meet the full-process collaborative constraints.

[0022] Disturbance identification and dynamic update decision module: Real-time monitoring of anomalies in production, calculation of structural stability indicators, and adaptive triggering of local adjustment or overall reconstruction strategies.

[0023] The results output module converts the generated batch plans into structured data and outputs it to other systems to guide actual production scheduling. All modules work together to achieve the described methods.

[0024] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is used to implement the above-described method.

[0025] The beneficial effects of this invention are: (1) This invention provides a joint batching method for steel production orders in collaborative production scenarios of multiple processes such as steelmaking, continuous casting, and hot rolling. This method constructs a joint batching decision at the order level that can be executed across multiple processes and can be dynamically maintained with production disturbances. Specifically, the production order is the basic decision object. At the order level, multiple attributes such as steel type, specification parameters, process path, delivery time and production resource status are uniformly modeled, and a joint batching scheme is formed under the constraints of multi-process collaboration. This can avoid the problem of orders being independently batched in different processes and frequently splitting or reconstructing in downstream processes from the source, and improve the consistency and executability of joint batches in the multi-process production chain. (2) By introducing a similarity assessment mechanism for joint execution of multiple processes in the order grouping stage, and taking into account the ability of order combinations to maintain structural consistency in the multi-process production process, the present invention can characterize the cross-process stable execution characteristics of order combinations in advance in the batch formation stage, reduce cross-process structural conflicts, and reduce the additional production costs caused by batch adjustment and process switching. (3) By constructing a dynamic update mechanism for joint batching driven by disturbance type, the present invention can adaptively maintain the joint batching scheme under dynamic production conditions such as order insertion, order adjustment and changes in equipment or material status, and maintain the stability of batch structure and the continuity of production execution while ensuring the executability of multiple processes. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of order similarity evaluation and order grouping for multi-process joint batching decision-making in an embodiment of the present invention; Figure 2 This is a schematic diagram of joint batch dynamic update decision driven by perturbation type in an embodiment of the present invention. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention. The technical features involved in the various embodiments of the invention described below can be combined with each other as long as they do not conflict with each other.

[0028] It should be further noted that although the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than that shown or described in the flowchart.

[0029] Unless otherwise defined, the technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains; the terminology used herein is for the purpose of describing embodiments of the invention only and is not intended to limit the invention.

[0030] To address the problem that order batching in existing steel production processes is typically carried out separately in different processes such as steelmaking, continuous casting, and hot rolling, with batching decisions being independent of each other, the resulting batch structure is difficult to maintain consistency across multiple production processes. This is especially true when there are dynamic fluctuations in equipment and material status due to order insertion, changes in order attributes, delivery date adjustments, and changes in equipment and material status. Existing batching schemes often need to be frequently split or restructured in downstream processes, leading to increased waiting time across processes, decreased capacity utilization, and reduced production organization efficiency. Moreover, these problems exhibit a superimposed and amplified characteristic under multi-process constraints. This invention proposes a joint batching method and system for steel production orders based on multi-process collaboration.

[0031] Meanwhile, even when multi-process collaborative scheduling or production planning optimization methods are introduced in existing technologies, they are usually based on a predetermined single-process batch structure. Their optimization targets mainly focus on the work sequence or timing, failing to provide a unified model of the batch formation mechanism at the order level. Therefore, it is difficult to eliminate the structural inconsistencies caused by independent batching of processes at the source. Furthermore, these methods often treat the batch structure as a static result, lacking a systematic characterization of the relationship between the batch formation process and the execution conditions of multiple processes. Simultaneously, when disturbances such as order insertions, order adjustments, or changes in equipment status occur during production, existing methods generally lack an effective mechanism to dynamically maintain the batch structure in response to changes in production conditions, resulting in insufficient stability and executability of batch schemes across multi-process links.

[0032] Based on the above problems, the design objective of this invention is to: take production orders as the basic object, uniformly model the multi-dimensional attributes of orders such as steel type, specifications, process path, delivery time, and production resource status, and coordinate the grouping, joint batching generation, and subsequent maintenance processes of orders under multi-process constraints. Under the collaborative constraints of multi-process processes, a joint batching scheme that can be executed across steelmaking, continuous casting, and hot rolling processes is formed, thereby improving the consistency and executability of batch structure in multi-process production links in complex and dynamic production environments. At the same time, by introducing a dynamic update mechanism oriented towards production disturbances, the joint batching scheme can be evaluated and adjusted after generation as orders and production conditions change, ensuring the executability of multi-process processes while taking into account the stability and continuity of production organization.

