PCB Board BOM Data Parsing Method and System for SMT Production

By performing entity association and board structure identification on multi-source PCB board BOM data, the problems of inconsistent component identification and inaccurate quantity calculation were solved, enabling accurate statistics of component quantity and traceability of abnormal data in SMT production, and improving the efficiency and accuracy of production data preparation.

CN123115121APending Publication Date: 2026-11-17JIANGSU TAIHU HUIYUN DATA SYST CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202611564514.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-10
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In SMT production, existing technologies struggle to accurately identify and associate PCB board BOM data from different sources, resulting in inconsistent component identification, missing or conflicting fields, and difficulty in reflecting actual production material requirements. In particular, quantity calculations are inaccurate in panelization and array device scenarios, and the lack of a unified verification mechanism makes it difficult to locate and trace abnormal data.

Method used

By acquiring multi-source engineering files and mapping them to a unified field system, entity association is performed using device reference numbers. The number of valid sub-boards is identified by combining the board structure. Field values ​​are cross-compared to identify array devices and deconstruct them into independent unit records. A manufacturing bill of materials is generated and an anomaly report is output.

Benefits of technology

It enables accurate conversion of device quantity, reduces the risk of double counting or omission, accurately converts the internal unit quantity of array devices, generates a traceable manufacturing bill of materials, and improves the efficiency and accuracy of production data preparation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123115121A_ABST
    Figure CN123115121A_ABST
Patent Text Reader

Abstract

This invention discloses a method and system for parsing PCB board BOM data for SMT production, relating to the field of SMT intelligent manufacturing technology. The method includes the following steps: acquiring the bill of materials, board layout graphics, and mounting coordinate files; parsing and mapping them to a unified field system; forming device entities by associating multi-source records with device reference numbers, and determining layer attributes and the number of valid sub-boards based on the board layout graphics; cross-validating candidate values ​​for target fields, determining target field values, and recording anomalies; identifying array devices, generating independent unit records based on the internal unit definitions of the standard library, and calculating statistical quantities by combining the number of array devices, the quantity statistical range, the number of internal units, and the number of valid sub-boards; identifying substitute materials, test points, and dummy devices, aggregating valid records by material code, and generating a manufacturing bill of materials and anomaly reports. This invention improves the accuracy of multi-source BOM parsing and material quantity statistics.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of SMT intelligent manufacturing technology, specifically to a method and system for parsing PCB board BOM data for SMT production. Background Technology

[0002] As electronic products evolve towards higher integration, miniaturization, and flexible manufacturing with diverse product types, the sources of PCB material data used in SMT production are becoming increasingly complex, typically including BOMs, coordinate files, PCB design data, and component attribute information. Due to differences in design software, file formats, and enterprise data standards, issues such as inconsistent component identification, missing or conflicting fields, and discrepancies between material quantities and the actual board structure can easily arise. Especially in panelization, array devices, and special packaging scenarios, traditional BOM data struggles to directly and accurately reflect actual production material requirements, impacting the efficiency and accuracy of production data preparation.

[0003] In existing technologies, one type of method mainly focuses on the parsing and correction of multi-source design data in the SMT production process. This involves extracting component information from BOM files, component coordinate files, Gerber files, or EDA design files, converting data from different sources into a unified data format or intermediate data structure, and establishing a correspondence between data based on component reference numbers, coordinate information, and a component database. On this basis, data usable for production is formed through methods such as coordinate consistency detection, board simulation, or error compensation. For example, CN107133724A parses the bill of materials, component coordinates, and PCB production data separately, establishes a mapping relationship between the parsing results and components in the database, and performs coordinate and angle correction in conjunction with a virtual board surface. CN102930114B extracts component information from EDA, Gerber, and BOM files to form a unified intermediate file, and performs coordinate consistency detection between BOM data and Gerber data based on component reference numbers. Another type of method mainly addresses the problem of diverse BOM file formats and inconsistent field expressions. By identifying file types, headers, and material parameter fields, the original BOM data is cleaned, fields are converted, and formats are standardized. It is then combined with a material database to perform material matching or generate a standardized BOM. For example, CN113627893B cleans and converts electronic component data by identifying BOM file types and column header data, and then encapsulates it into unified BOM parsing data. CN105761045A generates BOM data for PCB soldering by matching the original BOM generated by the PCB design device with the material data stored in the warehouse.

[0004] However, the above solutions still have the following problems: the same device in different source files lacks a stable entity association relationship, and it is difficult to determine reliable data when there is a field conflict; the effective number of sub-boards and the range of device numbers under panelization conditions are difficult to identify accurately, which can easily lead to duplicate calculations or omissions; array devices are usually recorded as single devices and cannot be accurately converted according to their internal independent units; at the same time, there is a lack of a unified verification mechanism between device attributes and SMT manufacturing requirements, and it is difficult to locate abnormal data in a timely manner and form traceable results.

[0005] Therefore, this invention provides a method and system for parsing PCB board BOM data for SMT production. Summary of the Invention

[0006] The purpose of this invention is to provide a method and system for parsing PCB board BOM data for SMT production, so as to solve the existing problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a PCB board BOM data parsing method for SMT production, comprising the following steps: S1. Obtain and parse multi-source engineering files including bill of materials files, board graphic files and mounting coordinate files according to file type, map the component fields to a unified field system, and generate standard component records with data source identifiers; S2. Using the component reference number as the primary key, associate standard component records from different data sources as component entities; obtain the layer attributes and effective sub-boards of the component entities based on the layer information, pad information, and board edge contour information of the board surface graphic file; S3. Cross-compare the candidate field values ​​corresponding to the same target field in the device entity, determine the candidate field values ​​that match the verification as the target field values, mark the fields that do not match the verification as abnormal fields and generate abnormal records; S4. Identify array devices based on the target field value, query the device standard library to obtain the internal unit definition, generate independent unit records according to the internal unit definition, and obtain the internal unit statistical quantity based on the number of array devices in the corresponding device entity of the array device and its corresponding quantity statistical range, the number of internal units in the internal unit definition and the number of valid sub-boards. S5. Identify substitute material relationships, test points, and dummy devices for non-array device entity records and independent cell records. Separate test point records and dummy device records. Aggregate device records that participate in material quantity statistics according to material codes. Generate manufacturing bill of materials and anomaly reports.

