A production method and system of a multi-model merging engineering technique

Abstract

The application relates to a production method and system of a multi-model merging engineering technology, and relates to the field of engineering data merging, which comprises the following steps: collecting data verification standards and multi-model layout target files; determining preliminary data according to the data verification standards and the multi-model layout target files, and configuring data quality rules; defining a rule system according to business requirements based on the data quality rules; matching corresponding auditing algorithms according to the rule system, and determining a mapping relationship; generating a display report according to the rule system and the multi-model layout target files; obtaining an association configuration result based on the rule system and the mapping relationship; performing real-time processing based on the association configuration result and the preliminary data, screening and marking to obtain abnormal data and target data; performing data adjustment based on the target data and the abnormal data to obtain execution data; and performing layout production by using the execution data of the multi-model merging. The application has the effects of real-time adjustment of production data and cost reduction.

Description

Technical Field

[0001] This invention relates to the field of engineering data merging, and in particular to a production method and system for merging multi-model engineering technologies. Background Technology

[0002] As electronics manufacturing enters the era of "multiple varieties and small batches," PCB boards, as the support for electronic components and the carrier of electrical connections, are crucial to electronics manufacturing.

[0003] In the current precision manufacturing field, such as for electronic circuit boards and high-precision structural components, the production of high-reliability products generally adopts a single-panel production model. This model first completes the circuit design and layout of a single PCB board according to requirements, clarifying the component layout and circuit routing. Then, the substrate cutting, drilling, copper plating, circuit fabrication, solder masking, and character printing are carried out in sequence. Next, the precise placement and soldering of components are performed. Finally, appearance inspection and electrical performance testing are conducted to eliminate defects. The entire process is carried out independently on a single board. By handling each board separately in each process, cross-interference between processes is avoided, laying the foundation for the product's dimensional accuracy and electrical performance stability.

[0004] Because the single-plate production method uses only one plate, each process only processes a single plate, making it impossible to process multiple plates together in batches. Producing each process separately increases production costs, and the process data cannot be adjusted in real time. Summary of the Invention

[0005] In order to adjust production data in real time and reduce costs, this invention provides a production method and system for merging engineering technologies for multiple models.

[0006] In a first aspect, the present invention provides a production method for merging multiple models of engineering technology, employing the following technical solution: A production method for multi-model integrated engineering technology includes: Collect data verification standards and multi-model composite target files in business scenarios; Preliminary data was determined based on data verification standards and multi-model panel target files, and data quality rules were configured accordingly. Based on data quality rules, a rule system is defined according to preset business requirements; The corresponding auditing algorithm is obtained by matching according to the rule system, and the mapping relationship is determined; A presentation report is generated based on the rule system and multi-model imposition target files; Based on the rule system and mapping relationships, settings can be imported and completed through a visual interface, and associated configuration results can be obtained; Real-time processing is performed based on the associated configuration results and preliminary data to filter and mark abnormal data and target data. After adjusting the target data and abnormal data, the execution data is obtained, and the execution data of multiple models is used for panel production.

[0007] By adopting the above technical solution, a rule system is constructed and audit algorithms are mapped after collecting data verification standards and multi-model layout target files. The configuration is completed through visual import, and abnormal and target data are filtered in real time to generate execution data for layout. After adjustment, the execution data is output for layout production, realizing full-link automation from demand collection to production execution. By merging the production of multiple models into one project, production data can be adjusted in real time to reduce costs.

[0008] Optional methods for identifying outlier and target data include: The analytical objective for data accuracy in data acquisition business scenarios; The core dimensions of the scoring are determined based on the associated configuration results and multi-model panelized target files; Match the scoring criteria for each dimension based on the core dimensions; The quality score is determined based on the scoring criteria and preliminary data; The scoring comparison results are determined based on preset quality acceptance thresholds and quality scores; Based on the associated configuration results and preliminary data, determine the anomaly type classification rules and match the corresponding anomaly type labels; Based on the matching anomaly type labels and scoring comparison results, combined with the preset anomaly judgment conditions, preliminary data is filtered to obtain qualified data and abnormal data. Based on the analysis objectives, a data audit model is determined, and qualified data is substituted into the data audit model to obtain the target data.

[0009] By adopting the above technical solution, the preliminary data is scored in five dimensions using the core dimension scoring criteria. The qualified data selected by combining the anomaly type label and the audit model is used as the target data, so that the anomaly judgment is shifted from experience to quantification, reducing the risk of missed detection and ensuring the high quality of the patchwork data source.

[0010] Optional methods for determining quality scores include: Collect source and target datasets from business scenarios; Based on the source dataset and the target dataset, the target field, target set, source field, and source set are extracted respectively. Based on the source dataset and the target dataset, a mapping verification is performed to determine the number of discrepancies, the number of timely anomalies, and the number of valid anomalies; The complete number of anomalies is determined by validating the target field and the source field. The accurate number of anomalies is determined by matching the target set and the source set. A five-dimensional score is calculated based on the number of accurate outliers, the number of complete outliers, the number of differential outliers, the number of timely outliers, the number of valid outliers, and the scoring criteria. The quality score is calculated based on the five-dimensional scoring.

[0011] By adopting the above technical solution, fields and sets are extracted based on source data and target datasets, and the number of outliers (difference, timeliness, effectiveness, completeness, and accuracy) is calculated. The comprehensive five-dimensional score reflects the true quality of the data, provides an interpretable basis for subsequent rule optimization, and enhances the objectivity and persuasiveness of the score.

[0012] Optional methods for determining anomaly type classification rules include: Based on the associated configuration results and combined with the preset anomaly judgment conditions, the screening criteria for abnormal data are determined. Based on the screening criteria and preliminary data, the labeling rules are determined, and the preliminary data is processed to generate data verification results; The data verification results are compared with the screening criteria, and based on the preliminary data comparison, the violation rules corresponding to each anomaly are recorded synchronously. The rules for classifying anomaly types in the rule system are determined based on violation rules and marking rules.

[0013] By adopting the above technical solution, violation rules and marking rules are used to generate anomaly type classification rules, forming a reusable rule base, enabling rapid location and dynamic expansion of anomaly causes, shortening the subsequent troubleshooting cycle for similar anomalies and improving maintenance efficiency.

