KR102509361B1 - Smart BOM system - Google Patents

Abstract

According to the present invention, a smart BOM system comprises: an information input unit in which bill of material (BOM) data is uploaded; a determination unit that analyzes the uploaded BOM data and determines part information from the BOM data; and an information providing unit that searches for the part information determined by the determination unit through a preset search engine to provide search result data.

Description

Smart BOM system {Smart BOM system}

The present invention relates to a smart BOM system.

In order to manufacture and produce an item, it goes through a production plan that oversees a series of information flows from the purchase of materials to the delivery date of the product, and an action plan to operate the manufacturing line within the period set in the production plan.

Here, the purchase of materials has been made in such a way that a customer requests a quotation, a person in charge checks the quotation, and then creates a quotation.

In addition, a lot of physical time was required because the person in charge had to check the price and inventory of each part before making a quote.

In particular, the person in charge had to search the Internet or contact the company by individual part number in order to check the existing parts inventory and price.

Here, the customer must deliver the part information to the person in charge. At this time, an order statement, that is, a bill of material (BOM), which is a collection of parts information to be purchased by the customer, is delivered.

BOM is a word mainly used in electronics, and refers to a collection of data such as parts necessary for manufacturing a product into one file.

For example, if parts are specified in the circuit design stage, the ECAD side records the names of the parts required for the BOM, and later sends this BOM to the parts site, and the person in charge of the parts site checks the orderable parts (stock, price) on the Internet. and write a quote.

Since the BOM contains information on multiple parts, not just one part, it was difficult to find each part and check stock, supplier availability, and lowest price.

In addition, since the BOM format used by each company is different and the file format is different, it is impossible to analyze with traditional keyword crawling technology, so there is a problem that can only be done manually.

Republic of Korea Patent Publication 10-2019-0114323

An object of the present invention, which has been devised to solve the problems described above, is to determine part information from various BOM data uploaded by a user, search through a preset search engine, and quickly obtain detailed information of parts included in the BOM data. This is to provide a smart BOM system that can automatically and quickly proceed with quotation and ordering while simultaneously confirming.

According to the smart BOM system of the present invention for achieving the object as described above, BOM (Bill of Material) data is uploaded information input unit; a determination unit that analyzes the uploaded BOM data and determines part information from the BOM data; and an information providing unit that searches parts information determined by the determination unit through a preset search engine and provides search result data.

In addition, the determination unit may use an artificial intelligence learning unit for learning column and row data from learning data in the same format as the BOM data, and a first keyword having the same meaning as or the same meaning as the header keyword through learning of the learning data. A keyword collection unit that collects and collects a second keyword that is the same as or has the same meaning as an item keyword, and determines whether the first keyword and the second keyword match from the BOM data, so as to determine column and row data of the BOM data. and an extraction unit for determining whether is a header keyword or item keyword and extracting parts information.

In addition, the information providing unit, an artificial intelligence search unit for searching the parts information extracted from the extraction unit with a preset search engine, and extracting search data searched by the artificial intelligence search unit for preset provided information and displaying it on a display unit includes the display.

In addition, the BOM data includes header information of one or more of product name, reference, part number, description, Q'ty (order quantity), manufacturer, and package. , Each header information includes at least one item information, and the information provided by the display unit includes the name of a supplier capable of supplying a part that matches the part information of the BOM data, and the price sold by the supplier. information, inventory information that can be supplied by the supplier, and information on suppliers that can supply the parts other than the supplier, and the display unit sorts and displays based on price information of the provided information, A quotation generating unit for generating an automatic quotation of parts included in the BOM data with the price information of the provided information; And a data management unit for modifying and managing the BOM data uploaded to the information input unit; more includes

As described above, the effect of the present invention is to determine various parts information from the BOM data uploaded by the user, search with a preset search engine, quickly check the detailed information of the parts included in the BOM data, and at the same time automatically can provide a smart BOM system that can quickly process quotations and orders.

1 is a configuration diagram showing a smart BOM system according to a preferred embodiment of the present invention.
Figure 2 is an exemplary view showing the BOM data format of the smart BOM system according to a preferred embodiment of the present invention.
3 is a state diagram showing a state of uploading BOM data of a smart BOM system according to a preferred embodiment of the present invention.
4 is an exemplary view showing BOM data uploaded to the information input unit of the smart BOM system according to a preferred embodiment of the present invention.
5 is an exemplary view showing search result data according to the information providing unit of the smart BOM system according to a preferred embodiment of the present invention.
Figure 6 is an exemplary view showing in detail the supplier information of the smart BOM system according to a preferred embodiment of the present invention.
7 is an exemplary diagram showing a data management unit of a smart BOM system according to a preferred embodiment of the present invention.