[0008] A further improvement of this invention is that the generation process of the standard device record includes: For multi-source engineering files of different file types, select the corresponding file parsing rules; use the file parsing rules to extract the original device fields from each multi-source engineering file, and normalize at least one of the field name, data format of the field value, and coordinate format of the original device fields; map the normalized device fields to the unified field system; associate the mapped device fields with at least one of the source file identifier, source type, and original field value to obtain the standard device record; The multi-source engineering files include at least one of schematic diagram files, assembly drawing files, and engineering change files. The unified field system includes component reference number, component value, package specification, component model, coordinate position, layer, material code, and required quantity.

[0009] A further improvement of this invention is that the process for determining the layer attributes and the number of effective sub-boards of the device entity includes: Read the layer attributes of the pads corresponding to the device entity, match the layer attributes with the layer names and layer types in the board surface graphic file, and determine the top layer attribute, bottom layer attribute or inner layer attribute corresponding to the device entity. Extract the board edge elements from the board surface graphic file and form candidate closed contours based on the connection relationship between the board edge elements. Based on the size, shape and inclusion relationship of the candidate closed contours, exclude the hole and slot contours located inside the product contour and the outer process contours surrounding multiple complete product areas. Determine the number of complete product contours with corresponding size and shape and repeated arrangement as the number of effective sub-boards. The determined layer attributes are written into the corresponding device entity, and the number of valid sub-boards is associated with the device entity as a quantity conversion parameter.

[0010] A further improvement of the present invention is that the process of cross-comparing the candidate field values ​​in S3 includes: Extract candidate field values ​​from different project files within the same device entity based on the target field; After unifying the character format of the candidate field values, compare whether the candidate field values ​​are consistent. For the numerical candidate field values ​​with units of measurement, unify the units. Determine the allowable range based on the data source precision and / or engineering rules corresponding to the candidate field values, and determine whether the numerical difference of the candidate field values ​​is within the allowable range. The candidate field values ​​that meet the consistency criteria in the comparison results are determined as the corresponding target field values; Fields whose comparison results do not meet the consistency criteria are identified as abnormal fields, and the corresponding device reference number, the value of each candidate field, and the data source identifier are saved.

[0011] A further improvement of the present invention is that the identification and conversion of the array device includes: Extract at least one of the following from the device entity: package specification, device model, and device value, as a query feature for the array device; The array device query features are matched with the array device feature index pre-established in the device standard library, and the device entities with corresponding internal unit definitions are identified as array devices. Read the number of internal cells, type of internal cells, and identifier of internal cells recorded in the matched internal cell definition, generate independent cell records according to the internal cell identifiers, and associate the independent cell records with the device reference number of the original array device; The number of internal cells corresponding to the independent cell record is determined based on the number of single-board array devices, the number of effective sub-boards, and the number of internal cells of a single array device.

[0012] A further improvement of this invention is that the process of generating the manufacturing bill of materials includes: Extract material codes that conform to the preset material code format from non-array device entity records and independent cell records. For the same device record with multiple material codes, distinguish the main material code and the substitute material code according to the field position or substitute material identifier corresponding to the material code, and establish the substitute material relationship between the main material code and the substitute material code. Based on at least one of the device reference number, package specification, device value, and object marker in the mounting coordinate file in the record to be identified, the record to be identified is matched with the preset test point identification features and dummy device identification features, and the matched test point record and dummy device record are separated from the device record participating in the material statistics. The device records participating in the material statistics are grouped according to the material code, the required quantity corresponding to the same material code is accumulated, and a manufacturing bill of materials is generated based on the accumulation result and the substitute material relationship.

[0013] A further improvement of the present invention is that the device standard library stores the correspondence between device features and internal unit definitions, and the internal unit definition includes at least the number of internal units, the type of internal unit, and the internal unit identifier; The quantity statistical range of the array device corresponding to the device is determined. The quantity statistical range is determined based on the quantity range identifier in the multi-source engineering file, the configuration information exported from the engineering file, and / or the distribution of device coordinates in the corresponding area of ​​the effective sub-board. When the quantity statistics range is within the range of a single valid sub-board, the total number of array devices is determined according to the following relationship: ,in, This indicates the number of array devices within a single effective sub-board. Indicates the number of valid sub-boards; When the quantity statistics range is already the range of the entire panel, the number of array devices corresponding to the entire panel range is determined as the total number of array devices. Furthermore, the internal unit count will not be repeatedly calculated based on the number of valid sub-boards; the internal unit count will be determined according to the following relationship: ,in, This indicates the number of internal cells in a single array device as defined in the internal cell definition. This indicates the statistical quantity of the internal units; When there are array devices with different internal unit definitions, the corresponding internal unit count is determined according to the number of devices, the count range, and the number of internal units for each array device.

[0014] A further improvement of this invention is that the identification process of the test point record and the dummy device record includes: Extract at least one feature from the device reference number, package specification, device value, and object marker in the mounting coordinate file of the record to be identified; The features are matched with preset test point identification rules and fake device identification rules, wherein the test point identification rules and fake device identification rules include at least one of the following: reference number rules, package rules, device value rules and object marking rules; Based on the matching results of the features with the test point identification rules and the fake device identification rules, the record to be identified is determined to be a test point record, a fake device record, or a normal device record.

[0015] A further improvement of the present invention is that, when the multi-source engineering file includes an engineering change file with handwritten annotations, it further includes: identifying at least one graphic mark from the engineering change file, including the handwritten annotation area and the arrow, connecting line, and circled area associated with the handwritten annotation area; Candidate device entities are determined based on the pointing, connection, inclusion, and / or spatial relationships between the graphic markers and the areas where the device reference numbers are located; When the candidate device entity can be uniquely identified, the corresponding device entity is identified as the target device entity; when there are multiple candidate device entities that cannot be excluded, an engineering change association exception is generated, and the automatic execution of the corresponding engineering change is stopped. The character content in the handwritten annotation area is recognized, and the recognition result is parsed into addition, deletion or field modification operations and corresponding changed field values ​​according to the engineering change rules associated with the current engineering task. For deletion or field modification operations, the status or field of the corresponding target device entity is updated; for addition operations, a new standard device record is created based on the identified new device information. Re-execute at least one of the corresponding processing steps S2 to S5 based on the changed field to update the manufacturing bill of materials and save the source of change, device reference number, value before change, and value after change.