[0014] Optional methods for configuring the data audit model include: The dataset features are obtained by defining the scope of the source dataset and the target dataset based on the analysis objectives. Based on the characteristics of the dataset, determine the relationship mapping between the source dataset and the target dataset to obtain the mapping rules between fields; The partitioning configuration results are obtained by partitioning based on the characteristics of the dataset. Based on the analysis objectives and partition configuration results, the model parameters of the data audit model are determined; Determine the model's accuracy threshold based on model parameters and mapping rules; The model's runtime verification logic is determined based on model parameters and accuracy thresholds, and the data audit model is obtained by configuring the runtime verification logic.

[0015] By adopting the above technical solution, the partitioning granularity and mapping rules are configured based on the characteristics of the dataset. Combined with the accuracy threshold setting, the running verification logic is made so that the audit model fits the actual business granularity, ensuring that high computing efficiency and low false alarm rate are maintained even with large amounts of data.

[0016] Optionally, the partitioning configuration results obtained by partitioning based on dataset features include: Business partitioning requirements for collecting datasets to be processed; Based on the characteristics of the dataset and the business partitioning requirements, the appropriate target partitioning dimensions are selected; Determine the corresponding partition granularity based on the target partition dimension and the source dataset; Based on the partitioning granularity and the target dataset, formulate specific partitioning rules; The dataset features are partitioned according to the partitioning rules to obtain the partitioning configuration results.

[0017] By adopting the above technical solutions, target partition dimensions are selected according to business partition requirements and partition rules are formulated. The dataset is logically segmented, reducing the amount of data to be verified in a single instance, achieving parallel processing and elastic resource scheduling, significantly alleviating computing pressure and shortening the overall audit time.

[0018] Optional, generating a presentation report includes: Collect the rule system parameters and data entity characteristics of the rule system; Based on the characteristics of data entities, determine the hierarchical dimensions of the topology graph; Based on the hierarchical dimensions and multi-model panelized target files, determine the configuration items for the graphical representation of the topology diagram; The audit algorithm type is determined based on the matching of rule system parameters; Based on the audit algorithm type and rule system parameters, determine the visualization configuration items and monitoring interface types of the operating platform; Based on the monitoring interface type, a report is generated by combining graphical and visual configuration items.

[0019] By adopting the above technical solution, the system parameters and data entity characteristics are collected, the audit algorithm type is matched, and the graphical display and monitoring interface configuration is determined. The topology display report is automatically generated, making complex rules and results clear and concise, and improving efficiency.

[0020] Optionally, methods for determining execution data include: Backtracking results of collected abnormal data; Determine the deviation value based on target data and outlier data; Based on the deviation value and the preset anomaly judgment conditions, specific data items that meet the standards are selected from the abnormal data; Based on specific data items and backtracking results, analyze and determine the core causes of data anomalies; Based on the core cause, match the corresponding adjustment strategy; Adjustment strategies are applied to abnormal data to obtain adjusted execution data.

[0021] By adopting the above technical solutions, abnormal data is traced back and specific data items are selected by combining deviation values. After analyzing the core reasons, adjustment strategies are matched and execution data is generated, forming a data quality self-healing mechanism to prevent abnormalities from repeatedly entering the production process and continuously ensuring the stability of the panel input.

[0022] Optionally, specific methods for using merged execution data from multiple models for panel production include: Collect dimensional characteristics of multiple model panel designs; Based on the execution data, extract the core execution parameters; The arrangement rules for the panels are determined based on size characteristics and core parameters; Based on the core parameters and arrangement rules, a layout combination operation is performed on the adjusted data to generate a preliminary layout graphic. Adjust the layout rules based on the initial layout graphic and the preset single-page size; The layout is then used for panel production based on the adjusted arrangement rules.

[0023] By adopting the above technical solution, arrangement rules are formulated based on the core parameters and size characteristics of the execution data. After generating the initial layout graphics, the layout is then dynamically adjusted according to the single-page size rules for further production. This achieves adaptive optimization of the layout, reduces material waste, and improves the page utilization rate of multi-model layouts.

[0024] Secondly, this application provides a production system for merging multiple models of engineering technology, which adopts the following technical solution: A production system for integrating multiple models of engineering technology includes: The acquisition module is used to acquire data verification standards and multi-model panel target files; Memory, used to store programs that implement production methods for any multi-model merging engineering technology; The processor loads and executes programs from memory.

[0025] In summary, this application includes at least one of the following beneficial technical effects: 1. After collecting data verification standards and multi-model panelization target files, a rule system is constructed and audit algorithms are mapped. The configuration is completed through visual import. Abnormal and target data are filtered in real time and execution data is generated for panelization. After adjustment, the execution data is output for panelization production, realizing full-link automation from demand collection to production execution. By merging the production of multi-model projects, production data can be adjusted in real time to reduce costs. 2. The initial data is scored in five dimensions using the core dimension scoring criteria. The qualified data is selected by combining the anomaly type labels and the audit model as the target data, so that the anomaly judgment is shifted from experience to quantification, reducing the risk of missed detection and ensuring the high quality of the patchwork data source. 3. Based on the source and target datasets, extract fields and sets, calculate the number of outliers in terms of difference, timeliness, effectiveness, completeness, and accuracy, and comprehensively score the five dimensions to reflect the true quality of the data, providing an interpretable basis for subsequent rule optimization and enhancing the objectivity and persuasiveness of the scoring. Attached Figure Description

[0026] Figure 1 This is a flowchart of a production method for a multi-model merging engineering technology according to an embodiment of the present invention; Figure 2 This is a flowchart of a method for determining abnormal data and target data according to an embodiment of the present invention. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0028] This application discloses a production method for merging multiple models of engineering technology.

[0029] Reference Figure 1 A production method for multi-model integrated engineering technology includes the following steps: Step S100: Collect data verification standards and multi-model panel target files in the business scenario.

[0030] Data verification standards refer to the accuracy, completeness, and consistency requirements when assembling PCB files.

[0031] Multi-model layout target file refers to the final file required after integrating product data files of multiple models into the same production layout.

[0032] When editing PCB files, technicians open the production management platform, log in to their accounts, enter the relevant modules, and manually input specific standards, such as the model names to be merged and the standard dimensions of a single board, thereby obtaining data verification standards and multi-model panelization target files.

