Automated Production Process Control Method and System Based on Design Model Analysis

Abstract

The present invention provides an automated production process control method and system based on design model analysis. The method includes: obtaining a building block design model corresponding to a sorting order number from a manufacturing execution system, and performing model analysis to generate a material list; allocating tasks to a vibratory bowl in the manufacturing execution system according to the materials in the material list to obtain a sorting task plan corresponding to the sorting order number; associating an empty material box in the manufacturing execution system with the sorting task plan to obtain an association relationship, and releasing the empty material box into the production line according to the association relationship; performing sorting tracking on the empty material boxes put into the production line according to the sorting task plan until the sorting task plan corresponding to the sorting order number is completed. By combining design model analysis with automated control, the present invention realizes the dynamic planning of sorting tasks and the efficient management of the production line, can flexibly adapt to diverse production requirements, and significantly improves the intelligent level and production efficiency of the automatic sorting production line.

Description

Technical Field

[0001] The present invention relates to the technical field of edge computing, and particularly to an automated production process control method and system based on design model analysis. Background Art

[0002] As an efficient automated feeding device, a vibrating bowl is widely used in various production lines. Through vibration, the vibrating bowl can arrange materials in an orderly state and eliminate materials that are stacked or in incorrect postures, thereby realizing stable feeding operations. In the mass production lines of traditional standardized products, vibrating bowls are usually used in conjunction with conveyor belts and other devices, and the discharge quantity of each material is set to a fixed value. For example, in the automatic sorting line of a standard building block set, multiple vibrating bowls are responsible for putting a preset number of building blocks into the flowing material boxes respectively to complete the sorting task of fixed orders. This control method is suitable for the production of single-batch standardized products and can effectively improve the production line efficiency.

[0003] However, with the increasing market demand for diversified and personalized products, the traditional fixed discharge method is unable to cope when faced with personalized products with multiple types and small batches or even different orders. Since the required quantity of each material is not fixed, the existing system lacks flexibility and cannot meet the changing production requirements. At the same time, the sorting task plan, material box management, and production line scheduling still rely on manual work or preset rules, with limited automation, which easily leads to material waste, reduced efficiency, and extended production cycles. Summary of the Invention

[0004] The main objective of the present invention is to solve the technical problems in the existing automatic sorting production line, such as the inability to flexibly adapt to the production requirements of multiple types of small batches or personalized products, resulting in inflexible sorting task plans, fixed control of vibrating bowls, and low production line scheduling efficiency;

[0005] The first aspect of the present invention provides an automated production process control method based on design model analysis. The automated production process control method based on design model analysis includes:

[0006] Obtain a sorting order number, obtain a building block design model corresponding to the sorting order number from the manufacturing execution system, and perform model analysis on the building block design model to obtain a material list;

[0007] Allocate tasks to the vibrating bowls in the manufacturing execution system according to the materials in the material list to obtain a sorting task plan corresponding to the sorting order number;

[0008] Associate the empty material boxes in the manufacturing execution system with the sorting task plan to obtain an association relationship, and release the empty material boxes to the production line according to the association relationship;

[0009] Sort and track the empty material boxes input into the production line according to the sorting task plan until the sorting task plan corresponding to the sorting order number is completed, and output the sorted material boxes from the production line.

[0010] Optionally, in the first implementation manner of the first aspect of the present invention, the obtaining the sorting order number, obtaining the building block design model corresponding to the sorting order number from the manufacturing execution system, and performing model analysis on the building block design model to obtain the material list includes:

[0011] Obtain the sorting order number, obtain the building block design model corresponding to the sorting order number from the manufacturing execution system, and perform three-dimensional structure decomposition on the building block design model to obtain the spatial position relationship data of the building block components in the building block design model;

[0012] Perform topological sorting processing on the building block components according to the spatial position relationship data to obtain a building block assembly order list, and perform parameter extraction processing on each building block component in the building block assembly order list to obtain a corresponding component feature set;

[0013] Query the preset standard building block library according to the component feature set to obtain the standard building block models corresponding to each building block component, and perform classification and counting processing on the standard building block models to obtain a standardized material list.

[0014] Optionally, in the second implementation manner of the first aspect of the present invention, the performing three-dimensional structure decomposition on the building block design model to obtain the spatial position relationship data of the building block components in the building block design model includes:

[0015] Perform voxelization processing on the building block design model to obtain a three-dimensional voxel matrix composed of multiple voxel units;

[0016] Perform connected component analysis processing on the three-dimensional voxel matrix to obtain multiple independent connected components, where each connected component represents a building block component;

[0017] Perform boundary extraction processing on each connected component to obtain the three-dimensional boundary point set of each building block component, and perform minimum bounding box calculation on each building block component according to the three-dimensional boundary point set to obtain the spatial position and size information of each building block component;

[0018] Perform relative coordinate conversion processing on the spatial position and size information of all building block components to obtain the relative spatial position relationship between the building block components, and use the spatial position, size information, and the relative spatial position relationship as the spatial position relationship data of the building block components in the building block design model.

[0019] Optionally, in the third implementation manner of the first aspect of the present invention, the task assignment to the vibratory bowl in the manufacturing execution system according to the materials in the material list to obtain the sorting task plan corresponding to the sorting order number includes:

[0020] Matching the materials in the material list with the preset automatically sortable material types to obtain an automatically sortable material subset;

[0021] Performing combinatorial optimization processing on the materials in the automatically sortable material subset according to the preset material constraint rules to obtain an optimal box - dividing scheme;

[0022] Performing assignment processing on the vibratory bowl according to the optimal box - dividing scheme and the status information of the vibratory bowl in the manufacturing execution system to obtain the sorting task plan corresponding to the sorting order number, where the sorting task plan includes an out - feeding sub - task list of the vibratory bowl.

[0023] Optionally, in the fourth implementation manner of the first aspect of the present invention, the performing combinatorial optimization processing on the materials in the automatically sortable material subset according to the preset material constraint rules to obtain an optimal box - dividing scheme includes:

[0024] Performing feature extraction processing on the materials in the automatically sortable material subset to obtain a material feature vector set;

[0025] Constructing a material compatibility graph according to the material feature vector set and the preset material constraint rules, and performing community detection processing on the material compatibility graph to obtain an initial material grouping;

[0026] Performing grouping optimization processing on the initial material grouping through a preset genetic optimization algorithm to obtain an optimal box - dividing scheme.