Advantages and features of the present invention, and methods of achieving them, will become clear with reference to the detailed description of the following embodiments taken in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only these embodiments make the disclosure of the present invention complete, and common knowledge in the art to which the present invention belongs. It is provided to completely inform the person who has the scope of the invention, and the present invention is only defined by the scope of the claims. Like reference numbers designate like elements throughout the specification.

Hereinafter, the present invention will be described with reference to drawings for explaining a smart BOM system according to embodiments of the present invention.

Referring to FIG. 1 , the smart BOM system according to the present invention includes an information input unit 10, a determination unit 20 and an information providing unit 30.

First, in the information input unit 10, BOM (Bill of Material) data is uploaded.

At this time, the BOM data is made in a certain format. For example, it may be made of a sheet composed of rows and columns as shown in FIG. 2, but is not limited thereto. That is, the BOM data file as shown in FIG. 2 is uploaded as shown in FIG. 3 .

BOM data includes header information of one or more of item name, reference, part number, description, Q'ty (order quantity), manufacturer, and package, and each Each header information includes at least one item information. For example, the number of parts of the BOM data and the item name of each part can be checked through the item information of the header information 'product name'.

When the above-described BOM data is uploaded by the user, the determination unit 20 analyzes the uploaded BOM data and determines part information from the BOM data.

At this time, the BOM data may be formed in any one file format among file formats such as xlsx, csv, xls, and ods. In addition, the BOM data may include header information and item information in various formats for each file format. Regarding the form, for example, if header information is described in a row or column direction, item information may be described in a column or row direction.

Here, the file format is described in detail: CSV (*.csv), Microsoft Excel Worksheet (*.xls, *.xlsx, *.xlt, *.xltx), Portable Document Format (*.pdf), Tab Delimited Text (*.txt), Web Page (*.htm, *.html), XML Spreadsheet (*.xml), etc., so description is not limited to one.

In this way, as the determination unit 20 analyzes various file formats and various forms to determine each part information, even if the file format and format used by each user are different, header information and item information are obtained from the BOM data uploaded by each user. It is possible to analyze and determine part information.

In this way, in order for the determination unit 20 to analyze the BOM data consisting of various file formats and formats, it is preferable that BOM data consisting of various file formats and formats, that is, learning data through artificial intelligence is performed.

Looking at artificial intelligence, there are several technologies in the field of artificial intelligence. First, machine learning is self-learning using input information given only basic rules. An artificial neural network is a machine learning model modeled after the structure of a human neuron. Deep learning is a study that mainly deals with artificial neural network techniques in which artificial neurons between input and output are stacked and connected. That is, rather than a single layer, it is made up of multiple layers, just like a real brain. Cognitive computing refers to programs or solutions that use machine learning to solve specific cognitive tasks. Finally, neuromorphic computing is a hardware implementation of artificial neural networks.

In addition, the determination unit 20 may analyze and determine the BOM data uploaded by the user as learned after learning the learning data through various artificial intelligence techniques described above.

Accordingly, in detail, the determination unit 20 includes an artificial intelligence learning unit 21, a keyword collection unit 22 and an extraction unit 23.

First, the artificial intelligence learning unit 21 learns column and row data from learning data including a plurality of BOM data composed of the above-described various formats and file formats.

At this time, the learning data may be made in any one file format among xlsx, csv, xls, and ods. At least, it is preferable that one or more formats are also variously configured for each file format. Here, the BOM data may consist of data collected from forms and file formats used by users who actually order parts, but are not limited thereto.

The artificial intelligence learning unit 21 learns header information and item information from the above-described BOM data.

Here, the artificial intelligence learning unit 21 may be formed of a TF-IDF (Term Frequency-Inverse Document Frequency) algorithm. TF-IDF is a method of weighting the importance of each word in the DTM using the frequency of a word and the inverse document frequency (taking a specific formula for the frequency of a document).

More specifically, after creating the DTM, TF-IDF weights are assigned. At this time, TF-IDF assigns a value in such a way that a high weight is given to words that frequently appear in individual documents and a penalty is given to words that appear frequently in all documents.