[0033] In this application, "order" is used to represent a production demand unit in steel production, which includes at least information such as steel type, specifications, process path, delivery window, and order quantity; "multi-process" includes interconnected production processes such as steelmaking, continuous casting, and hot rolling; "order cluster" is used to represent an order grouping structure with similar multi-dimensional attributes and the feasibility of joint execution of multiple processes under current production conditions; "joint batch" is used to represent an order combination batch structure that can be executed across processes under the constraints of multi-process collaboration.

[0034] This invention provides a method and system for joint batching of steel production orders based on multi-process collaboration. It is designed for production organization scenarios in the steel production process where multiple processes, such as steelmaking, continuous casting, and hot rolling, are sequentially connected and strongly constrained and coupled. By uniformly modeling and representing multi-source information such as steel type, specifications, process path, delivery date, and production resource status of production orders, and making grouping and joint batching decisions for orders under multi-process collaborative constraints, a joint batching scheme that can be consistently executed in the multi-process production chain is generated, so as to achieve consistency and executability of batch structure across processes.

[0035] Unlike existing technologies that typically form batches only within a single process and then passively split or recombine them in downstream processes based on local constraints, this invention does not retroactively modify existing batches during the multi-process execution phase. Instead, it establishes the relationships between orders at the batching decision stage, addressing the collaborative execution needs of multiple processes. Furthermore, it introduces a control-based decision-making mechanism during order grouping, joint batch generation, and dynamic maintenance to reduce cross-process structural conflicts and batch reorganization frequency from the outset. The resulting joint batching results can serve as a structured input between the planning and execution layers, providing stable and scalable batch structure support for subsequent multi-process production organization and scheduling.

[0036] To achieve the above objectives, this invention proposes a method for joint batching of steel production orders based on multi-process collaboration, the overall technical concept of which is as follows: First, a set of steel production orders and related production status information are acquired. The steel production orders include at least steel type information, specification parameters, process path information, delivery date information, and order quantity information. The production status information includes at least the equipment availability status, material availability status, and production resource constraints for processes such as steelmaking, continuous casting, and hot rolling. Based on the above order information and production status information, a unified model is constructed for the orders and their execution conditions in the multi-process production chain, forming an order status description for joint batching decisions. Subsequently, based on the order status description, the multi-source attributes of the orders are quantified to construct a multi-dimensional feature representation of orders that reflects the batching requirements under multi-process collaboration conditions. Based on this multi-dimensional feature representation, an order similarity evaluation mechanism for multi-process joint execution is constructed. The ability to maintain consistency across multiple processes is introduced as a control factor during the similarity evaluation process to suppress or filter order combinations that do not possess stable multi-process execution conditions, thereby forming an initial order cluster structure with cross-process collaboration potential at the order grouping stage.

[0037] Furthermore, after the initial order cluster is formed, the feasibility of each order cluster in the target processes such as steelmaking, continuous casting and hot rolling is jointly evaluated. The feasibility evaluation includes at least process capacity constraints, process path matching relationships and equipment availability conditions. During the order cluster expansion process, bottleneck processes that have a dominant constraint effect on the formation of joint batches are identified simultaneously, so that the joint batching results are jointly limited by the tightest constraints of multiple processes at the generation stage.

[0038] Furthermore, during production operations, when disturbance events such as order insertion, order cancellation, delivery date adjustment, and changes in equipment or material status occur, the disturbance events are identified, and the impact of the disturbance on the combined batch structure is assessed through a structural stability determination mechanism. Based on the determination result, the combined batch structure is adaptively adjusted locally, reorganized, or reconstructed as a whole to maintain the stability and continuity of the combined batching scheme while ensuring the executability of multiple processes.

[0039] Finally, the generated or updated joint batching scheme is output to the production planning system or production execution system as a structured input for multi-process production organization and scheduling decisions, which guides the subsequent production planning and execution.

[0040] Example 1: This example provides a joint batching method for steel production orders based on multi-process collaboration. This method is designed for production organization scenarios involving sequential connections and strong constraints between multiple processes such as steelmaking, continuous casting, and hot rolling. Using production orders as the basic decision-making object, it generates joint batching schemes that can be executed across processes under multi-process collaboration constraints, and dynamically maintains these schemes during production operation.