[0033] A production management platform is a digital tool built around the entire production process of an enterprise. It is pre-configured software used for order receiving, planning, production execution, quality control, inventory management, and data analysis.

[0034] Step S101: Determine preliminary data based on data verification standards and multi-model panel target files, and configure the resulting data quality rules.

[0035] Preliminary data refers to the original PCB file data extracted from the production management platform for panelization and merging, such as model size or process parameters.

[0036] Data quality rules refer to collective rule clauses used to assess whether data quality is up to standard.

[0037] According to the data verification standards, the raw data (such as board thickness and hole diameter parameters) of the PCB files are extracted from the production management platform. The raw dataset obtained after cleaning is the preliminary data.

[0038] By accessing the corresponding module in the production management platform, developers can use the platform to formulate the organization, process, and evaluation rules for quality assessment of the data quality of the file assembly according to data verification standards and multi-model assembly target files, thereby obtaining data quality rules and promptly identifying and resolving data quality issues.

[0039] Step S102: Based on data quality rules, a rule system is obtained according to the preset business requirements.

[0040] A rules system refers to a set of relevant quality rules when processing data in a PCB file.

[0041] Business requirements refer to the adjustments needed based on the client's content. These are pre-set by technical personnel according to the actual situation and will not be elaborated upon here.

[0042] The data quality rules are modified according to business needs. When the primary key uniqueness and data integrity are high, the requirement is to configure the automatic labeling rule, while the corresponding business logic requires manual configuration of the rule, thus obtaining the rule system. Primary key uniqueness means that the value of the primary key in the database table must be unique, which is used to ensure that each record in the table can be uniquely identified.

[0043] Step S103: Obtain the corresponding audit algorithm based on the rule system and determine the mapping relationship.

[0044] Auditing algorithms refer to algorithms used for quality checks on relevant data.

[0045] Mapping relationship refers to the correspondence between the rule system and the audit algorithm.

[0046] The data quality rules are based on the following matching algorithms: null value checking corresponds to a null value scanning algorithm, duplicate data checking corresponds to a deduplication comparison algorithm, and numerical range checking corresponds to a threshold judgment algorithm. Based on the monitoring algorithms in the workflow, the data acquisition layer sets up an interface status monitoring algorithm to monitor the running status of files / databases / acquisition interfaces; the data processing layer sets up a task scheduling algorithm to monitor the scheduling time and execution success rate of data processing tasks; and the result output layer sets up a report generation algorithm to periodically generate text / graphical monitoring reports according to templates. The "rule-algorithm" correspondence is visually managed in the runtime platform, allowing real-time viewing of algorithm execution results. If missed or false detections occur, the algorithm matching logic is optimized in reverse.

[0047] The method for determining the mapping relationship is common knowledge known to those skilled in the art, and will not be elaborated here.

[0048] Step S104: Generate a presentation report based on the rule system and multi-model imposition target files.

[0049] A presentation report refers to a visual report of the results of a data quality inspection.

[0050] The specific method for generating the presentation report is described in steps S700 to S705, and will not be repeated here.

[0051] Step S105: Based on the rule system and mapping relationship, import and complete the settings through the visual interface, and obtain the associated configuration results.

[0052] A visual interface refers to a graphical user interface, which is pre-configured by technical personnel on the production management platform according to the actual situation, and will not be elaborated on here.

[0053] The associated configuration result refers to the configuration result after the rule and algorithm are associated.

[0054] The association configuration settings can be imported through a visual interface, allowing users to import the rule system and mapping relationships, configure the "rule activation status and algorithm execution order", and generate association configuration results that can be directly used for subsequent data processing.

[0055] For example, after dragging and dropping SQL (Structured Query Language) data through a visual IDE (Integrated Development Environment) interface, and importing the relevant data, the configuration is performed in the order of format first and size second, thus obtaining the associated configuration results.

[0056] Step S106: Based on the associated configuration results and preliminary data, perform real-time processing to filter and mark abnormal data and target data.

[0057] Abnormal data refers to data that does not conform to data quality rules.

[0058] Target data refers to data that conforms to data quality rules.

[0059] The methods for determining abnormal data and target data are described in steps S200 to S207, and will not be repeated here.

[0060] Step S107: After adjusting the data based on the target data and abnormal data, the execution data is obtained, and the multi-model merged execution data is used for panel production.

[0061] Execution data refers to preliminary data that has been adjusted and is now usable for production.

[0062] The method for determining the execution data is described in steps S800 to S805, and will not be repeated here.

[0063] The specific methods for performing panel production are described in steps S900 to S905, and will not be repeated here.

[0064] Reference Figure 2 The methods for identifying abnormal data and target data include the following steps: Step S200: Collect the data accuracy analysis target in the business scenario.

[0065] The analysis objective refers to the specific target requirements used for data accuracy analysis.

[0066] The analysis objective is to determine the preset targets stored in the relevant modules of the production management platform during the PCB production process.

[0067] Step S201: Determine the core dimensions of the score based on the associated configuration results and the multi-model panel target file.

[0068] Core dimensions refer to the key dimensions used to score the data quality of a document.

[0069] Core Dimensions: Five levels of data quality dimensions (accuracy, completeness, variance, timeliness, and validity), because this dimension covers key indicators related to data accuracy (such as accuracy being related to size error and validity being related to format accuracy).

[0070] The rule system included in the associated configuration results (such as "null value check rule, format validation rule, and size compliance rule") corresponds to the basic dimensions of data quality—null value check corresponds to "completeness", format validation corresponds to "validity", and size compliance corresponds to "accuracy". This determines the candidate core dimensions for scoring. The multi-model composite target file is the final business goal. If the relevant content appears in the goal, the dimension is retained; otherwise, the dimension is excluded. Its data quality requirements further clarify the priority of the core dimensions, thus selecting five dimensions.

[0071] Step S202: Match the scoring criteria for each dimension based on the core dimensions.

[0072] The scoring criteria refer to the scoring rules for each dimension when scoring data quality.

[0073] Explanation of the scoring criteria: Accuracy is used to determine whether data matches a specified target value, such as the ratio of successfully validated records to the total number of records for monetary value verification; completeness is used to determine whether data is missing, including missing records, missing fields, and missing attributes; diversity is used to determine whether data records are duplicated and whether attributes are duplicated; a common metric is whether Hive table primary key values ​​are duplicated; timeliness is used to determine how quickly data reaches a specified target; validity is used to determine whether data conforms to agreed-upon rules such as type, format, and data range.