[0016] On the other hand, the present invention provides a PCB board BOM data parsing system for SMT production, comprising: The multi-source file parsing module is used to parse multi-source engineering files, convert device fields to a unified field system, and output standard device records with data source identifiers; The device entity association and board structure identification module is used to read the standard device record and establish device entities according to the device reference number, and output the board layer information and effective sub-board number corresponding to the device entity based on the layer information, pad information and board edge structure in the board graphic file. The field verification module is used to read the multi-source candidate field values ​​in the device entity, perform cross-source consistency verification on the multi-source candidate field values, and output the target field value and / or abnormal records; The array device deconstruction module is used to read the verified target field value and query the device standard library. Based on the matched internal cell definition, it generates an independent cell record associated with the original array device and determines the internal cell count based on the number of devices corresponding to the array device and its count range, the number of valid daughter boards and the number of internal cells. The manufacturing attribute processing and bill of materials output module is used to receive non-array device entity records and independent unit records, determine the device records participating in the material quantity statistics according to the test point identification rules and dummy device identification rules, establish the substitution material relationship between the master material code and the substitution material code, and aggregate the device records participating in the material quantity statistics according to the material code to generate a manufacturing bill of materials.

[0017] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention first performs entity association on the device records in multi-source PCB data, and determines the effective number of sub-boards and the range of device numbers by combining board structure identification. This solves the problems of device records being fragmented in different files and the easy confusion between the number of single boards and the number of the entire panel. It achieves accurate conversion of the actual number of devices put into production and reduces the risk of double counting or omission.

[0018] 2. By querying the device standard library, the number of internal units of the array device is determined, and the array device is deconstructed into independently countable internal unit records. The total number of array devices is then combined to perform quantity conversion. This solves the problem that the traditional BOM only counts single packaged devices and cannot reflect the actual demand for its internal independent units. It realizes the accurate conversion of the array device from the package level quantity to the unit level quantity required for manufacturing. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the PCB board BOM data parsing method in an embodiment of this application. Figure 2 This is a flowchart illustrating the identification of PCB board edge contour and effective sub-board quantity in an embodiment of this application. Figure 3 This is a schematic diagram illustrating the identification of the PCB board edge contour and the number of effective sub-boards in an embodiment of this application. Figure 4 This is a schematic diagram illustrating the disassembly and quantity conversion of the array devices in the embodiments of this application. Detailed Implementation

[0020] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations thereof. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0021] The term "and / or" is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can represent three cases: A exists alone, A and B exist simultaneously, and B exists alone.

[0022] This embodiment addresses the issue of different expression methods, data granularities, and potential contradictions among various engineering data generated during the design, manufacturing, and production preparation processes of the same PCB product. For example, BOM files typically focus on material codes, material names, quantities, and location numbers; mounting coordinate files mainly reflect component reference numbers, mounting coordinates, rotation angles, and mounting surfaces; and board layout graphics files such as Gerber and ODB++ reflect pads, layers, edges, and the actual physical structure of the PCB. For the same actual component, each of these different files may contain partial descriptions, but no single file inherently possesses all the correct manufacturing information.

[0023] Example 1 This embodiment provides a method for parsing PCB board BOM data for SMT production. The specific process is as follows: Figure 1 As shown, it includes the following steps: S1. Perform unified parsing of multi-source engineering files to form standard device records that can be correlated and compared: To prevent incorrect combination of data from different product versions, the system establishes the current project task before file parsing and associates the product code, PCB code, project version, and input file identifier with the same project task. The product code and version information are preferentially obtained from the BOM header, PCB project attributes, product information exported from PLM or ERP, and project file names. When the input file itself cannot uniquely provide a version, the engineer specifies it when creating the project task, and simultaneously saves the specified result and the corresponding original file, thus ensuring that all subsequent records have clear project version boundaries.

[0024] For BOM files, the system reads the actual component reference designation, component value, component model, package specification, material code, quantity, replacement material, and material status. For mount coordinate files, it reads the component reference designation, X coordinate, Y coordinate, mount surface, rotation angle, and package or component value information present in the file. For board surface graphics files such as Gerber and ODB++, it reads the layers, pads, board edge elements, drill holes, slots, and available component or package objects.

[0025] The aforementioned file can be parsed using existing Excel parsing, text parsing, Gerber parsing, ODB++ parsing, or corresponding EDA file reading methods.

[0026] Because different files may use different names for the same technical fields, the system converts the original fields to a unified field system based on the field mapping rules bound to the current engineering task. For example, RefDes in the BOM and Designator in the mounting file can both be converted to a unified "device reference number," and fields such as Part No. and Material Code are converted to a unified "material code" according to the corresponding file template. The field mapping rules are formed from the company's existing BOM templates, equipment export templates, EDA file field definitions, and historical mapping relationships confirmed by engineers, and are saved as a rule base version. Therefore, when a new file template appears, the system does not arbitrarily determine the meaning of fields based on string similarity, but retains the unrecognized original fields and forms new mapping rules after confirmation.

[0027] Device tag numbers serve as primary keys for cross-file associations in subsequent steps; therefore, differences arising solely from representation need to be eliminated before association. The system removes meaningless spaces and unifies character case according to the tag naming rules used in the current project, and parses multiple tag numbers or consecutive tag number expressions. For example, multiple tag numbers connected by explicit delimiters such as commas are split into independent tag numbers; for expressions like R1-R4, they are only expanded into R1, R2, R3, and R4 if the preceding and following prefixes are the same and the numeric parts can form a continuous integer sequence. The delimiters, valid tag number prefixes, and consecutive tag number expressions used are derived from the enterprise tag number rules or the corresponding BOM template definition. For tag numbers that cannot be uniquely parsed according to existing rules, fuzzy string algorithms are not used for forced conversion; instead, the original expression is retained and marked as data to be verified.

[0028] Besides the different field names, the coordinates of different project files are also a significant reason why direct comparison is not possible. Surface mount coordinates may use mm or mil, PCB design files may use different origins, bottom-plane exported data may contain mirror images, and scanned or modified images may use pixel coordinates. Therefore, in this embodiment, spatial data from different files is mapped to a unified PCB coordinate system before being entered into device verification.