[0041] In this embodiment, the steel production order set is first obtained. The steel production order includes at least steel type information, specification parameter information, process path information, delivery date information, and order quantity information; at the same time, it acquires production status information related to order execution, which includes at least the equipment availability status, material availability status, and production resource constraint information for processes such as steelmaking, continuous casting, and hot rolling.

[0042] Based on the aforementioned order information and production status information, the execution conditions of orders in the multi-process production chain are uniformly modeled to form a set of order status descriptions. This set of order status descriptions serves as the input to the joint batching algorithm, supporting subsequent order grouping, joint batch generation, and dynamic update decisions. The output of the joint batching algorithm is a joint batching scheme formed under multi-process collaborative constraints, which can serve as a structured input to a production planning system or a production execution system.

[0043] Based on this, the multi-source attributes of orders are quantified to construct a multi-dimensional feature representation of orders that reflects the batching requirements of orders under multi-process conditions. This multi-dimensional feature representation includes at least steel grade code features, specification parameter features, process path features, delivery window features, and order quantity features. Normalization is then used to eliminate the influence of different dimensional attributes on subsequent calculations, thus forming an order feature vector. The order feature vector is represented as follows: in, Indicates the first Feature vector of each order; Indicates the steel type characteristics of the order. Indicates specification parameters and characteristics, Indicates process path characteristics, Indicates delivery window characteristics, The order quantity feature is represented by the order feature vector, which serves as the basic input for subsequent order similarity assessment and order grouping calculation.

[0044] like Figure 1 As shown, after completing the order status modeling, the order grouping stage begins. In this embodiment, the order grouping process is abstracted as a clustering decision-making process with joint batch similarity as the core criterion. However, this process does not only group orders based on static attributes, but explicitly introduces the ability to maintain consistency across multiple processes as a control factor for similarity calculation.

[0045] In this embodiment, the order With orders The basic attribute similarity between them is denoted as It is used to characterize the proximity of orders in terms of static attributes such as steel type, specifications, and delivery time.

[0046] At the same time, define the ability to maintain consistency across multiple processes. It is used to characterize the ability of order combinations to maintain batch structure stability during the joint execution of multiple processes such as steelmaking, continuous casting and hot rolling.

[0047] In this embodiment, the basic attribute similarity is calculated using a weighted combination of multi-dimensional attribute similarity, and its expression is: in, The number of attribute dimensions; and These represent orders. With orders The feature value on the i-th attribute dimension; This represents the similarity function for the corresponding attribute dimension; is the attribute weight coefficient, and .

[0048] Unlike existing technologies that use multi-process factors as similarity weighting terms, in this embodiment, the ability to maintain consistency across multiple processes is not used as a regular feature in similarity weighting. Instead, it serves as a dynamic inhibition factor in the joint batch similarity calculation process, controlling whether order pairs are eligible to enter candidate clustering calculations. Specifically, the joint batch similarity is defined as: in, Indicates the similarity of combined batches; This indicates the similarity evaluation results of orders in terms of basic attributes such as steel type, specifications, and delivery time. This represents the ability coefficient of an order to maintain structural consistency during the joint execution of multiple processes; A preset threshold for maintaining consistency across multiple processes.

[0049] When order pairs are difficult to maintain a stable batch structure during the joint execution of multiple processes, even if their basic attributes are highly similar, they are suppressed from entering the subsequent clustering calculation process, thereby avoiding the formation of unstable joint batches in advance during the order grouping stage.

[0050] After completing the joint batch similarity calculation, based on The order set is grouped to form an initial order cluster set. The order cluster is used to characterize a combination of orders that have the potential for multi-process joint execution under current production conditions.

[0051] To construct the initial order cluster, this invention employs an iterative expansion method based on similarity. Specifically, seed orders are used as initial cluster elements, and candidate orders are sequentially merged into the clusters according to their joint batch similarity from high to low. The expansion process can be represented as follows: in, Indicates the first Order clusters in the next iteration; This indicates candidate orders to be merged.

[0052] After obtaining the initial set of order clusters, the feasibility of each order cluster in the multi-process production chain is jointly evaluated.