[0074] Step S203: Determine the quality score based on the scoring criteria and preliminary data.

[0075] The quality score refers to the overall score of the data quality.

[0076] First, calculate the sum of the scores of each rule participating in the evaluation under a single dimension to obtain the single dimension score of the task. Then, divide the sum of the scores of the tasks participating in the evaluation of this dimension by the total number of tasks participating in the evaluation of this dimension to obtain the single quality dimension score. The sum of the scores of multiple quality dimensions participating in the evaluation divided by the sum of the number of all participating dimension items is the quality score.

[0077] For example: first calculate the score of each rule (e.g., 95 points for the null rule and 100 points for the deduplication rule), then calculate the score of each dimension (completeness dimension = (95+100) / 2 = 97.5), and finally calculate the quality score (sum of scores of the five dimensions / 5).

[0078] Step S204: Determine the scoring comparison result based on the preset quality qualification threshold and quality score.

[0079] The quality acceptance threshold refers to the minimum standard for data quality acceptance. It is preset by technical personnel according to the actual situation and will not be elaborated here.

[0080] The rating comparison result refers to the comparison result between the quality score and the threshold.

[0081] A quality pass threshold (e.g., 80 points) is set, and the quality score is compared with the threshold. A score ≥80 is considered "pass," and <80 is considered "fail." The difference is recorded (e.g., a difference of 5 points between 75 and 80 points) for subsequent screening. The quality pass threshold is preset by technical personnel and will not be elaborated here.

[0082] Step S205: Determine the anomaly type classification rules and match the corresponding anomaly type labels based on the associated configuration results and preliminary data.

[0083] Anomaly classification rules refer to the rules for classifying abnormal data.

[0084] Anomaly type labels refer to the classification labels for abnormal data.

[0085] According to the established rules for classifying abnormal types, tags are matched, and abnormal data is automatically tagged through the data middle platform of the production management platform.

[0086] Step S206: Based on the matched anomaly type labels and scoring comparison results, and combined with the preset anomaly judgment conditions, preliminary data is filtered to obtain qualified data and abnormal data.

[0087] Anomaly detection conditions refer to the conditions for judging whether data is abnormal. These are preset by technical personnel according to the actual situation and will not be elaborated here.

[0088] Based on the scoring comparison results (passing score ≥ 80 points) and the abnormal type label (no label means passing), combined with the abnormal judgment criteria, qualified data and abnormal data are screened and determined.

[0089] Data with a score ≥ 80 and no label is considered qualified data, while data with a score < 80 or a label is considered abnormal data. Thus, qualified data and abnormal data are obtained. A score ≥ 80 and no label is considered an abnormal data condition.

[0090] Step S207: Determine the data audit model based on the analysis objective, and substitute the qualified data into the data audit model to obtain the target data.

[0091] A data audit model refers to a mathematical model used for data quality inspection.

[0092] The method for determining the data audit model is described in steps S500 to S505, and will not be repeated here.

[0093] Substitute qualified data into the model to output target data that meets the accuracy requirements.

[0094] The method for determining the quality score includes the following steps: Step S300: Collect the source dataset and target dataset in the business scenario.

[0095] The source dataset refers to the collection of original data sources, providing customers with initial file data.

[0096] The target dataset refers to the set of data that needs to meet the standards, which is the dataset of the defined production results.

[0097] The source dataset is obtained from the relevant file module in the production management platform, and the target dataset is obtained from the imported external file.

[0098] Step S301: Based on the source dataset and the target dataset, extract the target field, target set, source field, and source set respectively.

[0099] The target field refers to the field in the target dataset used to specify production compliance requirements.

[0100] Source fields refer to the fields in the source dataset used to describe the core attributes of the data.

[0101] The target set refers to the set of data in the target dataset that meets production requirements.

[0102] The source set refers to the collection of data in the source dataset that describes the core attributes of the data.

[0103] The source fields for data classification are extracted from the source dataset. The data without fields after extraction are the source set itself. Similarly, the target fields and target sets are extracted from the target dataset. The extraction methods are common knowledge known to those skilled in the art and will not be elaborated here.

[0104] Step S302: Perform mapping verification based on the source dataset and the target dataset to determine the number of discrepancies, the number of timely anomalies, and the number of valid anomalies.

[0105] The number of outliers refers to the number of differences that occur in the source data.

[0106] The number of timely anomalies refers to the number of time-sensitive anomalies that occur in the source data.

[0107] The number of valid outliers refers to the number of valid outliers that appear in the source data.

[0108] Select data from the source dataset that do not match the target dataset in the corresponding dimension, and count the selected data. The difference outlier is the number of duplicate records between the source dataset and the target dataset; the timeliness outlier is the number of records whose data submission delay exceeds 1 hour; and the validity outlier is the number of records with non-compliant format / range. This determines the difference outlier, the timeliness outlier, and the validity outlier.

[0109] Step S303: Determine the complete number of anomalies based on the target field and the source field.

[0110] The number of complete anomalies refers to the number of anomalies affecting data integrity.

[0111] The number of records with missing core fields (such as plate thickness and aperture) in the statistical source fields is the complete outlier count.

[0112] Step S304: Determine the accurate number of anomalies by matching the target set and the source set.

[0113] Accurate outliers refer to the number of outliers that affect the accuracy of the data.

[0114] Compare the source field values ​​with the target field values, count the number of mismatches, and record this count as the exact number of outliers. For example, if the source field specifies a plate thickness of 1.7mm and the target field specifies 1.5mm, Step S305: Calculate the five-dimensional score based on the number of accurate outliers, the number of complete outliers, the number of differential outliers, the number of timely outliers, the number of valid outliers, and the scoring criteria.

[0115] The five-dimensional score refers to the comprehensive score of data quality across five dimensions: accuracy, completeness, difference, timeliness, and effectiveness.

[0116] The scoring formula for difference, accuracy, completeness and validity is (1 - number of outliers / total number) * 100.

[0117] The formula for calculating timeliness is (1 - (threshold - delay time)) / threshold. If the delay time is greater than or equal to the threshold, the score is 0. The threshold is 1 hour by default.

[0118] Step S306: Calculate the quality score based on the five-dimensional score.