[0027] Optionally, in the fifth implementation manner of the first aspect of the present invention, the constructing a material compatibility graph according to the material feature vector set and the preset material constraint rules, and performing community detection processing on the material compatibility graph to obtain an initial material grouping includes:

[0028] Performing similarity calculation processing between materials according to the material feature vector set to obtain a material similarity matrix;

[0029] Performing fusion processing on the material similarity matrix and the preset material constraint rules to obtain a material compatibility weight matrix;

[0030] Constructing a weighted undirected graph according to the material compatibility weight matrix, and performing edge pruning processing on the weighted undirected graph to obtain a material compatibility graph;

[0031] Apply the multi-scale modularity maximization algorithm to the material compatibility graph for community detection processing to obtain a hierarchical community structure, where the communities in the hierarchical community structure represent highly compatible material groups;

[0032] Perform adaptive merging processing on the hierarchical community structure according to a preset community merging threshold to obtain an initial material grouping.

[0033] Optionally, in the sixth implementation manner of the first aspect of the present invention, the sorting and tracking of the empty material boxes input into the production line according to the sorting task plan until the sorting task plan corresponding to the sorting order number is completed, and outputting the sorted material boxes from the production line includes:

[0034] Sort and track the empty material boxes input into the production line according to the sorting task plan. When the empty material box reaches the corresponding position of the vibrating disk, check the number of the empty material box and the sorting task plan to confirm the correctness of the sorting task plan and the box order of the empty material box;

[0035] When it is confirmed that the sorting task plan and the box order of the empty material box are incorrect, analyze the missing material boxes in the sorting and tracking process and generate an alarm signal for the missing material boxes;

[0036] When it is confirmed that the sorting task plan and the box order of the empty material box are correct, control the vibrating disk to put a specified number of materials into the current empty material box according to the sorting task plan;

[0037] Control the material box to flow forward along the production line until all the vibrating disks have completed sorting, and output the sorted material boxes from the production line.

[0038] The second aspect of the present invention provides an automated production process control system based on design model analysis. The automated production process control system based on design model analysis includes:

[0039] A model analysis module, configured to obtain a sorting order number, obtain a building block design model corresponding to the sorting order number from a manufacturing execution system, and perform model analysis on the building block design model to obtain a material list;

[0040] A task allocation module, configured to allocate tasks to the vibrating disks in the manufacturing execution system according to the materials in the material list to obtain a sorting task plan corresponding to the sorting order number;

[0041] An association module, configured to associate the empty material boxes in the manufacturing execution system with the sorting task plan to obtain an association relationship, and release the empty material boxes into the production line according to the association relationship;

[0042] The sorting and tracking module is used to sort and track the empty material boxes put into the production line according to the sorting task plan until the sorting task plan corresponding to the sorting order number is completed, and output the sorted material boxes from the production line.

[0043] The above-mentioned automated production process control method and system based on design model analysis obtain the building block design model corresponding to the sorting order number from the manufacturing execution system, perform model analysis to generate a material list; allocate tasks to the vibratory bowls in the manufacturing execution system according to the materials in the material list to obtain the sorting task plan corresponding to the sorting order number; associate the empty material boxes in the manufacturing execution system with the sorting task plan to obtain an association relationship, and release the empty material boxes into the production line according to the association relationship; sort and track the empty material boxes put into the production line according to the sorting task plan until the sorting task plan corresponding to the sorting order number is completed. By combining design model analysis and automated control, the present invention realizes the dynamic planning of sorting tasks and the efficient management of the production line, can flexibly adapt to diverse production requirements, and significantly improves the intelligent level and production efficiency of the automatic sorting production line.

[0044] Other features and advantages of the present invention will be described in the following specification, and, in part, will be obvious from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention are achieved and obtained by the structures specifically pointed out in the specification, claims, and drawings.

[0045] To make the above objectives, features, and advantages of the present invention more obvious and understandable, the following specifically provides preferred embodiments and, in conjunction with the accompanying drawings, makes a detailed description as follows. Description of the Drawings

[0046] Figure 1 It is a schematic diagram of the first embodiment of the automated production process control method based on design model analysis in the embodiments of the present invention;

[0047] Figure 2 It is a schematic diagram of an embodiment of the automated production process control system based on design model analysis in the embodiments of the present invention. Detailed Embodiments

[0048] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0049] As used in the embodiments of the present invention, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally further include other unlisted steps or units, or may optionally further include other steps or units inherent to these processes, methods, products or devices.

[0050] For ease of understanding of this embodiment, a method for controlling an automated production process based on design model analysis disclosed in the embodiments of the present invention will be introduced in detail first. As Figure 1 shown, this method includes the following steps:

[0051] 101. Obtain a sorting order number, obtain a building block design model corresponding to the sorting order number from a manufacturing execution system, and perform model analysis on the building block design model to obtain a material list;

[0052] In an embodiment of the present invention, the obtaining a sorting order number, obtaining a building block design model corresponding to the sorting order number from a manufacturing execution system, and performing model analysis on the building block design model to obtain a material list includes: obtaining a sorting order number, obtaining a building block design model corresponding to the sorting order number from a manufacturing execution system, and performing three-dimensional structure decomposition on the building block design model to obtain spatial position relationship data of building block components in the building block design model; performing topological sorting processing on the building block components according to the spatial position relationship data to obtain a building block assembly order list, and performing parameter extraction processing on each building block component in the building block assembly order list to obtain a corresponding component feature set; querying a preset standard building block library according to the component feature set to obtain a standard building block model corresponding to each building block component, and performing classification and counting processing on the standard building block model to obtain a standardized material list.