[0053] In this embodiment, the multi-process executability assessment is not only used to determine whether a combined batch is executable, but is also further introduced as a reverse constraint condition for the order cluster expansion process.

[0054] Specifically, during the order cluster expansion process, the order cluster is monitored synchronously at each target process. Capacity constraints and process path constraints. When a certain process causes the demand of an order cluster to exceed the available capacity of that process for the first time, that process is determined to be the bottleneck process of the current order cluster, and further expansion of the order cluster is terminated. The termination criterion can be expressed as: in, For the target process set, Indicates order cluster In the process Available capabilities on This indicates the production demand corresponding to the order cluster. To terminate the further expansion of the order cluster.

[0055] Through the above mechanism, the order clustering process no longer uses mathematical similarity convergence as the sole termination condition, but uses the executability of multi-process engineering as the termination criterion for cluster expansion, so that the clustering results meet the requirements of multi-process collaborative execution at the formation stage.

[0056] During production operations, when disruptive events occur such as order insertion, order cancellation, delivery date adjustment, changes in equipment status, or changes in material availability, such as... Figure 2 As shown, disturbance events are identified and their impact on the structural stability of joint batches in a multi-process production chain is assessed.

[0057] In this embodiment, based on the degree of impact of the disturbance event on the joint batch structure, the disturbance event is divided into structure-independent disturbance and structure-dependent disturbance.

[0058] Among them, structure-independent disturbances refer to disturbance events that do not cause changes in the structure of the joint batch or only cause negligible structural changes; structure-dependent disturbances refer to disturbance events that cause significant changes in the order cluster structure, joint batch composition, or multi-process executability, and the changes are quantified by subsequent structural stability indicators.

[0059] In this embodiment, after detecting a disturbance event, the joint batching algorithm first calculates the joint batch structure stability index. The structural stability index is used to quantify the degree of structural change in the joint batch before and after the disturbance, and it can be expressed as: in, This indicates the change in the center position of the order cluster. Indicates the percentage change in the composition of a combined batch. This indicates the satisfaction status of multi-process executability constraints.

[0060] Among them, the change in order cluster structure The change in the center position of the order cluster in the feature space before and after the perturbation can be used to characterize the impact. The center of the order cluster is represented by the cluster center of the order feature vector, which is used to reflect the degree of impact of the perturbation on the overall structure of the order cluster.

[0061] In this embodiment, structural stability index It is not only used to evaluate the impact of disturbances, but also serves as a branch control variable in the joint batching algorithm. According to Based on the relationship with the preset threshold range, the algorithm adaptively enters different processing branches, and its branch selection rules are as follows: in, and This is the structural stability threshold. To avoid performing a structural update, In order to perform partial adjustments, To trigger a joint batch structure reconfiguration.

[0062] Among them, the , and These correspond to the three joint batch update strategies described in the claims: maintaining the existing joint batch structure, performing local adjustments on the affected order clusters, and performing overall reconstruction on the order cluster set.

[0063] When the disturbance has a small impact on the joint batch structure, the algorithm avoids unnecessary recalculation; when the disturbance has a significant impact on the joint batch structure, the algorithm triggers structural reorganization or overall reconstruction, thereby maintaining the stability and continuity of the joint batching scheme while ensuring the executability of multiple processes.

[0064] After generating or dynamically updating the joint batching scheme, the scheme is output to the production planning system or production execution system to guide subsequent multi-process production organization and scheduling. During actual production operation, the joint batching algorithm can be repeatedly executed within a preset time period or when a disturbance event is detected, so as to achieve continuous maintenance of the joint batching scheme in the multi-process production chain.

[0065] In this embodiment, the joint batching scheme is described and output in the form of structured data, including at least: the order set identifier contained in the joint batch, the execution path identifier corresponding to each target process, and the batch quantity parameter or total production quantity parameter used to characterize the scale of the joint batch, so as to support the subsequent multi-process production planning and production scheduling decision-making.