[0029] For structured engineering files that can directly read coordinate definitions, obtain the coordinate units, origin, coordinate axis direction, and Top / Bottom attributes from the file header, export attributes, layer attributes, or corresponding device export configuration; for image data without explicit PCB coordinate definitions, select board corners, fiducials, mounting holes, positioning holes, or other stable geometric features that can be identified simultaneously in the image data and PCB engineering data to establish a correspondence.

[0030] When only rotation, scaling, and translation relationships exist between the source coordinates and the PCB standard coordinates, the affine transformation relationship can be obtained based on at least three sets of non-collinear corresponding reference points: ,in, This represents the coordinates in the source file. Represents the transformed standard PCB coordinates, matrix and vectors The homography transformation relationship is actually obtained based on the corresponding reference points mentioned above. When the source is a photograph, scan, or other data with perspective distortion, a homography transformation relationship can be established based on no less than four sets of corresponding reference points.

[0031] Whether the bottom surface needs to be mirrored and the mirroring direction should be determined primarily based on the Top / Bottom coordinate definition of the project file itself. If the file does not explicitly specify this, it should be determined by the actual corresponding features such as the board frame and Fiducial, rather than adopting the fixed assumption that all bottom surfaces are mirrored in the same direction.

[0032] After the coordinate transformation is completed, the actual residual after the transformation is calculated based on the reference point used to establish the transformation relationship. This residual is subsequently used as one of the sources of error for position consistency judgment.

[0033] After the fields and coordinates are normalized, each source data is formed into a standard device record. The standard device record not only saves the normalized device reference number, device value, package, material code, coordinates and layer information, but also saves the source file, source type, original field name and original field value.

[0034] S2. In this embodiment, the standardized device reference number is used as the primary association key to associate BOM records, mounting records and board data with the same standard device reference number within the same engineering task as a single device entity.

[0035] For example, R1 in the BOM and R1 in the mounting coordinate file are first classified into the same R1 device entity. If both sources provide package fields, then two package candidate values ​​are saved simultaneously within the device entity. If the mounting coordinate file provides the device location, and the board surface graphic data can obtain the corresponding pad location, then the two are saved as candidate space information for the device entity.

[0036] In high-density PCBs, two different devices may be physically close together. If they are merged based on the closest distance, adjacent devices may be mistaken for the same device. Therefore, this embodiment uses the tag number as the primary basis for engineering semantic association, while coordinates mainly serve as subsequent verification. Only for special records with missing tag numbers or tag numbers that cannot be uniquely determined, spatial location is used as auxiliary information for generating candidate association relationships, and it needs to be confirmed as an anomaly before it can become a formal association result.

[0037] When identical tag numbers appear within the same data source but correspond to two different locations that cannot be explained by the source's coordinate precision, the system does not directly merge them but instead generates a duplicate tag number anomaly within the source. The basis for determining whether the positional differences can be explained by the data precision comes from the actual coordinate resolution of the source file and the residual obtained from the S1 coordinate transformation.

[0038] After the device entity is established, the system further extracts the actual layer information corresponding to the device from the board file. For data with clear object and layer attributes, such as ODB++, the physical layer to which the pad belongs can be read directly; for data organized in layer files, such as Gerber, the corresponding Top or Bottom side is determined based on the layer name of the current project and the layer mapping relationship confirmed by the enterprise. The Top / Bottom fields in the mount coordinate file are retained as another independent source. At this time, the system still does not directly overwrite the mount surface with the board data, but uses both as candidate layer values ​​to be compared in subsequent S3.

[0039] Since the quantities in the BOM may not directly reflect the actual panel structure, and the subsequent conversion of array devices and manufacturing requirements depends on how many complete product boards are actually contained in a production panel, relying solely on manually entered panel counts in the BOM can lead to significant errors in quantity if versions are out of sync. Therefore, S2 also needs to determine the effective number of sub-boards from the PCB layout, the process of which is as follows: Figure 2 As shown.

[0040] The system first determines the layer used to represent the PCB board edge or mechanical outline. This layer is preferentially obtained based on the layer attributes of the project file itself, such as ODB++; for Gerber files, it is determined based on the layer naming in the manufacturing data or the board edge layer mapping rules already confirmed by the enterprise.

[0041] The system then connects line segments, arcs, and existing closed shapes in the board edge layer to form candidate closed contours. For cases where the original file does not directly save topological connections, whether two primitives are connected is determined based on the coordinate resolution of the board edge file itself. For example, if the difference between the endpoints of adjacent primitives does not exceed one effective coordinate resolution unit in the file, it can be considered a difference in the representation of the same point caused by the file's discrete precision. Therefore, the closure determination uses the precision of the current board surface file itself, rather than a fixed millimeter value applicable to all PCBs.

[0042] For the obtained closed contour, the system calculates the outer width, outer height, contour shape, spatial position, and containment relationship between contours based on the actual geometric data. Closed areas located inside a complete product shape and represented as holes or slots in the mechanical layer or drilling data are not considered independent sub-boards; when an outer contour simultaneously surrounds multiple complete product areas of the same size and shape, the outer contour is considered as a candidate for the panelization process outer contour, rather than being incorrectly calculated as another product board.

[0043] When comparing whether multiple product candidate contours belong to the same sub-board structure, the system compares them based on the normalized contour primitives, width and height dimensions, and geometric relationships. The corresponding geometric allowable difference is still formed based on the coordinate resolution of the current board edge file.

[0044] Ultimately, complete product outlines that can be identified within the same peripheral area, correspond in size and shape, and are arranged in a repeating pattern are grouped into the same product outline group. The actual number of complete product outlines in this group is determined as the number of effective sub-boards. ; When there are gaps on the board edge, or when multiple contour groups may belong to the product area or the periphery of the process and cannot be reliably ruled out, the system does not rely on experience to guess. Instead, the number of valid sub-boards is set to a pending verification status, preventing this uncertain parameter from directly entering subsequent quantity conversion.

[0045] Figure 3 This is a schematic diagram showing the identification results of the PCB board edge contour and the number of effective sub-boards in this embodiment.