[0053] Specifically, it starts with obtaining the sorting order number, which can be done by the user manually entering it or the system automatically selecting it from the order queue. After obtaining the order number, the system connects to the Manufacturing Execution System (MES) through a preset API interface and uses this order number as a query parameter to retrieve the corresponding building block design model data from the MES database. This building block design model is usually stored in the form of a 3D CAD file and contains the complete structural information of the entire building block product. After obtaining the building block design model, the system performs a 3D structural decomposition on it. This process involves complex computer graphics algorithms. First, the system converts the 3D model into a voxel representation, that is, discretizes the continuous 3D space into a series of small cubes. Then, the connected component analysis algorithm is used to identify the sets of connected voxels, and each set represents an independent building block component. For each identified component, the system calculates its central coordinates, orientation, and size information in 3D space. By comparing the relative positions and orientations between different components, the system can construct a detailed spatial position relationship data structure, which not only contains the position information of each component but also includes the connection relationships and relative directions between the components.

[0054] Specifically, the system uses the spatial position relationship data obtained in the previous step to perform a topological sorting process on the building block components. Topological sorting is a graph theory algorithm used to arrange the nodes in a directed acyclic graph into a linear sequence such that all directed edges point from the front to the back in the sequence. In this scenario, each building block component is regarded as a node in the graph, and the dependency relationships between components (for example, one component must be installed before another component) are regarded as directed edges. The system first constructs a dependency graph and then uses depth-first search or the Kahn algorithm for topological sorting to obtain a building block assembly order list that conforms to the assembly logic. This list ensures that in the actual assembly process, each component can be installed after the installation of its dependent components is completed. At the same time, the system performs parameter extraction processing on each building block component in the building block assembly order list. This process involves the analysis of the geometric features, material properties, and functional characteristics of each component. The system calculates geometric parameters such as the volume, surface area, and centroid position of the component, identifies the material type of the component (such as plastic type, color, etc.), and determines its functional characteristics (such as connecting parts, decorative parts, etc.) according to the role of the component in the overall structure. These extracted parameters constitute the feature set of each component and provide a basis for subsequent standardization matching.

[0055] Specifically, after obtaining the component feature set, the system queries the preset standard building block library and maps each custom component to the closest standard building block model. The standard building block library is a pre-established database that contains detailed information on all available standard building block parts. The matching process uses a multi-dimensional feature comparison algorithm, considering similarities in multiple aspects such as geometry, size, and function. The system calculates the similarity score with all building blocks in the standard library for each custom component and selects the standard building block with the highest score as the matching result. If there is no completely matching standard building block, the system selects the closest model and records the difference information for subsequent processing. After the matching is completed, the system classifies and counts the obtained standard building block models. The classification process is based on a predefined building block category system, such as classification by shape (bricks, plates, shafts, etc.) or function (structural parts, connectors, decorative parts, etc.). The counting process counts the number of each standard building block model used in the entire design. Finally, the system integrates all the information to generate a standardized material list. This list contains the model, quantity, classification information, and any special requirements (such as specific colors or materials) of each standard building block.

[0056] Furthermore, the three-dimensional structural decomposition of the building block design model to obtain the spatial position relationship data of the building block components in the building block design model includes: voxelizing the building block design model to obtain a three-dimensional voxel matrix composed of multiple voxel units; performing connected domain analysis on the three-dimensional voxel matrix to obtain multiple independent connected domains, wherein each connected domain represents a building block component; performing boundary extraction on each connected domain to obtain a three-dimensional boundary point set of each building block component, and performing minimum bounding box calculation on each building block component based on the three-dimensional boundary point set to obtain spatial position and size information of each building block component; performing relative coordinate transformation on the spatial position and size information of all building block components to obtain the relative spatial position relationship between the building block components, and using the spatial position, size information and the relative spatial position relationship as the spatial position relationship data of the building block components in the building block design model.

[0057] Specifically, voxelization is the process of converting a continuous three-dimensional model into a discrete voxel representation. In specific implementation, the system first determines an appropriate voxel size, which needs to balance accuracy and computational efficiency. Then, the system divides the entire design space into equal-sized cubic grids. For each point in the model, the system checks whether it falls within a voxel. If it falls within a voxel, the voxel is marked as "real", otherwise it is marked as "empty". This process is completed by traversing all vertices and faces of the model. For complex surfaces, the system also needs to perform surface sampling to ensure that the surface is correctly voxelized. Finally, the entire building block design model is converted into a three-dimensional Boolean matrix, in which each element represents a voxel, a value of 1 indicates that the voxel is occupied, and a value of 0 indicates that the voxel is empty. This voxelized representation simplifies subsequent geometric analysis, making operations such as connectivity analysis and boundary extraction more direct and efficient.

[0058] Specifically, the system performs a connected domain analysis on the obtained three-dimensional voxel matrix. The purpose of connected domain analysis is to identify and separate independent components in the model. This process uses a three-dimensional connectivity algorithm, usually a depth-first search (DFS) or a breadth-first search (BFS). The algorithm starts with an occupied voxel and checks the voxels adjacent to its six faces. If the adjacent voxel is also occupied, it is marked as the same connected domain and the search continues. This process is performed recursively or iteratively until no more adjacent occupied voxels can be found. The algorithm then moves to the next unprocessed occupied voxel and starts a new connected domain search. Each connected domain is assigned a unique identifier. In a complex building block model, there may be components connected by small connection points, and the system needs to decide whether to regard these components as independent connected domains based on predefined connection rules. Ultimately, each connected domain represents an independent building block component, and the system generates a mapping that associates each voxel with the connected domain (i.e., building block component) to which it belongs. The result of this step is a labeled three-dimensional voxel matrix, in which the value of each non-zero element represents the identifier of the building block component to which it belongs.

[0059] Specifically, the system performs boundary extraction processing on each recognized connected component. The purpose of boundary extraction is to determine the exact shape and position of each building block component. In implementation, the system traverses all voxels in each connected component. For each voxel, it checks whether the voxels adjacent to its six faces belong to the same connected component. If at least one adjacent voxel does not belong to the same connected component (i.e., is empty or belongs to another component), then the current voxel is marked as a boundary voxel. The set of coordinates of all boundary voxels constitutes the three-dimensional boundary point set of the building block component. Then, the system uses these boundary point sets to calculate the minimum bounding box of each building block component. The minimum bounding box calculation usually uses the rotating calipers algorithm or the principal component analysis (PCA) method. The system first calculates the covariance matrix of the boundary point set, and then performs eigenvalue decomposition on this matrix to obtain three principal directions. These principal directions define the orientation of the minimum bounding box. Then, the system finds the maximum and minimum coordinates of the boundary points in these three directions to determine the size and position of the bounding box. The center point coordinates of the minimum bounding box are used as the position of the building block component, and the three dimensions of the bounding box represent the size of the component.