[0066] The joint batching decision-making method provided by this invention takes production orders as the basic object. It constructs an order similarity evaluation mechanism based on multi-dimensional attributes such as steel type, specifications, process path, and delivery date, and introduces the ability to maintain consistency across multiple processes as a screening condition to suppress order combinations that lack stable joint execution conditions, forming an initial order cluster structure. Based on this, it evaluates the joint executability of multiple processes and terminates order cluster expansion by using the bottleneck process that first triggers the constraint as a limiting condition, generating a joint batching scheme that satisfies multi-process collaborative constraints. When disturbances such as order changes or fluctuations in equipment or material status occur, a dynamic update mechanism based on batch structure stability judgment is constructed, adaptively selecting strategies of maintenance, partial adjustment, or overall reconstruction to achieve dynamic maintenance of the joint batching scheme. This invention improves the consistency and executability of cross-process batches and reduces the reconstruction frequency through an algorithmic control mechanism of "suppression screening - bottleneck limitation - stability-driven update," providing dynamically maintainable batching support for multi-process collaborative production.

[0067] Example 2: The following example is one of the preferred embodiments given based on Example 1, used to illustrate the parameter configuration method and strategy selection of the method of the present invention under different production conditions, and does not constitute a limitation on the scope of protection of the claims.

[0068] In this embodiment, the threshold for maintaining consistency across multiple processes Structural stability threshold and These are pre-defined control parameters used to adjust the screening intensity of order combinations entering the joint batching calculation process and the update sensitivity of the joint batching scheme after a disturbance occurs.

[0069] Under different production conditions, the above threshold parameters can be configured according to the production load level, order delivery time pressure, or equipment resource shortage. For example, when the production load is high or the delivery time constraint is strong, the multi-process consistency maintenance capability threshold can be appropriately increased to enhance the suppression effect on order combinations that do not have stable multi-process joint execution conditions; when the production rhythm is relatively stable, the threshold can be appropriately decreased to expand the candidate range of joint batching.

[0070] Furthermore, in this embodiment, the joint batch structure stability threshold can be adjusted based on historical production statistics or real-time production status information to change the sensitivity of the joint batching algorithm in determining which update branch to enter after a disturbance occurs. When the production system has high requirements for disturbance response, the threshold range can be narrowed to improve the algorithm's response sensitivity to changes in the joint batch structure; when the production system has high requirements for production continuity and structural stability, the threshold range can be appropriately widened to reduce unnecessary joint batch reconstruction.

[0071] Furthermore, in this embodiment, the bottleneck process reverse constraint mechanism can set different judgment priorities based on the constraint strength or importance of different processes in the production chain, thereby enabling differentiated termination strategies for the order cluster expansion process under multi-process conditions. This embodiment is not limited thereto, and the above parameter configuration methods and strategy selections can be adjusted according to specific production needs.

[0072] Based on the above method embodiments, the present invention also provides a steel production order joint batching system based on multi-process collaboration, which is used to implement the joint batching method described in the above embodiments.

[0073] As mentioned above, the system includes an order information acquisition module, an order feature modeling module, an order grouping module, a joint batching generation module, a disturbance identification and dynamic update decision-making module, and a result output module. These modules work together to complete order status modeling, order grouping decision-making, joint batching scheme generation, and dynamic maintenance under disturbance conditions.

[0074] In its implementation, the system can be deployed within a production planning system or a production execution system. The relevant functional modules can be implemented through software or a combination of hardware and software. This invention does not limit the specific implementation form of the system; any system capable of achieving the functions of the aforementioned combined batching method should be included within the scope of protection of this invention.

Claims

1. A method for joint batching of steel production orders based on multi-process collaboration, characterized in that, The method includes the following steps: S1: Obtain production order information and production status information, and construct an order feature vector based on the multi-dimensional attributes of the order; S2: Introduce the ability of order combinations to maintain batch structure consistency during multi-process joint execution as a control factor to suppress and screen order combinations that do not have the conditions for stable multi-process execution, and form an initial order cluster structure; S3: During the construction and expansion of order clusters, the executability of the order clusters on each target process is evaluated simultaneously. When any process triggers the constraint first, the corresponding process is identified as the bottleneck process and the order cluster expansion is terminated, generating a joint batch that satisfies the multi-process collaborative constraints.

2. The method for joint batching of steel production orders based on multi-process collaboration as described in claim 1, characterized in that, In step S1, the production order information includes: steel type information, specification parameter information, process path information, delivery date information, and order quantity information; the production status information includes: equipment availability status, material availability status, and production resource constraint information for processes such as steelmaking, continuous casting, and hot rolling.