[0046] S3. If, after forming the device entity in S2, fields from one type of file are directly overwritten with those from other sources, such as a fixed rule that the BOM always takes precedence, then maintenance errors in the BOM itself will unconditionally enter the production data. If a fixed rule that the mounting file takes precedence is also applied, errors will also occur when the mounting file is not synchronized with the latest version. Therefore, it is necessary to determine the consistency judgment method according to the data type and engineering meaning of the fields.

[0047] For character fields such as component values ​​and package specifications, character normalization in S1 is performed first. If an approved package alias relationship already exists in the enterprise component standard library, different aliases can be mapped to the same standard name before comparison; if no standard library reference exists, the same package is not automatically determined based on similar-looking characters.

[0048] For fields that serve as unique identifiers, such as material codes and device models, a complete comparison is performed after eliminating spaces, capitalization, and other semantically unrelated differences. For two different material codes, if the BOM or PLM data clearly indicates that they belong to a primary material and a substitute material relationship, then the substitute relationship is saved.

[0049] For numeric fields with units, first convert them to a unified unit, then determine whether the differences can be explained by data discrepancy errors based on the representation precision of the two sources. Let the smallest effective representation units of the corresponding fields from the two sources be respectively... and ,in , This is derived from the unit definitions and significant decimal places in their respective documents. In one implementation, it can be done as follows: This serves as the basis for the permissible difference formed by the precision of the two sources themselves.

[0050] For the coordinate field, the comparison is performed after the unified coordinate transformation is completed in S1. Assume the coordinate resolutions of the two sources are respectively... , The actual residuals obtained through the coordinate transformation of the reference point are respectively , In one possible implementation, it can be done according to: This forms the current coordinate comparison range between the two sources.

[0051] Calculate the distance between two candidate standard coordinates ,when If the positional difference can be explained by the original file precision and coordinate transformation error, it can be considered that the positional difference is explainable by the original file precision and coordinate transformation error; otherwise, it is recorded as a positional conflict.

[0052] Layer classification is a discrete field and does not use the numerical allowable difference mentioned above. When the board surface pad layer is determined to be Top, but the same component is recorded as Bottom in the mounting file, a layer conflict is directly formed, and two sources are retained.

[0053] The quantity in a BOM can also be independently verified using the component reference number. For example, the quantity field of a BOM record is... The number of valid position numbers is obtained after the deterministic splitting of the position number field in S1. .in Directly from the quantity field in the BOM. The result is derived from the parsing of the actual location number. When When this happens, the record can be reliably expanded into the corresponding device record; when When this happens, an anomaly occurs in the quantity and position number.

[0054] For array devices, each standard record must contain at least the query characteristics that can identify the device and the number of internal cells. Internal unit type and internal unit identifier.

[0055] When the current device query condition uniquely corresponds to a standard library definition, the system identifies the device as an array device; when there is no matching result, it does not automatically infer the number of internal units, but forms an array device to be verified record; when the same query condition corresponds to multiple standard library results with different internal structures, a standard library conflict exception is formed.

[0056] For a uniquely matching array device, the system generates multiple independent cell records based on the internal cell definitions in the standard library. For example, if RN1 corresponds to four internal cells, then RN1-U1, RN1-U2, RN1-U3, and RN1-U4 can be formed. Each independent cell record stores the physical identifier of the original array device RN1, so even if internal cells are used for material statistics later, the original physical package on the PCB can still be traced.

[0057] Let the number of array devices within a single effective sub-board be . When the BOM clearly belongs to a single-board BOM and its quantity has been verified by the location number in S3. It can be directly derived from the BOM of the single board; when the mounting coordinate file clearly covers only one valid sub-board, it can also be obtained from the actual number of components within the sub-board's range. .

[0058] Whether the quantity source falls within the single-board range is determined by not using the system default value, but prioritizing the acquisition from the BOM header, MES / ERP task information, project export instructions, or placement file export configuration. When the file does not explicitly specify, the coordinate coverage range can be checked in conjunction with the sub-board area obtained from S2. If the component coordinates fall only within one valid sub-board area, this can serve as an important basis for determining the single-board range; if the same group of components is repeatedly distributed across all sub-board areas, it indicates that the input data may have been expanded according to the entire panel. If a unique determination still cannot be made, the quantity caliber is marked as abnormal, and automatic multiplication and conversion are discontinued.

[0059] when The number of single boards has been determined, and S2 has obtained the number of valid sub-boards. At that time, the number of physical packages of array devices in the current complete panel is: ,in, Data from single-board components Count of the actual complete product outline from the PCB board surface.

[0060] Subsequently, based on the number of internal units in the standard library The number of internal units is as follows: Right now: , Data from single-board components in the current project. Based on the actual PCB panel structure, Standard structural definitions derived from the device itself.

[0061] If the input BOM or coordinate file clearly falls within the scope of a complete panel, then directly use the verified number of complete panels from that source as the basis for calculation. Do not multiply again .

[0062] S5. Process substitute materials, test points, and dummy devices at a unified manufacturing statistical granularity, and generate a manufacturing bill of materials: S5 receives two types of records: one is ordinary device entities that do not require internal disassembly, and the other is independent cell records of array devices formed by S4. Only records whose corresponding key fields have been verified by S3 or confirmed manually will enter the automated manufacturing statistics process.

[0063] The system first identifies the material code. The valid format of the material code comes from ERP, PLM, MES, or the enterprise's material coding specifications, including the code length, character composition, and classification rules currently used by the enterprise. When a field does not meet the material coding rules of the current version, the system retains its original data and generates a material code exception, rather than automatically correcting it to another existing code simply because its character format is similar.

[0064] When multiple material codes exist for the same component, the system further determines whether they belong to a primary material and a substitute material relationship. Substitution relationships can be directly derived from the substitution status field in the BOM, the substitute material master data in the PLM, or approved engineering substitution rules. The system saves the primary material code, the substitute material code, the source of the substitution relationship, and the corresponding version.

[0065] The terms "primary material" and "substitute material" indicate that the demand for the same device can be met by different materials, rather than representing that they must be consumed simultaneously during production. Therefore, in manufacturing demand statistics, the demand for the same device is only recorded once, and the substitute material is retained as a replaceable attribute of that demand, rather than having the same quantity accumulated in both the primary material and all substitute materials categories.