[0060] Specifically, the system performs relative coordinate transformation processing on the spatial position and size information of all building block components. The purpose of this step is to establish the relative spatial relationship between the building block components, which is crucial for understanding the structure of the entire building block model. First, the system selects a reference point, usually the geometric center of the entire model or the center of a specific component. Then, the position of each building block component is coordinate-transformed to represent its offset relative to this reference point. This process involves simple vector subtraction operations. In addition to the position, the system also needs to consider the orientation of the component. This is usually achieved by calculating the angle between the main axis of each component and the axes of the global coordinate system. For each pair of building block components, the system calculates the vector between their center points, which represents both the relative position and implies the direction information. At the same time, the system also calculates the shortest distance between the surfaces of the component bounding boxes, which helps to determine whether the components are directly connected. All this information - including the absolute position and size of each component, the relative position relationship, the orientation difference, and the surface distance - together constitute the complete spatial position relationship data. These data are organized into a structured data set, which contains the identifier of each component, its spatial attributes, and the relationship description with all other components.

[0061] 102. Assign tasks to the vibrating bowl in the manufacturing execution system according to the materials in the material list to obtain the sorting task plan corresponding to the sorting order number;

[0062] In an embodiment of the present invention, the task assignment to the vibratory bowl in the manufacturing execution system according to the materials in the material list to obtain the sorting task plan corresponding to the sorting order number includes: matching the materials in the material list with the preset automatically sortable material types to obtain an automatically sortable material subset; performing combinatorial optimization processing on the materials in the automatically sortable material subset according to the preset material constraint rules to obtain an optimal box-packing plan; and performing assignment processing on the vibratory bowl according to the optimal box-packing plan and the status information of the vibratory bowl in the manufacturing execution system to obtain the sorting task plan corresponding to the sorting order number, where the sorting task plan includes an outfeed subtask list of the vibratory bowl.

[0063] Specifically, first, the materials in the material list are matched with the preset automatically sortable material types. The system first obtains a complete material list from the manufacturing execution system, which contains all the building block parts required for the order. At the same time, the system maintains a preset database of automatically sortable materials, which stores all the material types that can be sorted by automated equipment (such as vibratory bowls) and their characteristics. The matching process uses a multi-dimensional feature comparison algorithm, considering multiple attributes such as the geometric shape, size, and weight of the materials. The system compares each item in the material list with the entries in the automatically sortable material database one by one and calculates a similarity score. If the similarity exceeds a predetermined threshold, the material is marked as automatically sortable. In addition, the system also needs to consider the special requirements of the materials, such as fragility and surface treatment, which may affect whether the materials are suitable for automatic sorting. Through this matching process, the system filters out all the materials that can be sorted by automated equipment to form an automatically sortable material subset. This subset not only contains the list of automatically sortable materials but also includes the detailed attributes and sorting requirements of each material, providing basic data for subsequent box-packing optimization and task assignment.

[0064] Specifically, the system performs combinatorial optimization on the materials in the automatically sortable material subset according to the preset material constraint rules. The material constraint rules are a series of predefined conditions used to determine which materials can be placed in the same sorting box. These rules include material compatibility (e.g., some materials cannot be in direct contact), weight limits, volume limits, etc. The system first constructs a material compatibility matrix representing the compatibility degree between each pair of materials. Then, heuristic algorithms (such as simulated annealing or genetic algorithms) are used to generate multiple candidate binning schemes. Each candidate scheme is an allocation of materials to sorting boxes. The optimization process considers multiple objectives, including minimizing the number of sorting boxes, maximizing the space utilization rate of each sorting box, minimizing the contact probability of incompatible materials, etc. The system scores each candidate scheme, and the scoring criteria include the degree of satisfaction of all constraint rules and the overall efficiency index. Through multiple iterations and optimizations, the system finally selects the scheme with the highest score as the optimal binning scheme. This optimal scheme not only satisfies all the material constraint rules but also achieves the best balance in terms of efficiency and resource utilization. The optimal binning scheme includes a detailed material list for each sorting box and the ideal arrangement of materials in the sorting box.

[0065] Specifically, the system performs allocation processing on the vibratory bowls according to the optimal binning scheme and the status information of the vibratory bowls in the manufacturing execution system. The status information of the vibratory bowls includes the current working status, remaining capacity, compatible material types, etc. of each vibratory bowl. The system first matches each material in the optimal binning scheme with the available vibratory bowls. The matching process considers multiple factors, such as the material compatibility of the vibratory bowls, the current load, the position, etc. For each sorting box, the system generates a series of vibratory bowl discharging instructions, specifying the material type and quantity that each vibratory bowl needs to discharge. During the allocation process, the system also needs to consider the balance of the overall production line to avoid situations where some vibratory bowls are overloaded while others are idle. In addition, the system also needs to consider the moving path of the sorting boxes on the production line to ensure that the discharging order of the vibratory bowls matches the arrival order of the sorting boxes. Through this allocation process, the system generates a detailed sorting task plan, which includes a list of discharging subtasks for each vibratory bowl. Each subtask specifies when the vibratory bowl needs to discharge to which sorting box, the material type and quantity of the discharge.

[0066] Further, the combinatorial optimization of the materials in the automatically sortable material subset according to the preset material constraint rules to obtain the optimal binning scheme includes: performing feature extraction processing on the materials in the automatically sortable material subset to obtain a set of material feature vectors; constructing a material compatibility graph based on the set of material feature vectors and the preset material constraint rules, and performing community detection processing on the material compatibility graph to obtain an initial material grouping; and performing grouping optimization processing on the initial material grouping through a preset genetic optimization algorithm to obtain the optimal binning scheme.