For example, assuming that there are a total of 5 documents, the word 'deep learning' appears in all 5 documents, and the word 'machine learning' appears frequently only in document 1, the value of TF-IDF is The word 'deep learning' is given low and the word 'machine learning' is given high.

The keyword collection unit 22 collects first keywords that are identical to or have the same meaning as the header keywords through learning of the learning data, and collects second keywords that are identical to or have the same meanings as the item keywords. The collected keywords are stored by dividing the first keyword and the second keyword.

In this way, the keyword collection unit 22 collects the first keyword corresponding to the header keyword and the second keyword corresponding to the item keyword, and vectorizes the collected keywords using the TF-IDF algorithm to prepare for BOM data analysis. .

The extraction unit 23 determines whether the first keyword and the second keyword match from the BOM data uploaded by the user, determines whether the column and row data of the BOM data are header keywords or item keywords, and extracts part information. .

At this time, the extraction unit 23 may be performed through a logistic regression algorithm among machine learning algorithms.

Here, machine learning algorithms are divided into supervised learning and unsupervised learning, and the logistic regression algorithm is a representative algorithm in supervised learning. In addition, machine learning algorithms include support vector machine (SVM), Naive Bayes classification, linear regression, quantile regression, random forest, and neural networks, etc.

Logistic regression is one of the regression algorithms that predict whether data will belong to a specific category with a non-linear, continuous probability between 0 and 1.

For example, pass is expressed as 1 and fail is expressed as 0. The problem of choosing one of these two options is used in binary classification.

Accordingly, the keywords are learned by the TF-IDF algorithm, and the keywords collected by the keyword collection unit 22 are used to determine whether the header keywords or item keywords are correct in the BOM data uploaded by the user using the logistic regression machine learning algorithm. It is determined by the extraction unit 23.

That is, the extractor 23 analyzes each row of the input BOM data to determine whether it corresponds to a header or an item, and then analyzes each column to determine which item it corresponds to.

For example, when BOM data as shown in FIG. 4 is uploaded, the extraction unit 23 analyzes each row to determine whether it is header information or item information. At this time, referring to FIG. 3, it can be determined that the columns of the third row are header information, and the columns from the fourth row are item information. Accordingly, by analyzing rows and columns of the BOM data, it is possible to identify each piece of PCB component information included in the BOM data. That is, detailed information of PCB parts information can be grasped through item information of each header information.

The information providing unit 30 searches the parts information determined by the determining unit 20 through a preset search engine and provides search result data.

Here, the information providing unit 30 includes an artificial intelligence search unit 31 and a display unit 32.

The artificial intelligence search unit 31 searches for parts information extracted by the extraction unit 23 through a preset search engine. For example, part information can be determined by the part number, and the artificial intelligence search unit 31 can search the part number (part name) in a preset search engine, but is not limited thereto.

In addition, the artificial intelligence search unit 31 may search for all part information that matches or is similar to part information. In addition, the artificial intelligence search unit 31 may search for a company name 33, price information 34, and inventory information 35 to be displayed on a display unit 32 to be described later and form a database.

In addition, the artificial intelligence search unit 31 may request additional information from a manager or user when search is not performed due to lack of parts information extracted from BOM data.

The display unit 32 extracts preset provided information from the search data searched by the artificial intelligence search unit 31 and displays it on the display unit.

The information provided by the display unit 32 includes the name of a supplier capable of supplying parts that match the part information of the BOM data 33, price information sold by the supplier 34, and inventory information available from the supplier. (35), may include any one or more of supplier information (36) that can supply parts other than suppliers, but is not limited thereto.

In addition, the display unit 32 may sort and display the price information 34 of the provided information. At this time, the display unit 32 may also display margin information according to a predetermined margin rate through the price information 34 .

For example, the display unit 32 may display the provided information on the display unit as shown in FIG. 5 . At this time, as shown in FIG. 5, when the display unit 32 displays the provision information on the display unit, the company name of a supplier that matches the supplier information included in the BOM data may be displayed preferentially.

In addition, as shown in FIG. 5 , the display unit 32 may display a packaging unit, a current stock, a minimum purchase quantity, a required quantity, an additional quantity, and a final quantity.

In addition, the display unit 32 can finally update the parts information and display the past time, so that the latest information can be further confirmed, and the order quantity can be easily adjusted and confirmed by displaying margin information for each quantity.

In addition, suppliers may supply parts through a plurality of online markets, and the supplied prices may be different. Accordingly, the display unit 32 may preferentially display the name of a supplier that provides the lowest price among suppliers that match the supplier information included in the BOM data, but is not limited thereto.