[0066] Before performing quantity aggregation, the system also needs to identify test points and dummy components. The corresponding identification rules come from the enterprise engineering rule library and can be formed by confirmed test point reference number rules, PCB object attributes, component value flags, package rules, mount object categories, and audited historical engineering rules, and include rule versions.

[0067] When a record to be processed matches a test point rule, its manufacturing attribute is set to test point; when it matches a dummy device rule, it is set to dummy device. Test points and dummy devices are not deleted from the engineering data, but their participation in the aggregation of actual material requirements is set to no, while the original device entity, data source, and matched rules are retained for engineers to track.

[0068] If the same record simultaneously satisfies contradictory test point rules and dummy device rules, or if two manufacturing attribute rules give classification results that cannot be met at the same time, then the record enters a manufacturing attribute abnormal state and will not participate in automatic material quantity aggregation until confirmed.

[0069] For ordinary devices whose BOM has been expanded according to position numbers and whose quantity and position number have passed S3 verification, the board quantity recorded for each device can be 1; for records with independent quantity fields that have been verified, their confirmed quantity is used; for internal cells of array devices, the corresponding cell quantity calculated in S4 is used. If the original data belongs to the board range, then it is combined with the quantity in S2. Convert to the target manufacturing range; if the original data already falls within the range of complete panels, the conversion will not be repeated.

[0070] Finally, the system groups the records that actually participated in the material statistics according to the material code.

[0071] Set material code The corresponding set of valid records is Therefore, the required quantity of this material to be manufactured is: ,in, This set includes only records with the same material code and a valid status in manufacturing statistics; test points, dummy devices, and unconfirmed critical anomaly records are not included.

[0072] For array devices that have been converted to internal cell statistics via S4, their independent cell records are used for aggregation, and the original physical package records are no longer added repeatedly according to the same caliber.

[0073] After aggregation is completed, a manufacturing bill of materials is generated. The manufacturing bill of materials can at least save the material code, material description, manufacturing requirement quantity, corresponding component reference number, main material and substitute material relationship, quantity statistics range, data version and review status.

[0074] At the same time, the system will generate a unified anomaly report for the tag number anomaly, coordinate anomaly, layer conflict, quantity and location number conflict, panel recognition anomaly, array device matching anomaly, material code anomaly, and manufacturing attribute classification anomaly generated during the aforementioned processing.

[0075] Example 2 Based on the inventive concept of Embodiment 1, this embodiment provides an implementation method for parsing engineering change documents. In actual SMT engineering management, after PCB products are released, component parameters, material codes, or component status may be modified through ECN or ECO files. Some engineering change documents are not fully structured data but contain handwritten text, arrows, circles, or lines. If only handwritten text is OCR-recognized, for example, to obtain a changed model number or deletion, it is still impossible to know which specific component the text refers to, and therefore the manufacturing BOM cannot be directly modified.

[0076] This embodiment first utilizes existing OCR and image recognition methods to obtain the handwritten text region, the recognized text, the OCR recognition confidence level, arrows, connecting lines, and selected areas. The recognition confidence level comes directly from the output of the adopted OCR model and is not fabricated by this system.

[0077] When the engineering change file contains graphics that can correspond to the PCB board surface, the page coordinates in the engineering change drawing are converted to the unified PCB coordinates in the same way as in S1, using board corners, fiducials, mounting holes or other corresponding references.

[0078] Subsequently, candidate associations between handwritten annotations and device entities are established based on graphical relationships. When an arrow or line clearly points to a device reference area, that device is considered a target candidate; when there is only one valid device reference within the selected area, that device is considered a target candidate; when there is no explicit graphical connection relationship, a candidate set can be generated based on the positional relationship between the annotation area and surrounding device areas, but a device is not forcibly identified as the target device simply because it is the closest.

[0079] The spatial tolerance required for determining the positions of endpoints, bounding boxes, etc. in an image is determined based on the actual pixel resolution of the image and the physical size after image coordinate transformation, rather than using a fixed number of millimeters that is unrelated to image quality.

[0080] The system will only write the changes obtained from OCR into the engineering change record of the target device when the target device can be uniquely identified; if there are multiple candidates that cannot be ruled out, an engineering change association anomaly will be generated.

[0081] Change types are identified based on the engineering change terminology rules currently used by the enterprise. For example, terms such as adding, deleting, replacing materials, and modifying packaging, along with their expressions, are stored in a versioned change rule library.

[0082] For newly added devices, the system creates a new standard device record and re-enters S2 and subsequent steps; for deletion operations, the status of the corresponding device in the current project version is set to invalid, so that it will no longer enter the current version of the manufacturing BOM, but the historical version data is retained; for field modifications, the value before modification, the value after modification, the ECN source, the project version, and the original OCR content are saved.

[0083] After the change is completed, it is not necessary to reprocess all engineering data unconditionally. Instead, corresponding steps are triggered based on the actual changed fields. For example, when the tag number changes, the device association is re-executed; when the package, device model, or device value changes, S3 verification is re-performed and it is determined whether array device parsing needs to be re-executed; when the material code, substitute material, or quantity changes, the manufacturing material aggregation in S5 is re-executed.

[0084] Example 3 This embodiment is based on an actual BOM file for illustration.

[0085] This BOM includes fields such as material number, name, material description, version, status, quantity, replacement status, and location number. An example material recorded in the actual materials includes material number 30080302000022, with a quantity of 2, and corresponding location numbers C86 and C87.

[0086] The system first parses the BOM fields according to S1, mapping data such as material number, quantity, location number, and substitution status to a unified field system.

[0087] For the above materials, the required quantity is 2, and the two device tag numbers C86 and C87 are obtained from the location number field. Therefore, two corresponding device records can be formed, and the consistency between the required quantity and the location number quantity can be checked.

[0088] When the required quantity differs from the number of position numbers obtained from parsing, an anomaly occurs indicating a mismatch between the generated quantity and the position numbers, rather than the position numbers being added or deleted automatically.

[0089] The actual BOM also contains material code 30080301000027 and its replacement part 30080301000047.

[0090] For this data, the system does not treat the two material codes as two materials that must be consumed simultaneously. Instead, it establishes a substitution relationship between the primary material 30080301000027 and the substitute material 30080301000047, thereby avoiding duplicate accumulation of procurement or production needs caused by substitute materials. This processing method is consistent with the principles in actual cases.