[0067] Specifically, first, feature extraction processing is performed on the materials in the automatically sortable material subset. Feature extraction is the process of converting various attributes of materials into numerical representations, which is crucial for subsequent compatibility analysis and optimization. The system first establishes a comprehensive feature set, including geometric features of materials (such as length, width, height, volume), physical features (such as weight, material, surface roughness), functional features (such as connection method, load-bearing capacity), and other relevant attributes (such as color, production batch). For each material, the system uses specialized algorithms and sensor data to quantify these features. For example, geometric features can be obtained through 3D scanning and image processing techniques, physical features can be measured by weight sensors and material analyzers, and functional features may need to be determined by combining design specifications and experimental data. All these features are organized into a multi-dimensional vector, with each dimension corresponding to a specific attribute. To ensure comparability between different features, the system also standardizes these feature values, usually using the Z-score or Min-Max scaling method. Finally, the system obtains a set of material feature vectors, where each vector represents a material, and each element of the vector is the standardized numerical value of the material on a specific attribute. This set of material feature vectors provides a mathematical basis for subsequent compatibility analysis, making it possible to compare and cluster materials.

[0068] Specifically, the system constructs a material compatibility graph based on the material feature vector set and the preset material constraint rules, and performs community detection processing on this graph. The process of constructing the material compatibility graph first involves calculating the compatibility scores between materials. The system uses the material feature vectors and the preset material constraint rules to evaluate the compatibility degree between each pair of materials. The material constraint rules can include weight limits, volume limits, material compatibility, etc. For example, two materials with similar properties may have a higher compatibility, while materials with a large volume difference may have a lower compatibility. The system transforms these rules into mathematical functions and operates on the feature vectors of each pair of materials to obtain a compatibility score between 0 and 1. Using these scores, the system constructs a weighted undirected graph, where the nodes represent materials and the weights of the edges represent the compatibility scores between materials. Next, the system performs community detection on this compatibility graph. The purpose of community detection is to find groups of highly interconnected nodes in the graph, and these groups represent sets of materials that are highly compatible with each other. Commonly used community detection algorithms include the Louvain method, label propagation algorithm, or spectral clustering. These algorithms divide communities by optimizing modularity or other community quality metrics. The result of community detection is a series of material groups, and the materials within each group have a relatively high compatibility with each other. These groups form the initial material grouping, providing a good starting point for subsequent optimization. The initial material grouping not only considers the inherent characteristics of the materials but also reflects the preset constraint rules, so it has practical operability.

[0069] Specifically, the system performs grouping optimization on the initial material grouping through a preset genetic optimization algorithm. The genetic algorithm is an optimization method that simulates the process of natural evolution and is particularly suitable for dealing with complex combinatorial optimization problems. In this scenario, each chromosome represents a binning scheme, and the gene sequence corresponds to the distribution of materials. The initial population is generated based on the initial material grouping obtained in the previous step, which ensures that the initial solution already has a certain quality. The design of the fitness function is crucial, and it needs to comprehensively consider multiple objectives, such as minimizing the number of sorting bins, maximizing space utilization, minimizing the contact probability of incompatible materials, etc. The system also needs to design appropriate gene crossover and mutation operations to generate new candidate solutions. The crossover operation can exchange some groupings in two parent solutions, while the mutation operation may randomly change the grouping of a certain material. In each generation of evolution, the system evaluates the fitness of all candidate solutions, selects the optimal individuals to be retained in the next generation, and introduces a certain degree of randomness to maintain population diversity. This process is iterated until a preset termination condition is reached, such as reaching the maximum number of iterations or the improvement in fitness is lower than the threshold. Finally, the system selects the individual with the highest fitness from the last generation population as the optimal binning scheme. This scheme not only meets all the preset constraints but also achieves the best balance among multiple optimization objectives. The optimal binning scheme includes a detailed material list for each sorting bin and the ideal arrangement of materials in the sorting bin.

[0070] Furthermore, constructing a material compatibility graph according to the material feature vector set and preset material constraint rules, and performing community detection processing on the material compatibility graph to obtain the initial material grouping includes: calculating the similarity between materials according to the material feature vector set to obtain a material similarity matrix; fusing the material similarity matrix and preset material constraint rules to obtain a material compatibility weight matrix; constructing a weighted undirected graph according to the material compatibility weight matrix, and performing edge pruning processing on the weighted undirected graph to obtain a material compatibility graph; applying a multi-scale modularity maximization algorithm to the material compatibility graph for community detection processing to obtain a hierarchical community structure, where the communities in the hierarchical community structure represent highly compatible material groups; adaptively merging the hierarchical community structure according to a preset community merging threshold to obtain the initial material grouping.

[0071] Specifically, first, similarity calculation processing is performed on materials based on the material feature vector set. The system uses the material feature vector set obtained in the previous step to calculate the similarity for each pair of materials. Commonly used similarity calculation methods include Euclidean distance, cosine similarity, or Mahalanobis distance, etc. Selecting an appropriate similarity metric method depends on the nature and distribution of the features. For example, for high-dimensional sparse feature vectors, cosine similarity is usually more suitable. The system traverses all possible pairs of materials, calculates the similarity between them, and fills the results into an n×n matrix, where n is the total number of materials. This matrix is the material similarity matrix, and each of its elements (i,j) represents the similarity between material i and material j. The matrix is symmetric because the similarity between material i and j is the same as that between j and i. The diagonal elements are usually set to 1, indicating that the material is completely similar to itself. This similarity matrix provides a basis for subsequent compatibility analysis and reflects the proximity of materials in the feature space.

[0072] Specifically, the system performs a fusion process on the material similarity matrix and the preset material constraint rules. The purpose of this step is to combine the feature-based similarity with the actual operation constraints to obtain a more practical compatibility metric. The preset material constraint rules include weight limits, volume limits, material compatibility, etc. The system quantifies these rules into numerical weights or penalty factors. For example, if the sum of the weights of two materials exceeds the preset limit, the system will reduce their compatibility score. The fusion process usually adopts methods such as weighted summation or multiplicative combination. For each pair of materials, the system calculates a comprehensive compatibility score that takes into account both the feature similarity and the degree of satisfaction of the constraint rules. The result is a new n×n matrix called the material compatibility weight matrix. Each element of this matrix represents the comprehensive compatibility score of the corresponding pair of materials after considering all constraints. The compatibility weight matrix can better reflect the suitability of material combinations in the actual sorting process than the simple similarity matrix.