[0119] Calculate the scores for each of the five dimensions according to the formula in the document, and the sum is the five-dimensional score.

[0120] For example: Accuracy = (1-6 / 100)×100=94, Completeness = (1-2 / 100)×100=98, Difference = (1-5 / 100)×100=95, Timeliness = 0 (3 items over 1 hour), Validity = (1-4 / 100)×100=96; Five-dimensional score = 94+98+95+0+96=383.

[0121] The method for determining the classification rules for anomaly types includes the following steps: Step S400: Based on the associated configuration results and combined with the preset anomaly judgment conditions, determine the screening criteria for abnormal data.

[0122] Screening criteria refer to the rules for filtering out abnormal data.

[0123] The rule system (such as null value checking rules and format validation rules) contained in the associated configuration results corresponds to the data quality filtering rules (such as null value checking corresponding to integrity and format validation corresponding to validity). The mapping relationship in the associated configuration results is used to generate executable validation logic, thereby obtaining the basic rules for filtering. Based on the specific limits in the anomaly judgment conditions, the filtering criteria for abnormal data are determined. The specific methods are common knowledge known to those skilled in the art and will not be elaborated here.

[0124] Step S401: Based on the screening criteria and preliminary data, determine the labeling rules and process the preliminary data to generate data verification results.

[0125] Labeling rules refer to rules used to quickly locate and label abnormal data.

[0126] Data validation results refer to the results used to filter qualified data in order to locate abnormal problems.

[0127] Based on the established abnormal data screening criteria (such as rules on format, size, and completeness), and combined with the preliminary data extracted from the business system, the marking rules for abnormal data are first clarified (such as using specific symbols and colors to mark the abnormal type and cause); then, the preliminary data is checked and abnormal items are marked one by one according to the marking rules, and finally, a data verification result containing the normal / abnormal status, abnormal type, and violation details of all data is generated.

[0128] Step S402: Compare the data verification results with the screening criteria. Based on the preliminary data comparison, record the violation rules corresponding to each anomaly.

[0129] Violation rules refer to the rules used to determine whether data conforms to production standards.

[0130] The data verification results are compared with the screening criteria, and the violation is determined based on the comparison results. The violation rules corresponding to each anomaly are recorded simultaneously.

[0131] Step S403: Determine the anomaly type classification rules in the rule system based on the violation rules and the marking rules.

[0132] Anomaly type classification rules refer to a standard system for classifying data anomalies.

[0133] By combining violation rules and labeling rules as anomaly type classification rules, a systematic classification of anomalous data can be achieved, facilitating subsequent targeted processing.

[0134] The configuration method for the data audit model includes the following steps: Step S500: Determine the dataset features by defining the range of the source dataset and the target dataset based on the analysis objective.

[0135] Dataset features refer to the set of data features involved in production data processing and layout.

[0136] Based on the key attributes that have appeared in the analysis target, the data range of the source / target dataset is limited, data that has not appeared is excluded, and the set of data that has appeared that meets the requirements is retained. This set is the dataset feature.

[0137] Step S501: Based on the characteristics of the dataset, determine the relationship mapping between the source dataset and the target dataset to obtain the mapping rules between fields.

[0138] Mapping rules refer to the correspondence between source fields and target fields.

[0139] By establishing a correspondence between source and target fields, fields with consistent attributes are matched to ensure clear field matching logic.

[0140] For example, the source field "Customer Plate Thickness" corresponds to the target field "Standard Plate Thickness", the source field "Customer Aperture" corresponds to the target field "Standard Aperture", and the source field "Customer Format" corresponds to the target field "Standard Format (RS-274X)".

[0141] Step S502: Configure partitions based on dataset features to obtain partition configuration results.

[0142] Partition configuration results refer to the explicit configuration scheme formed after dividing production data into different partitions and configuring them.

[0143] The specific methods are described in steps S600 to S604, and will not be repeated here.

[0144] Step S503: Based on the analysis objectives and partition configuration results, determine the model parameters of the data audit model.

[0145] Model parameters refer to the key configuration values ​​used to control the model's operating logic and accuracy.

[0146] Model parameters are obtained by inputting the analysis objectives and partitioning configuration results into a preset model parameter database. The model parameter database is a database that is pre-set by technicians according to the actual situation. The model parameter database contains the linear relationship between the analysis objectives and partitioning configuration results and the model parameters. The actual parameters are pre-set by technicians according to the actual situation, which will not be elaborated here.

[0147] Step S504: Determine the accuracy threshold of the model based on the model parameters and mapping rules.

[0148] Accuracy threshold refers to the minimum standard value used to ensure that PCB production data meets process requirements and to determine whether the product accuracy meets the standards.

[0149] Based on the model parameters (such as the specific parameters set in the data audit model, such as the size error range and format standards) and mapping rules (such as "customer board thickness" matching "standard board thickness"), the minimum verification accuracy rate that the preset quantification model needs to achieve is the accuracy threshold, to ensure that the model output results meet the data quality requirements of the business scenario. The core logic of the quantification model is to transform business rules (such as "size compliance") into executable mathematical formulas or code logic, which can be implemented by those skilled in the art using conventional tools.

[0150] For example, if the model parameters limit the size error to ≤0.1mm and the mapping rules require a field matching accuracy of 100%, then the accuracy threshold can be set to "model verification accuracy ≥95%" to ensure that the model effectively filters qualified data under parameter constraints and field correspondence logic.

[0151] Step S505: Determine the model's runtime verification logic based on model parameters and accuracy thresholds, and obtain the data audit model by configuring the runtime verification logic.

[0152] The execution verification logic refers to an ordered verification process in the data audit model used to verify whether the data meets the preset standards.

[0153] The validation logic is as follows: first, format validation; then, field integrity validation; then, size accuracy validation; and finally, timeliness validation.

[0154] The partitioning configuration result based on dataset features includes the following steps: Step S600: Collect the business partition requirements of the dataset to be processed.

[0155] Business partitioning requirements refer to the requirements for partitioning data according to business logic.

[0156] The customer-defined requirements are extracted from the production management platform to obtain the business segment requirements.

[0157] Step S601: Based on the characteristics of the dataset and the business partitioning requirements, select the appropriate target partitioning dimensions.

[0158] The target partition dimension refers to the key field dimension that classifies and divides the data based on key influencing factors at the target level.