[0091] For records with missing fields, inconsistent quantities and location numbers, or unclear substitution relationships, generate corresponding exception records for engineers to review.

[0092] It should be noted that this practical case only obtained the BOM file, and did not simultaneously obtain the Gerber / ODB++, Pick & Place coordinate files, and ECN data. Therefore, this practical data example is used to illustrate BOM field parsing, quantity checking, and substitute material relationship processing, and should not be used as the actual verification result for layer identification, valid sub-board quantity identification, or engineering change identification.

[0093] Example 4 This embodiment also provides a PCB board BOM data parsing system for performing the above method.

[0094] The multi-source file parsing module in this system receives the BOM, board graphic file, mounting coordinate file and optional project file associated with the current project task, and outputs standard device records with project task identifier, source file identifier and standard device reference number. At the same time, it writes the spatial data after unified coordinate transformation into the corresponding record.

[0095] The component entity association and board structure recognition module reads the aforementioned standard component records, establishes component entities using the standard component reference numbers, and simultaneously reads the pads, layers, and board edge structures from the board file to generate candidate layer values ​​and the number of valid sub-boards for the components. .

[0096] The field validation module reads candidate fields from multiple sources within the device entity and their source information. Based on the corresponding field type, it performs consistency checks on characters, encodings, values, coordinates, layers, and quantities, and writes the results back to the device entity. For conflicting data, the field validation module generates an exception record; for data that passes validation or is manually confirmed, it is set as a valid field for subsequent manufacturing processes.

[0097] The array device deconstruction module only reads the confirmed package, model, device value, or material code, and calls the current version of the device standard library. When a unique internal cell definition is obtained, an independent cell record for the array device is formed, and this is combined with the number of devices on the board. Number of effective sub-boards and the number of internal units in the standard library Generate the corresponding manufacturing statistics, such as Figure 4 As shown.

[0098] The manufacturing attribute processing and bill of materials output module reads ordinary device records and array independent cell records, and simultaneously calls the current version of material coding rules, test point rules, dummy device rules and substitute material data to determine the actual set of records participating in manufacturing statistics, and finally aggregates them according to material codes to form a manufacturing bill of materials.

[0099] When an engineering change document exists, the engineering change processing module obtains the corresponding identification result, writes the uniquely associated change information back to the corresponding device entity, and then triggers the corresponding processing module to re-execute based on the changed fields.

[0100] Each module stores version numbers for its field mapping rules, tag number rules, layer mapping rules, material coding rules, test point and dummy component rules, source credibility configuration, and component standard library. Engineering tasks also record the actual rules and standard library versions used, so even if enterprise rules or component libraries are updated subsequently, it is still possible to restore the data and rules that a historical manufacturing BOM was based on at that time.

[0101] The threshold, weight, coefficient and other settings involved in the above embodiments can be set by default according to the present invention, or can be set by those skilled in the art according to actual engineering needs.

[0102] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0103] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0104] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0105] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0106] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A method for parsing PCB board BOM data for SMT production, characterized by: Includes the following steps: S1. Obtain and parse multi-source engineering files including bill of materials files, board graphic files and mounting coordinate files according to file type, map the component fields to a unified field system, and generate standard component records with data source identifiers; S2. Using the component reference number as the primary key, associate standard component records from different data sources as component entities; obtain the layer attributes and effective sub-boards of the component entities based on the layer information, pad information, and board edge contour information of the board surface graphic file; S3. Cross-compare the candidate field values ​​corresponding to the same target field in the device entity, determine the candidate field values ​​that match the verification as the target field values, mark the fields that do not match the verification as abnormal fields and generate abnormal records; S4. Identify array devices based on the target field value, query the device standard library to obtain the internal unit definition, generate independent unit records according to the internal unit definition, and obtain the internal unit statistical quantity based on the number of array devices in the corresponding device entity of the array device and its corresponding quantity statistical range, the number of internal units in the internal unit definition and the number of valid sub-boards. S5. Identify substitute material relationships, test points, and dummy devices for non-array device entity records and independent cell records. Separate test point records and dummy device records. Aggregate device records that participate in material quantity statistics according to material codes. Generate manufacturing bill of materials and anomaly reports.

2. The PCB board BOM data parsing method for SMT production according to claim 1, characterized in that: The process of generating the standard device record includes: For multi-source engineering files of different file types, select the corresponding file parsing rules; use the file parsing rules to extract the original device fields from each multi-source engineering file, and normalize at least one of the field name, data format of the field value, and coordinate format of the original device fields; map the normalized device fields to the unified field system; associate the mapped device fields with at least one of the source file identifier, source type, and original field value to obtain the standard device record; The multi-source engineering files include at least one of schematic diagram files, assembly drawing files, and engineering change files. The unified field system includes component reference number, component value, package specification, component model, coordinate position, layer, material code, and required quantity.

3. The PCB board BOM data parsing method for SMT production according to claim 1, characterized in that: The process of determining the layer attributes and effective number of sub-boards of the device entity includes: Read the layer attributes of the pads corresponding to the device entity, match the layer attributes with the layer names and layer types in the board surface graphic file, and determine the top layer attribute, bottom layer attribute or inner layer attribute corresponding to the device entity. Extract the board edge elements from the board surface graphic file and form candidate closed contours based on the connection relationship between the board edge elements. Based on the size, shape and inclusion relationship of the candidate closed contours, exclude the hole and slot contours located inside the product contour and the outer process contours surrounding multiple complete product areas. Determine the number of complete product contours with corresponding size and shape and repeated arrangement as the number of effective sub-boards. The determined layer attributes are written into the corresponding device entity, and the number of valid sub-boards is associated with the device entity as a quantity conversion parameter.

4. The PCB board BOM data parsing method for SMT production according to claim 1, characterized in that: The process of cross-comparing candidate field values ​​in S3 includes: Extract candidate field values ​​from different project files within the same device entity based on the target field; After unifying the character format of the candidate field values, compare whether the candidate field values ​​are consistent. For the numerical candidate field values ​​with units of measurement, unify the units. Determine the allowable range based on the data source precision and / or engineering rules corresponding to the candidate field values, and determine whether the numerical difference of the candidate field values ​​is within the allowable range. The candidate field values ​​that meet the consistency criteria in the comparison results are determined as the corresponding target field values; Fields whose comparison results do not meet the consistency criteria are identified as abnormal fields, and the corresponding device reference number, the value of each candidate field, and the data source identifier are saved.