[0073] Specifically, the system constructs a weighted undirected graph based on the material compatibility weight matrix and performs edge pruning on this graph. In this graph, each node represents a material, and the weight of the edge corresponds to the score in the compatibility weight matrix. After constructing the complete graph, the system performs edge pruning, aiming to remove weak compatibility connections and highlight strong compatibility relationships. Pruning usually adopts the threshold method, that is, removing all edges with weights lower than a certain threshold. The selection of the threshold needs to balance the connectivity and sparsity of the graph, and the optimal threshold can be determined through statistical analysis or cross-validation. The graph after edge pruning is the final material compatibility graph. This graph retains the most important compatibility relationships, simplifies the subsequent community detection process, and also reduces the influence of noise. The system applies the multi-scale modularity maximization algorithm to the material compatibility graph for community detection processing. Multi-scale modularity maximization is an efficient community detection method that can identify community structures at different scales. The algorithm first calculates the modularity gain between each pair of nodes at the finest granularity, and then iteratively merges the node pairs that can maximize the modularity. This process generates a hierarchical community structure, gradually merging from the single materials at the finest granularity to larger groups. At each level, the algorithm calculates the modularity of the current partition and records the composition of the community. This hierarchical structure allows observing the clustering of materials at different granularities, providing a basis for subsequent flexible grouping.

[0074] Specifically, the system adaptively merges the hierarchical community structure according to a preset community merging threshold. The purpose of this step is to transform the hierarchical structure into practically usable material groupings. The community merging threshold defines when to stop the merging process, and it can be based on the modularity gain, community size, or other custom criteria. The system starts from the finest granularity and traverses the hierarchical structure step by step, stopping when it encounters a level that meets the threshold condition. During this process, the system also considers actual operation constraints, such as the maximum sorting box capacity, to adjust the merging decision. The finally obtained initial material groupings are a series of material groups, and the materials within each group have a high degree of compatibility.

[0075] 103. Associate the empty material boxes in the manufacturing execution system with the sorting task plan to obtain an association relationship, and release the empty material boxes to the production line according to the association relationship;

[0076] In one embodiment of the present invention, the system first obtains information about available empty cassettes from the cassette management module of the manufacturing execution system, including the unique identifier, size, status, etc. of the cassettes. At the same time, the system also obtains the sorting task plans to be executed from the task scheduling module, and these plans contain detailed information about each task, such as the type, quantity, priority, etc. of the materials to be sorted. The association process uses an intelligent matching algorithm that considers multiple factors to determine the best cassette-task pairing. The algorithm first checks whether the physical characteristics of the cassette meet the task requirements, such as whether the size is sufficient to accommodate the predetermined quantity of materials. Then, the algorithm considers the priority and estimated execution time of the task, and tries to ensure that high-priority tasks can be allocated to suitable cassettes in a timely manner. In addition, the algorithm also needs to consider the balance of the overall production line to avoid the situation where some cassettes are overused while others are idle. During the matching process, the system uses the dynamic programming method to optimize the overall matching effect and ensure the maximization of resource utilization. Once the best cassette-task pairing is determined, the system creates a corresponding association record in the database. This association record contains the unique identifier of the cassette, the task ID assigned to it, the estimated start and end times, and the bill of materials required for the task. At the same time, the system updates the status of the cassette and the task, marks the cassette as "assigned", and the task as "ready to execute". The establishment of this association relationship not only ensures that each task has a corresponding cassette, but also provides a clear execution guide for the subsequent production process. Next, based on the established association relationship, the system starts to execute the release process of the empty cassette. The release process first involves performing necessary preparation work on the cassette. The system sends instructions to the cassette preparation station, instructing the staff or automated equipment to perform operations such as cleaning, inspection, and label pasting on the specified cassette. The label usually contains the task ID, target material information, etc., which is convenient for subsequent identification and tracking. After the preparation is completed, the system updates the status of the cassette to "ready". Then, based on the current status of the production line and the task priority, the system decides the specific timing to release the cassette. This decision-making process considers multiple factors, including the number of cassettes already on the production line, the load conditions of each workstation, and the urgency of the task associated with the cassette. The system uses a real-time scheduling algorithm to optimize the release timing to ensure the continuous and efficient operation of the production line. When the conditions are met, the system sends a release instruction to the control device at the cassette entrance. The release instruction triggers the mechanical device at the cassette entrance to physically push the prepared cassette onto the conveyor belt of the production line. At the same time, the system updates the status of the cassette to "entered the line" and starts to track the movement of the cassette on the production line. The system uses a sensor network distributed throughout the production line to monitor the position and status of the cassette in real time. This real-time tracking not only ensures that the cassette moves along the predetermined path, but also provides a mechanism for timely detection and response to possible abnormal situations (such as jammed materials, misentry, etc.).

[0077] 104. Sort and track the empty bins input into the production line according to the sorting task plan until the sorting task plan corresponding to the sorting order number is completed, and output the sorted bins from the production line.

[0078] In an embodiment of the present invention, the step of sorting and tracking the empty bins input into the production line according to the sorting task plan until the sorting task plan corresponding to the sorting order number is completed, and outputting the sorted bins from the production line includes: sorting and tracking the empty bins input into the production line according to the sorting task plan. When the empty bin reaches the corresponding position of the vibrating disk, check the number of the empty bin and the sorting task plan, and confirm the correctness of the sorting task plan and the bin order of the empty bin. When it is confirmed that the sorting task plan and the bin order of the empty bin are incorrect, analyze the missing bins in the sorting and tracking process and generate an alarm signal for the missing bins. When it is confirmed that the sorting task plan and the bin order of the empty bin are correct, control the vibrating disk to input a specified quantity of materials into the current empty bin according to the sorting task plan. Control the bin to flow forward along the production line until all the vibrating disks have completed sorting, and output the sorted bins from the production line.