In addition, as shown in FIG. 6, for the supplier information 36 displayed on the display unit, the margin for each packaging unit, current inventory, minimum purchase quantity, required quantity, additional quantity, final quantity, update time, and quantity for each supplier By displaying information, you can easily adjust and confirm the order quantity and display it in detail.

In addition, when displaying the supplier information 36, the display unit 32 arranges and displays the available supplier information 36 in order of price, but may display the lowest price first, but is not limited thereto.

In addition, the user can select a part displayed on the display unit 32 through a checkbox as shown in FIG. 5 and store it in a 'cart' or 'MYBOM' shown in FIG. 7 . These stored details can be ordered and paid online at any time.

In addition, a search function for checking order details and delivery details, a function for requesting a refund/return/AS according to an order, and a function for providing reserve funds according to purchases may be provided.

On the other hand, the smart BOM system according to the present invention may further include a quote generator and a data management unit.

The quotation generating unit may generate an automatic quotation of parts included in the BOM data as the price information 34 of the provided information. Accordingly, as the search is completed, it is possible to immediately check the quotation, so that quick confirmation and ordering can be made.

The data management unit corrects and manages the BOM data uploaded to the information input unit 10 .

Referring to FIG. 7 , the uploaded BOM data is information determined through the determination unit 20, and the header name and number of parts can be checked, the BOM data can be downloaded, and the BOM data can be checked through the viewing function. there is.

As described above, the smart BOM system of the present invention determines parts information from various BOM data uploaded by the user, searches with a preset search engine, and provides the lowest price information among parts of the company desired by the user, thereby allowing the user to view the BOM data. You can quickly check the detailed information of the parts included in the BOM, so you can quickly correct errors, etc. BOM data management is also effective by easily performing information modification, deletion, and addition of information.

Those skilled in the art to which the present invention pertains will understand that the present invention can be embodied in other specific forms without changing its technical spirit or essential features. Therefore, the embodiments described above should be understood as illustrative in all respects and not limiting. The scope of the present invention is indicated by the claims to be described later rather than the detailed description above, and all changes or modifications derived from the meaning and scope of the claims and equivalent concepts thereof are included in the scope of the present invention. should be interpreted In addition, the operation sequence of the components described in the above-described process does not necessarily have to be performed in a time-series order, and even if the execution sequence of each component and step is changed, if the gist of the present invention is satisfied, these processes will fall within the scope of the present invention. Of course you can.

10: information input unit 20: determination unit
21: artificial intelligence learning unit 22: keyword collection unit
23: extraction unit 30: information provision unit
31: artificial intelligence search unit 32: display unit
33: Company name 34: Price information
35: inventory information 36: supplier information

Claims (4)

An information input unit in which BOM (Bill of Material) data is uploaded;
a determination unit that analyzes the uploaded BOM data and determines part information from the BOM data; and
An information providing unit that searches the parts information determined by the determination unit with a preset search engine and provides search result data;
The judge,
An artificial intelligence learning unit for learning column and row data from learning data in the same format as the BOM data;
A keyword collection unit that collects first keywords identical to header keywords and collects second keywords identical to item keywords through learning of the learning data;
An extraction unit for determining whether the first keyword and the second keyword match from the BOM data, determining whether column and row data of the BOM data are header keywords or item keywords, and extracting part information,
The BOM data includes header information of one or more of product name, reference, part number, description, Q'ty (order quantity), manufacturer, and package, respectively. Includes at least one item information for each header information of,
The artificial intelligence learning unit extracts column and row data from BOM data composed of any one of xlsx, csv, xls, and ods file formats, learns header information and item information from the extracted data, and Depending on the frequency of occurrence of item information, weights are assigned differently to learn,
The information provision department,
An artificial intelligence search unit for searching the parts information extracted from the extraction unit with a preset search engine;
A display unit extracting preset provided information from the search data searched by the artificial intelligence search unit and displaying it on a display unit;
The information provided by the display unit includes the name of a supplier capable of supplying a part that matches the parts information of the BOM data, price information sold by the supplier, inventory information that the supplier can supply, and other than the supplier. Including any one or more of the supplier information that can supply the parts,
The display unit sorts and displays the price information of the provided information,
A quotation generating unit for generating an automatic quotation of parts included in the BOM data with the price information of the provided information; and
a data management unit for modifying and managing the BOM data uploaded to the information input unit; Smart BOM system that includes. delete delete delete

Images