3. The method for joint batching of steel production orders based on multi-process collaboration as described in claim 1, characterized in that, In step S1: The multidimensional attributes of orders are quantified to construct a multidimensional feature representation of orders that reflects the batching requirements of orders under multi-process conditions; The multidimensional features of the order include steel type code features, specification parameter features, process path features, delivery window features, and order quantity features; The features of different dimensions are then normalized to form an order feature vector, which is expressed as follows: in, For the first in the production order set One order, x i For order feature vectors; Indicates the steel type characteristics of the order. Indicates specification parameters and characteristics, Indicates process path characteristics, Indicates delivery window characteristics, This indicates the order quantity characteristic.

4. The method for joint batching of steel production orders based on multi-process collaboration as described in claim 1, characterized in that, In step S2, the ability of order combinations to maintain batch structure consistency during the joint execution of multiple processes is used as a dynamic inhibition factor in the joint batch similarity calculation process; The formula for calculating the joint batch similarity is as follows: in, Indicates the similarity of combined batches; Indicates order With orders Similarity in basic attributes; This represents a coefficient indicating the ability of an order to maintain batch structure consistency during the joint execution of multiple processes. A preset threshold for the ability to maintain consistency across multiple processes; When the consistency retention coefficient of the multi-process structure of an order combination meets the preset threshold condition, the corresponding order combination is allowed to participate in the construction of the order cluster; otherwise, the corresponding order combination is suppressed so that it does not participate in the joint batch calculation.

5. The method for joint batching of steel production orders based on multi-process collaboration according to claim 4, characterized in that, In the process of forming the initial order cluster structure, seed orders are used as initial cluster elements, and the similarity of the joint batching is determined. The candidate orders are merged in descending order of quality. The expansion process can be represented as follows: in, Indicates the first Order clusters after the next iteration; Indicates the first Order clusters in the next iteration; This indicates candidate orders to be merged; Obtain the initial order cluster set .

6. The method for joint batching of steel production orders based on multi-process collaboration according to claim 5, characterized in that, Step S3: During the order cluster expansion process, perform a multi-process executability assessment simultaneously; When there is a target process satisfy At that time, the process This process was identified as a bottleneck, and the order cluster was terminated. Further expansion; When order cluster All processes in the target process set P satisfy the following: At that time, determine the order cluster The conditions for forming a joint batch are met, and corresponding joint batch candidates are generated; in, Indicates order cluster In the process Available capabilities on This indicates the production requirements corresponding to the order cluster; , For the target process set.

7. The method for joint batching of steel production orders based on multi-process collaboration as described in claim 1, characterized in that, The method further includes: S4: During production operations, when order or production status disturbances occur, the system adaptively selects an update strategy of maintaining, partially adjusting, or reconstructing the entire batching scheme based on the impact of the disturbance on the stability of the combined batch structure and the executability of multiple processes, thereby achieving dynamic maintenance of the combined batching scheme. S5: Output the updated joint batching scheme to the production planning system or production execution system. The joint batching scheme is used to describe the batch structure across processes and its executability characteristics in the multi-process production chain, and serves as a structured input for multi-process production organization and scheduling decisions.

8. The method for joint batching of steel production orders based on multi-process collaboration according to claim 7, characterized in that, In step S4, the disturbance events include one or more of the following: order insertion, order cancellation, delivery date adjustment, equipment status change, and material availability status change; After detecting a disturbance event, the joint batch structural stability index is first calculated. , represented as: in, This indicates the change in the center position of the order cluster; Indicates the percentage change in the composition of a combined batch; This indicates the satisfaction status of multi-process executability constraints; Set structural stability threshold and According to the joint batch structural stability index The relationship with the preset threshold range is used to adaptively select a joint batch update strategy, which is expressed as follows: in, and This is the structural stability threshold, and ; This indicates that no structural update will be performed, and the existing joint batch structure will remain unchanged; This indicates that a partial adjustment or partial reorganization will be performed on the order cluster containing the affected orders; This indicates that a joint batch structure reconstruction has been triggered, and an overall reconstruction of the order cluster set and the joint batch structure has been performed.

9. A joint batching system for steel production orders based on multi-process collaboration, characterized in that, The system includes an order information acquisition module, an order status modeling module, an order grouping module, a joint batch generation module, a disturbance identification and dynamic update decision module, and a result output module. The modules work together to implement the method described in any one of claims 1-8.

10. A computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the method of any one of claims 1-8.

Images