5. The PCB board BOM data parsing method for SMT production according to claim 1, characterized in that: The identification and conversion of the array device includes: Extract at least one of the following from the device entity: package specification, device model, and device value, as a query feature for the array device; The array device query features are matched with the array device feature index pre-established in the device standard library, and the device entities with corresponding internal unit definitions are identified as array devices. Read the number of internal cells, type of internal cells, and identifier of internal cells recorded in the matched internal cell definition, generate independent cell records according to the internal cell identifiers, and associate the independent cell records with the device reference number of the original array device; The number of internal units corresponding to the independent unit record is determined based on the number of single-board array devices, the number of effective sub-boards, and the number of internal units in a single array device.

6. The PCB board BOM data parsing method for SMT production according to claim 1, characterized in that: The process of generating the manufacturing bill of materials includes: Extract material codes that conform to the preset material code format from non-array device entity records and independent cell records. For the same device record with multiple material codes, distinguish the main material code and the substitute material code according to the field position or substitute material identifier corresponding to the material code, and establish the substitute material relationship between the main material code and the substitute material code. Based on at least one of the device reference number, package specification, device value, and object marker in the mounting coordinate file in the record to be identified, the record to be identified is matched with the preset test point identification features and dummy device identification features, and the matched test point record and dummy device record are separated from the device record participating in the material statistics. The device records participating in the material statistics are grouped according to the material code, the required quantity corresponding to the same material code is accumulated, and a manufacturing bill of materials is generated based on the accumulation result and the substitute material relationship.

7. The PCB board BOM data parsing method for SMT production according to claim 5, characterized in that: The device standard library stores the correspondence between device features and internal unit definitions. The internal unit definition includes at least the number of internal units, the type of internal unit, and the internal unit identifier. The quantity statistical range of the array device corresponding to the device is determined. The quantity statistical range is determined based on the quantity range identifier in the multi-source engineering file, the configuration information exported from the engineering file, and / or the distribution of device coordinates in the corresponding area of ​​the effective sub-board. When the quantity statistics range is within the range of a single valid sub-board, the total number of array devices is determined according to the following relationship: ,in, This indicates the number of array devices within a single effective sub-board. Indicates the number of valid sub-boards; When the quantity statistics range is already the range of the entire panel, the number of array devices corresponding to the entire panel range is determined as the total number of array devices. Furthermore, the internal unit count will not be repeatedly calculated based on the number of valid sub-boards; the internal unit count will be determined according to the following relationship: ,in, This indicates the number of internal cells in a single array device as defined in the internal cell definition. This indicates the statistical quantity of the internal units; When there are array devices with different internal unit definitions, the corresponding internal unit count is determined according to the number of devices, the count range, and the number of internal units for each array device.

8. The PCB board BOM data parsing method for SMT production according to claim 6, characterized in that: The process of identifying the test point records and dummy device records includes: Extract at least one feature from the device reference number, package specification, device value, and object marker in the mounting coordinate file of the record to be identified; The features are matched with preset test point identification rules and fake device identification rules, wherein the test point identification rules and fake device identification rules include at least one of the following: reference number rules, package rules, device value rules and object marking rules; Based on the matching results of the features with the test point identification rules and the fake device identification rules, the record to be identified is determined to be a test point record, a fake device record, or a normal device record.

9. The PCB board BOM data parsing method for SMT production according to claim 1, characterized in that: When the multi-source engineering file includes an engineering change file with handwritten annotations, it further includes: identifying at least one graphic mark from the engineering change file, including a handwritten annotation area and an arrow, line, and circle area associated with the handwritten annotation area; Candidate device entities are determined based on the pointing, connection, inclusion, and / or spatial relationships between the graphic markers and the areas where the device reference numbers are located; When the candidate device entity can be uniquely identified, the corresponding device entity is identified as the target device entity; when there are multiple candidate device entities that cannot be excluded, an engineering change association exception is generated, and the automatic execution of the corresponding engineering change is stopped. The character content in the handwritten annotation area is recognized, and the recognition result is parsed into addition, deletion or field modification operations and corresponding changed field values ​​according to the engineering change rules associated with the current engineering task. For deletion or field modification operations, the status or field of the corresponding target device entity is updated; for addition operations, a new standard device record is created based on the identified new device information. Re-execute at least one of the corresponding processing steps S2 to S5 based on the changed field to update the manufacturing bill of materials and save the source of change, device reference number, value before change, and value after change.

10. A PCB board BOM data parsing system for SMT production, used to execute the PCB board BOM data parsing method for SMT production as described in any one of claims 1-9, characterized in that: The system includes: The multi-source file parsing module is used to parse multi-source engineering files, convert device fields to a unified field system, and output standard device records with data source identifiers; The device entity association and board structure identification module is used to read the standard device record and establish device entities according to the device reference number, and output the board layer information and effective sub-board number corresponding to the device entity based on the layer information, pad information and board edge structure in the board graphic file. The field verification module is used to read the multi-source candidate field values ​​in the device entity, perform cross-source consistency verification on the multi-source candidate field values, and output the target field value and / or abnormal records; The array device deconstruction module is used to read the verified target field value and query the device standard library. Based on the matched internal cell definition, it generates an independent cell record associated with the original array device and determines the internal cell count based on the number of devices corresponding to the array device and its count range, the number of valid daughter boards and the number of internal cells. The manufacturing attribute processing and bill of materials output module is used to receive non-array device entity records and independent unit records, determine the device records participating in the material quantity statistics according to the test point identification rules and dummy device identification rules, establish the substitution material relationship between the master material code and the substitution material code, and aggregate the device records participating in the material quantity statistics according to the material code to generate a manufacturing bill of materials.

Citation Information

Patent Citations

  • Visual Inspection Methods for Design Manufacturability of Electronic Products (EDA)

    CN102930114B

  • Printed circuit board BOM sorting method and system

    CN105761045A

  • Correction method and system for surface mounting machine importing production data

    CN107133724A

  • A method for parsing and recognizing BOM files and an electronic device

    CN113627893B