[0159] The target partition dimension is obtained by inputting the dataset features and business partition requirements into a preset target partition dimension database. The target partition dimension database is a database that is preset by technical personnel according to the actual situation. The target partition dimension database contains a linear relationship table between the dataset features and business partition requirements and the target partition dimension. The linear relationship table is preset by technical personnel according to the actual situation, and will not be described in detail here.

[0160] Step S602: Determine the corresponding partition granularity based on the target partition dimension and the source dataset.

[0161] Partition granularity refers to the fineness of data division, such as time data.

[0162] After dividing the source dataset according to the target partitioning dimension, the granularity of production batch, delivery date, and customer level is determined.

[0163] For example: Production batch granularity: 1 batch per day (e.g., batch 20240501, batch 20240502); Delivery time granularity: by day (e.g., rush orders: same day / next day, regular orders: 3-7 days); Customer level granularity: by level (e.g., Class A, Class B, Class C).

[0164] Step S603: Based on the partition granularity and the target dataset, formulate specific partitioning rules.

[0165] Partitioning rules refer to the specific logical conditions or expressions used to divide data.

[0166] By combining the partition granularity (the fineness of data division) with the target dataset (preset compliant data standards), the partition dimensions corresponding to the granularity are transformed into clear logical conditions, forming specific rules for standardized data partitioning. This ensures that the partitioned data meets the requirements of the target dataset, supporting subsequent data processing and production scheduling. The specific methods are common knowledge known to those skilled in the art and will not be elaborated here.

[0167] For example: Production batch rules: Data submission time between 2024050100:00-23:59 → 20240501 batch area; Delivery time rules: Delivery time ≤ 3 days → Urgent order area, 3 < Delivery time ≤ 7 days → Regular order area; Customer level rules: Customer level A is in the A category area, B is in the B category area, and C is in the C category area.

[0168] Step S604: Configure the dataset features to partition according to the partitioning rules to obtain the partitioning configuration results.

[0169] The partition configuration result refers to the output result used for subsequent data processing and production scheduling after the partitioning rules have completed the data division.

[0170] In the data middleware of the production management platform, partitioning configuration is executed according to rules—database partitioning: creating partition tables, file partitions, creating subfolders, and outputting partitioning configuration results for subsequent data processing and scheduling. The specific methods are common knowledge known to those skilled in the art and will not be elaborated here.

[0171] Generating a presentation report involves the following steps: Step S700: Collect the rule system parameters and data entity characteristics of the rule system.

[0172] Rule system parameters refer to the configurable parameter items that constitute the rule system.

[0173] Data entity characteristics refer to the data fields or entity attributes that participate in rule operations.

[0174] The rule system parameters and data entity features are extracted from the production management platform. The rule system parameters are the configurable values ​​of the rules in the platform; the data entity features are the PCB data fields that participate in the rule calculation.

[0175] Step S701: Determine the hierarchical dimensions of the topology graph based on the characteristics of the data entities.

[0176] Hierarchical dimensions refer to the hierarchical fields or classification dimensions used to construct the topology graph structure.

[0177] Based on the characteristics of data entities, the data flow is layered. In this embodiment, the data flow process consists of a data source layer, a data acquisition layer, a data cleaning layer, a data verification layer, and a data application layer. The data entities corresponding to each layer are the layering dimensions.

[0178] Step S702: Determine the graphical display configuration items of the topology diagram based on the hierarchical dimensions and multi-model panel target files.

[0179] The graphical display configuration items refer to the set of parameters used to control the visual display of the topology diagram.

[0180] Based on the target files of multiple layout models, and combined with the hierarchical dimension, the visual parameters such as node style, node color, connection style and display information are configured and modified to obtain the result of the graphical display configuration items.

[0181] Step S703: Determine the audit algorithm type based on the matching of rule system parameters.

[0182] Audit algorithm type refers to the category of algorithms used for data quality verification.

[0183] Based on the parameters of the rule system, the algorithm type is determined. The specific matching method is common knowledge known to those skilled in the art and will not be elaborated here.

[0184] Step S704: Based on the audit algorithm type and rule system parameters, determine the visualization configuration items and monitoring interface type of the running platform.

[0185] Visual configuration items refer to the configuration parameters displayed on the platform interface.

[0186] The monitoring interface type refers to the interface category definition used for system monitoring.

[0187] Based on the audit algorithm type and rule system parameters, the visualization configuration items and monitoring interface types are determined. The visualization configuration items include chart type, refresh rate, and display range. The monitoring interface types include API interface, database interface, and file interface. The specific matching methods are common knowledge known to those skilled in the art and will not be elaborated here.

[0188] Step S705: Based on the monitoring interface type, a display report is obtained by combining graphical and visual configuration items.

[0189] The presentation report refers to the final generated visual report that includes graphics and configuration.

[0190] It integrates real-time data, topology maps, and visualization charts obtained from monitoring interfaces to generate display reports, such as data flow topology maps, quality score trends over the past 7 days, anomaly type statistics (pie charts), and target achievement rates, which can be displayed in real time on a large data dashboard.

[0191] The method for determining execution data includes the following steps: Step S800: Collect the backtracking results of abnormal data.

[0192] The backtracking results refer to the tracking information on the source and processing path of abnormal data.

[0193] The backtracking results are obtained by tracing the "source-processing path-abnormal node" of abnormal data.

[0194] Example: Backtracking results of an abnormal file: The source is an old version file submitted by customer A. The processing path is customer upload, ETL extraction (no format conversion), verification (format abnormality). The abnormal node is that the format conversion script was not triggered in the ETL extraction stage, which is recorded in the backtracking report.

[0195] Step S801: Determine the deviation value based on the target data and the abnormal data.

[0196] The deviation value refers to the difference between the target data and the actual data.

[0197] The difference between the target data (standard value) and the abnormal data (actual value) is the deviation value.

[0198] Step S802: Based on the deviation value and the preset anomaly judgment conditions, filter out the specific data items in the abnormal data that meet the standards.

[0199] The specific data item refers to the specific field or record that was determined to be abnormal.

[0200] Based on the deviation value, specific fields / records that meet the abnormality judgment conditions are selected as specific data items. The specific method is common knowledge known to those skilled in the art and will not be elaborated here.