[0079] Specifically, first sort and track the empty bins input into the production line according to the sorting task plan. The system uses a sensor network distributed throughout the production line to monitor the position and status of the bins in real time. Each bin is equipped with a unique identification tag, such as RFID or QR code, enabling the system to accurately track the movement of each bin. When a bin enters the production line, the system starts recording its movement trajectory and compares this information with the predetermined sorting task plan in real time. This continuous tracking not only ensures that the bins move along the predetermined path but also provides a basis for the timely detection of abnormal situations. When the empty bin reaches the corresponding position of the vibrating disk, the system checks the number of the empty bin and the sorting task plan. This checking process is a key step in ensuring the accuracy of the entire sorting process. The system first confirms that the bin has reached the specified vibrating disk position through a position sensor and then reads the unique identifier of the bin. Next, the system queries the currently active sorting task plan to check whether the bin is the next bin that should reach the vibrating disk according to the plan. This checking process not only verifies the identity of the bin but also ensures that the order in which the bins arrive is consistent with the sorting plan. This dual-verification mechanism effectively prevents sorting errors that may be caused by incorrect bin order or mislabeling.

[0080] Specifically, if the system finds that the sorting task plan does not match the order of the empty boxes during the verification process, the error handling process will be triggered. First, the system will immediately suspend the operation of the current vibration plate to prevent the wrong material from being put in. Then, the system will analyze the historical data during the sorting tracking process to try to identify possible missing boxes. This analysis process involves backtracking the movement trajectory of the box to check whether any box has been accidentally removed on the way or failed to enter the production line correctly. Once the missing box is identified, the system will generate an alarm signal containing detailed information about the missing box. This alarm signal will not only be displayed on the console, but also sent to relevant personnel through preset communication channels (such as SMS and email). At the same time, the system will start a predefined exception handling process, which may include suspending the production line in the relevant area, issuing sound and light alarms, etc., to attract the attention of the operator and take necessary corrective measures.

[0081] When the sorting task plan and the order of the empty material box are confirmed to be correct, the system will control the vibration plate to put the specified amount of materials into the current empty material box according to the sorting task plan. This process first involves the precise adjustment of the vibration plate control system. The system will adjust the vibration frequency and amplitude of the vibration plate according to the current task requirements to ensure that the materials can be transported out at the appropriate speed and posture. At the same time, the system will activate the counting device matched with the vibration plate, such as a photoelectric sensor or a weight sensor, to accurately measure the amount of materials put in. When it is detected that the amount of materials put in reaches the specified value, the system will immediately stop the output of the vibration plate. The entire feeding process is under real-time monitoring. If any abnormality is found, such as material jamming or counting errors, the system will immediately interrupt the operation and issue an alarm. After completing the feeding of a vibration plate, the system will control the material box to flow forward along the production line until the sorting tasks of all vibration plates are completed. During the movement of the material box, the system continues to track its position and repeats the aforementioned verification and feeding process at each vibration plate position. The system will also update the completion status of each material box in real time and record the type and quantity of materials that have been put in. When the box has completed all the scheduled sorting tasks, the system will mark it as "sorting completed". Finally, the system controls the conveyor belt to transport the sorted box to the exit of the production line. At the exit, the system will perform a final verification to confirm that the contents of the box completely match the sorting task plan. If the verification passes, the box will be output to the designated collection area; if any mismatch is found, the system will direct the box to the exception handling area for further inspection.

[0082] In this embodiment, a building block design model corresponding to a sorting order number is obtained from a manufacturing execution system, and model analysis is performed to generate a material list. According to the materials in the material list, tasks are assigned to the vibratory bowls in the manufacturing execution system to obtain a sorting task plan corresponding to the sorting order number. An association is made between the empty material boxes in the manufacturing execution system and the sorting task plan to obtain an association relationship, and the empty material boxes are released into the production line according to the association relationship. The sorting of the empty material boxes put into the production line is tracked according to the sorting task plan until the sorting task plan corresponding to the sorting order number is completed. By combining design model analysis and automated control, the present invention realizes the dynamic planning of sorting tasks and the efficient management of the production line, can flexibly adapt to diverse production requirements, and significantly improves the intelligent level and production efficiency of the automatic sorting production line.

[0083] The above describes the automated production process control method based on design model analysis in the embodiments of the present invention. Next, the automated production process control system based on design model analysis in the embodiments of the present invention will be described. Please refer to Figure 2 , an embodiment of the automated production process control system based on design model analysis in the embodiments of the present invention includes:

[0084] A model analysis module 201, configured to obtain a sorting order number, obtain a building block design model corresponding to the sorting order number from a manufacturing execution system, and perform model analysis on the building block design model to obtain a material list;

[0085] A task assignment module 202, configured to assign tasks to the vibratory bowls in the manufacturing execution system according to the materials in the material list to obtain a sorting task plan corresponding to the sorting order number;

[0086] An association module 203, configured to associate the empty material boxes in the manufacturing execution system with the sorting task plan to obtain an association relationship, and release the empty material boxes into the production line according to the association relationship;

[0087] A sorting tracking module 204, configured to track the sorting of the empty material boxes put into the production line according to the sorting task plan until the sorting task plan corresponding to the sorting order number is completed, and output the sorted material boxes from the production line.

[0088] In the embodiments of the present invention, the automated production process control system based on design model analysis runs the above-mentioned automated production process control method based on design model analysis. The automated production process control system based on design model analysis obtains the building block design model corresponding to the sorting order number from the manufacturing execution system, performs model analysis to generate a material list; assigns tasks to the vibrating trays in the manufacturing execution system according to the materials in the material list to obtain the sorting task plan corresponding to the sorting order number; associates the empty material boxes in the manufacturing execution system with the sorting task plan to obtain an association relationship, and releases the empty material boxes to the production line according to the association relationship; performs sorting tracking on the empty material boxes put into the production line according to the sorting task plan until the sorting task plan corresponding to the sorting order number is completed. The present invention combines design model analysis with automated control to achieve dynamic planning of sorting tasks and efficient management of the production line, can flexibly adapt to diverse production requirements, and significantly improves the intelligent level and production efficiency of the automatic sorting production line.