[0201] Example: Data ID001, plate thickness field 1.7mm (e.g., deviation 0.2mm); Data ID005, format field old version (e.g., qualitative deviation), clearly indicating the location of the anomaly.

[0202] Step S803: Based on the specific data items and backtracking results, analyze and determine the core reasons for the data anomalies.

[0203] The core reason refers to the fundamental problem that leads to data anomalies.

[0204] The core cause was analyzed, and the specific data items and backtracking results were combined to determine the core cause. The specific methods are common knowledge known to those skilled in the art and will not be elaborated here.

[0205] Step S804: Match the corresponding adjustment strategy based on the core cause.

[0206] Adjustment strategy refers to the operational plan used to correct abnormal data.

[0207] The core cause matching strategy is used to adjust subsequent steps. The specific methods are common knowledge known to those skilled in the art and will not be elaborated here.

[0208] If the customer enters an incorrect value, contact the customer service to correct the plate thickness; if the ETL process fails to convert the data, add a format conversion script; if duplicate data is entered, add a deduplication check to the system.

[0209] Step S805: Perform adjustment operations on the abnormal data using the adjustment strategy to obtain the adjusted execution data.

[0210] Execution data refers to data that can be used for production after anomaly corrections are completed.

[0211] Perform operations on abnormal data according to the adjustment strategy, such as filling in missing fields, converting formats, correcting values, and deleting duplicate records; merge the adjusted data with the target data to form the execution data, which must meet the document production data standards.

[0212] The specific method for using merged execution data from multiple models for panel production includes the following steps: Step S900: Collect the dimensional characteristics of multiple model panels.

[0213] Dimensional characteristics refer to the length, width, thickness, and other dimensional parameters of the objects to be assembled.

[0214] Open the files for each model with the software, enter the measurement mode, measure the length, width, and thickness to obtain the dimensions of Model A and the single plate, and determine the dimensional characteristics by combining them with the maximum processing size of the production equipment.

[0215] Step S901: Extract the core execution parameters based on the execution data.

[0216] Core parameters refer to the key data fields or values ​​that affect the layout result.

[0217] Key fields affecting panel layout are extracted from the execution data, such as the length / width / thickness of each model, splicing gap, single panel load-bearing capacity, and process compatibility, to form a core parameter table.

[0218] Step S902: Determine the arrangement rules of the panel based on size characteristics and core parameters.

[0219] Arrangement rules refer to the layout logic used for page layout, including the number of rows and columns, spacing, etc.

[0220] Taking multi-model PCB panelization as an example, based on dimensional characteristics (single board internal dimensions 510mm×410mm, requiring 5mm space for rulers on each side of the board; Model A dimensions 100mm×80mm) and core parameters (3mm splicing gap to avoid transportation collisions), the arrangement is simulated using panelization software: The available space in the length direction is the length of a single panel minus the side panel edges (510-2×5=500mm). The space occupied by a single Model A is its own length plus the gap (100+3=103mm). The calculation yields 500÷103≈4.3. Rounding down, 4 can be arranged, and the actual occupied length is 4×100+3×3=409mm (≤500mm, which meets the space requirements). The available space in the width direction is the width of the single panel minus the side panel edges (410-2×5=400mm). The space occupied by a single Model A is its own width plus the gap (80+3=83mm). The calculation yields 400÷83≈4.6. Rounding down, 4 can be arranged, and the actual occupied width is 4×80+3×3=329mm (≤400mm, which meets the space requirements). The final arrangement rules were determined as follows: Model A is arranged in four horizontal rows along the length of the single plate and four vertical rows along the width, with a 3mm gap between adjacent models and a 5mm ruler space reserved on the edge of the plate to ensure that the space utilization rate after assembly meets the standards and is within the processing range of the production equipment.

[0221] Step S903: Based on the core parameters and arrangement rules, perform a layout combination operation on the adjusted data to generate a preliminary layout graphic.

[0222] Preliminary layout graphics refer to the graphical output result after the initial layout is completed.

[0223] Import the execution data into the software, perform the layout combination according to the arrangement rules, mark the position coordinates of each model, output the preliminary layout graphic, and visually display the arrangement of each model on the single plate.

[0224] Step S904: Adjust the layout rules of the layout according to the initial layout graphic and the preset single-page size.

[0225] The single-page size refers to the standard size of a single page unit, which is preset by technicians according to the actual situation and will not be elaborated here.

[0226] Check whether the preliminary layout graphics conform to the preset single-page size, and adjust the layout rules as needed. The specific methods are common knowledge known to those skilled in the art and will not be elaborated here.

[0227] Example: If the initial layout width exceeds 400mm (e.g., 329+120+3=452mm), then the adjustment rule is: arrange 3 Model A in the width direction (3×80+2×3=246mm), add 1 Model B (120+3=123mm, total 246+123=369mm≤400), ensuring that the single-page space utilization rate after adjustment is ≥90% and within the processing range of the equipment.

[0228] Step S905: Produce the layout using the adjusted arrangement rules.

[0229] Imposition production refers to the actual imposition manufacturing process carried out according to the final rules.

[0230] The adjusted arrangement rules and panel layout are imported into the PCB production system. The subsequent steps are material cutting, cleaning, circuit printing, and testing.

[0231] For details, the substrate is cut to a single-plate size of 510×410mm, the substrate surface is cleaned, multiple models of circuits are printed according to the layout pattern, solder mask printing, drilling, AOI testing are performed, and the process is executed according to the process requirements. Finally, the production of multiple models is completed by merging and panelizing, realizing the production of multiple models at one time and saving costs.

[0232] Based on the same inventive concept, embodiments of the present invention provide a production system for merging multiple engineering technologies, comprising: The acquisition module is used to acquire data verification standards, multi-model panel target files, analysis targets, source datasets, target datasets, business partitioning requirements, rule system parameters, data entity characteristics, backtracking results, and size characteristics. Memory, used to store programs that implement production methods for any multi-model merging engineering technology; The processor loads and executes programs from memory.