[0089] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described system or device and unit can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0090] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0091] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An automated production process control method based on design model analysis, characterized in that: The automated production process control method based on design model analysis includes: Obtaining a sorting order number, obtaining a building block design model corresponding to the sorting order number from a manufacturing execution system, and performing model analysis on the building block design model to obtain a material list; The materials in the material list are matched with the preset types of materials that can be automatically sorted to obtain a subset of materials that can be automatically sorted; the materials in the subset of materials that can be automatically sorted are subjected to feature extraction processing to obtain a material feature vector set; similarity calculation processing is performed on the materials according to the material feature vector set to obtain a material similarity matrix; the material similarity matrix and the preset material constraint rules are fused to obtain a material compatibility weight matrix; a weighted undirected graph is constructed according to the material compatibility weight matrix, and the weighted undirected graph is subjected to edge pruning processing to obtain a material compatibility graph; a multi-scale model is applied to the material compatibility graph. The blockiness maximization algorithm is used to perform community detection processing to obtain a hierarchical community structure, wherein the communities in the hierarchical community structure represent highly compatible material groups; the hierarchical community structure is adaptively merged according to a preset community merging threshold to obtain an initial material grouping; the initial material grouping is grouped and optimized by a preset genetic optimization algorithm to obtain an optimal boxing scheme; the vibration plate is allocated according to the optimal boxing scheme and the status information of the vibration plate in the manufacturing execution system to obtain a sorting task plan corresponding to the sorting order number, wherein the sorting task plan includes a list of material discharging subtasks of the vibration plate; Associating the empty material box in the manufacturing execution system with the sorting task plan to obtain an association relationship, and releasing the empty material box to the production line according to the association relationship; According to the sorting task plan, the empty material boxes put into the production line are sorted and tracked until the sorting task plan corresponding to the sorting order number is completed, and the sorted material boxes are output from the production line.

2. The automated production process control method based on design model analysis according to claim 1, characterized in that: The obtaining of the sorting order number, obtaining the building block design model corresponding to the sorting order number from the manufacturing execution system, and performing model analysis on the building block design model to obtain the material list includes: Obtaining a sorting order number, obtaining a building block design model corresponding to the sorting order number from a manufacturing execution system, and performing a three-dimensional structural decomposition on the building block design model to obtain spatial position relationship data of building block components in the building block design model; Performing topological sorting processing on the building block components according to the spatial position relationship data to obtain a building block assembly sequence list, and performing parameter extraction processing on each building block component in the building block assembly sequence list to obtain a corresponding component feature set; A preset standard building block library is queried according to the component feature set to obtain a standard building block model corresponding to each building block component, and the standard building block models are classified and counted to obtain a standardized material list.

3. The automated production process control method based on design model analysis according to claim 2 is characterized in that: The three-dimensional structural decomposition of the building block design model to obtain spatial position relationship data of building block components in the building block design model includes: voxelize the building block design model to obtain a three-dimensional voxel matrix composed of a plurality of voxel units; Performing connected domain analysis on the three-dimensional voxel matrix to obtain a plurality of independent connected domains, wherein each connected domain represents a building block component; Perform boundary extraction processing on each connected domain to obtain a three-dimensional boundary point set of each building block component, and calculate the minimum bounding box of each building block component based on the three-dimensional boundary point set to obtain the spatial position and size information of each building block component; Relative coordinate transformation is performed on the spatial position and size information of all building block components to obtain the relative spatial position relationship between the building block components, and the spatial position, size information and the relative spatial position relationship are used as the spatial position relationship data of the building block components in the building block design model.

4. The automated production process control method based on design model analysis according to claim 1, characterized in that: The step of sorting and tracking the empty material boxes put into the production line according to the sorting task plan until the sorting task plan corresponding to the sorting order number is completed, and outputting the sorted material boxes from the production line includes: According to the sorting task plan, the empty material boxes put into the production line are sorted and tracked. When the empty material boxes arrive at the corresponding position of the vibration plate, the numbers of the empty material boxes and the sorting task plan are checked to confirm the correctness of the sorting task plan and the material box sequence of the empty material boxes; When it is confirmed that the sorting task plan and the material box sequence of the empty material box are wrong, the missing material boxes in the sorting tracking process are analyzed and an alarm signal of the missing material box is generated; When it is confirmed that the sorting task plan and the material box sequence of the empty material box are correct, the vibration plate is controlled to put a specified amount of material into the current empty material box according to the sorting task plan; Control the material boxes to flow forward along the production line until all the vibration plates have been sorted, and then output the sorted material boxes from the production line.

5. An automated production process control system based on design model analysis, characterized in that: The automated production process control system based on design model analysis includes: A model analysis module, used to obtain a sorting order number, obtain a building block design model corresponding to the sorting order number from a manufacturing execution system, and perform model analysis on the building block design model to obtain a material list; The task allocation module is used to match the materials in the material list with the preset types of materials that can be automatically sorted to obtain a subset of materials that can be automatically sorted; perform feature extraction on the materials in the subset of materials that can be automatically sorted to obtain a material feature vector set; perform similarity calculation on the materials based on the material feature vector set to obtain a material similarity matrix; perform fusion processing on the material similarity matrix and the preset material constraint rules to obtain a material compatibility weight matrix; construct a weighted undirected graph based on the material compatibility weight matrix, perform edge pruning on the weighted undirected graph to obtain a material compatibility graph; apply the material compatibility graph The multi-scale modularity maximization algorithm is used to perform community detection processing to obtain a hierarchical community structure, wherein the communities in the hierarchical community structure represent highly compatible material groups; the hierarchical community structure is adaptively merged according to a preset community merging threshold to obtain an initial material grouping; the initial material grouping is grouped and optimized by a preset genetic optimization algorithm to obtain an optimal boxing scheme; the vibration disk is allocated according to the optimal boxing scheme and the status information of the vibration disk in the manufacturing execution system to obtain a sorting task plan corresponding to the sorting order number, wherein the sorting task plan includes a list of material discharging subtasks of the vibration disk; An association module, used for associating the empty material box in the manufacturing execution system with the sorting task plan to obtain an association relationship, and releasing the empty material box to the production line according to the association relationship; The sorting tracking module is used to sort and track the empty material boxes put into the production line according to the sorting task plan until the sorting task plan corresponding to the sorting order number is completed, and the sorted material boxes are output from the production line.

Images