[0233] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0234] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for producing printed circuit boards, characterized in that, include: Collect data verification standards and multi-model panelization target files in business scenarios. Data verification standards refer to the accuracy, completeness, and consistency requirements when panelizing PCB files. Multi-model panelization target files refer to the final files required after integrating product data files of multiple models into the same production panel. Preliminary data was determined based on data verification standards and multi-model panel target files, and data quality rules were configured accordingly. Based on data quality rules, a rule system is defined according to preset business requirements; The corresponding auditing algorithm is obtained by matching according to the rule system, and the mapping relationship is determined; A presentation report is generated based on the rule system and multi-model imposition target files; Based on the rule system and mapping relationships, settings can be imported and completed through a visual interface, and associated configuration results can be obtained; Real-time processing is performed based on the associated configuration results and preliminary data to filter and mark abnormal data and target data. After adjusting the target data and abnormal data, the execution data is obtained, and the execution data of multiple models is used for panel production. Methods for identifying anomalous and target data include: The analytical objective for data accuracy in data acquisition business scenarios; The core dimensions of the scoring are determined based on the associated configuration results and multi-model panelized target files; Match the scoring criteria for each dimension based on the core dimensions; The quality score is determined based on the scoring criteria and preliminary data; The scoring comparison results are determined based on preset quality acceptance thresholds and quality scores; Based on the associated configuration results and preliminary data, determine the anomaly type classification rules and match the corresponding anomaly type labels; Based on the matching anomaly type labels and scoring comparison results, combined with the preset anomaly judgment conditions, preliminary data is filtered to obtain qualified data and abnormal data. Determine the data audit model based on the analysis objectives, and substitute qualified data into the data audit model to obtain the target data; Methods for determining quality scores include: In the business scenario, the source dataset and the target dataset are collected. The source dataset refers to the set of original data sources, and the target dataset refers to the set of data that needs to meet the standard. Based on the source dataset and the target dataset, target fields, target sets, source fields, and source sets are extracted respectively. Target fields refer to the fields in the target dataset used to clarify the data for production compliance requirements. Source fields refer to the fields in the source dataset used to describe the core attributes of the data. Target sets refer to the data set in the target dataset that meets production requirements. Source sets refer to the data set in the source dataset that describes the core attributes of the data. Based on the source dataset and the target dataset, a mapping verification is performed to determine the number of discrepancies, the number of timely anomalies, and the number of valid anomalies. The number of discrepancies refers to the number of discrepancies in the source data, the number of timely anomalies refers to the number of timeliness anomalies in the source data, and the number of valid anomalies refers to the number of validity anomalies in the source data. The complete number of anomalies is determined by validating the target field and the source field. The complete number of anomalies refers to the number of anomalies that affect data integrity. The exact number of outliers is determined by matching the target set and the source set. The exact number of outliers refers to the number of outliers that affect the accuracy of the data. A five-dimensional score is calculated based on the number of accurate outliers, the number of complete outliers, the number of differential outliers, the number of timely outliers, the number of valid outliers, and the scoring criteria. The quality score is calculated based on the five-dimensional scoring.

2. The method for producing printed panels according to claim 1, characterized in that, Methods for determining exception type classification rules include: Based on the associated configuration results and combined with the preset anomaly judgment conditions, the screening criteria for abnormal data are determined. Based on the screening criteria and preliminary data, the labeling rules are determined, and the preliminary data is processed to generate data verification results; The data verification results are compared with the screening criteria, and based on the preliminary data comparison, the violation rules corresponding to each anomaly are recorded synchronously. The rules for classifying anomaly types in the rule system are determined based on violation rules and marking rules.

3. The method for producing printed panels according to claim 2, characterized in that, The configuration methods for the data audit model include: The dataset features are obtained by defining the scope of the source dataset and the target dataset based on the analysis objectives. Based on the characteristics of the dataset, a relational mapping is performed between the source dataset and the target dataset to obtain the mapping rules between fields; The partitioning configuration results are obtained by partitioning based on the characteristics of the dataset. Based on the analysis objectives and partition configuration results, the model parameters of the data audit model are determined; Determine the model's accuracy threshold based on model parameters and mapping rules; The model's runtime verification logic is determined based on model parameters and accuracy thresholds, and the data audit model is obtained by configuring the runtime verification logic.

4. The method for producing printed panels according to claim 3, characterized in that, The partitioning configuration results obtained based on dataset features include: Business partitioning requirements for collecting datasets to be processed; Based on the characteristics of the dataset and the business partitioning requirements, the appropriate target partitioning dimensions are selected; Determine the corresponding partition granularity based on the target partition dimension and the source dataset; Based on the partitioning granularity and the target dataset, formulate specific partitioning rules; The dataset is partitioned according to the partitioning rules to obtain the partitioning configuration results.

5. The method for producing printed panels according to claim 1, characterized in that, The generated presentation report includes: Collect the rule system parameters and data entity characteristics of the rule system; Based on the characteristics of data entities, determine the hierarchical dimensions of the topology graph; Based on the hierarchical dimensions and multi-model panelized target files, determine the configuration items for the graphical representation of the topology diagram; The audit algorithm type is determined based on the matching of rule system parameters; Based on the audit algorithm type and rule system parameters, determine the visualization configuration items and monitoring interface types of the operating platform; Based on the monitoring interface type, a report is generated by combining graphical and visual configuration items.

6. The method for producing printed panels according to claim 1, characterized in that, Methods for determining execution data include: Backtracking results of collected abnormal data; Determine the deviation value based on target data and outlier data; Based on the deviation value and the preset anomaly judgment conditions, specific data items that meet the standards are selected from the abnormal data; Based on specific data items and backtracking results, analyze and determine the core causes of data anomalies; Based on the core cause, match the corresponding adjustment strategy; Adjustment strategies are applied to abnormal data to obtain adjusted execution data.

7. The method for producing printed panels according to claim 1, characterized in that, Specific methods for using merged execution data from multiple models for panel production include: Collect dimensional characteristics of multiple model panel designs; Based on the execution data, extract the core execution parameters; The arrangement rules for the panels are determined based on size characteristics and core parameters; Based on the core parameters and arrangement rules, a layout combination operation is performed on the adjusted data to generate a preliminary layout graphic. Adjust the layout rules based on the initial layout graphic and the preset single-page size; The layout is then used for panel production based on the adjusted arrangement rules.

8. A panel production system, characterized in that, include: The acquisition module is used to acquire data verification standards and multi-model imposition target files; A memory for storing a program that implements any one of claims 1 to 7 of a panel production method; A processor is used to load and execute programs from memory.

Images