Invoice matching method, invoice matching system, and invoice matching program

Abstract

This invention provides an invoice matching method, system, and program that reduce the burden of matching work. [Solution] The method reads multiple invoice data and total invoice amounts, associates the invoice amount with each invoice record, identifies the total invoice amount, reads multiple purchase data including transaction codes associated with purchase amounts, identifies which invoice record each purchase data corresponds to, identifies by transaction code, classifies the data into three categories: a first invoice record where the invoice amount associated with the invoice record matches the purchase amount associated with the paired purchase record; a second invoice record where a purchase record was paired but the invoice amount associated with the invoice record does not match the purchase amount associated with the purchase record; and a third invoice record where no purchase record was paired. The difference between the total purchase amount, the sum of the amounts with differences, and the sum of the unrecorded amounts, and the total invoice amount is then calculated.

Description

Technical Field

[0001] The present invention relates to a bill matching method, a bill matching system, and a bill matching program.

Background Art

[0002] As one of the general affairs operations, there is an operation (so-called "matching") of visually matching a bill from a supplier with purchase data (also called purchase confirmation data) managed by the company itself to identify a discrepancy between the two. This matching operation is very burdensome.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The problems of the present invention are to reduce the burden of the matching operation or to identify a business partner or a billing party from a bill.

Means for Solving the Problems

[0005] As an example, the following solutions are provided.

[0006] [0] A bill reading unit reads bill data including a plurality of bill records and information on the total bill amount, associates a bill amount with each of the plurality of bill records, and identifies the total bill amount; A purchase data reading unit reads purchase data including a plurality of purchase records each associated with a purchase amount; ​The pairing means pairs each of the plurality of invoice records with each of the plurality of purchase records, and each of the invoice records, The first invoice record can be paired with any purchase record, and the invoice amount associated with that invoice record matches the purchase amount associated with the paired purchase record. A second invoice record was paired with one of the purchase records, but the invoice amount associated with that invoice record did not match the purchase amount associated with the paired purchase record. The third invoice record, which is not paired with any purchase record, is classified as follows: The method for calculating the total purchase amount includes the step of calculating the total purchase amount, which is the sum of the purchase amounts associated with purchase records paired with any invoice record, The means for identifying differences in amounts includes the steps of identifying the difference in amounts between the invoice amount associated with the second invoice record and the purchase amount associated with the paired purchase record, and calculating the total amount of the differences in amounts, The means for identifying unrecorded amounts includes the steps of identifying the invoice amount associated with the third invoice record as an unrecorded amount and calculating the total amount of the unrecorded amounts, An invoice matching method that includes the step of calculating the difference between (1) the sum of the total purchase amount, the sum of the amount differences, and the sum of the unrecorded amounts, and (2) the total invoice amount.

[0007] [1] The invoice reading means reads invoice data including multiple invoice records and information on the total invoice amount, associates the invoice amount with each of the multiple invoice records, and identifies the total invoice amount. The purchase data reading means includes the step of reading purchase data which includes multiple purchase records, each of which is associated with a purchase amount, and each of the purchase records includes a transaction code, The transaction identification means includes the steps of identifying the purchase record that corresponds to one of the plurality of invoice records, and identifying the transaction code included in the identified purchase record, The pairing means pairs each of the plurality of invoice records with each of the plurality of purchase records that includes the identified transaction code, and each of the invoice records, The first invoice record can be paired with any purchase record, and the invoice amount associated with that invoice record matches the purchase amount associated with the paired purchase record. A second invoice record was paired with one of the purchase records, but the invoice amount associated with that invoice record did not match the purchase amount associated with the paired purchase record. The third invoice record, which is not paired with any purchase record, is classified as follows: The method for calculating the total purchase amount includes the step of calculating the total purchase amount, which is the sum of the purchase amounts associated with purchase records paired with any invoice record, The means for identifying differences in amounts includes the steps of identifying the difference in amounts between the invoice amount associated with the second invoice record and the purchase amount associated with the paired purchase record, and calculating the total amount of the differences in amounts, The means for identifying unrecorded amounts includes the steps of identifying the invoice amount associated with the third invoice record as an unrecorded amount and calculating the total amount of the unrecorded amounts, An invoice matching method that includes the step of calculating the difference between (1) the sum of the total purchase amount, the sum of the amount differences, and the sum of the unrecorded amounts, and (2) the total invoice amount.

[0008] [2] Each of the aforementioned billing records includes a transaction identification code, Each of the aforementioned purchase records includes a transaction identification code, The invoice matching method according to [1], wherein the transaction identification means identifies the purchase record containing a transaction identification code that corresponds to a transaction identification code contained in any of the plurality of invoice records.

[0009] [3] The invoice reading means reads invoice data including multiple invoice records and information on the total invoice amount, associates the invoice amount with each of the multiple invoice records, and identifies the total invoice amount, wherein each of the invoice records includes one or more predetermined items and a transaction identification code. A purchase data reading means reads purchase data, which includes multiple purchase records, each of which is associated with a purchase amount, and each of the purchase records includes one or more predetermined items, a transaction code, and a transaction identification code. Transaction identification method, (1) Identify one or more purchase records that include one or more specified items that match one or more specified items included in any of the multiple invoice records, (2) For each purchase code that shares a transaction code with one or more identified purchase records, the transaction identification code contained in each purchase record is matched with the invoice record containing the corresponding transaction identification code. Purchase records for which no invoice record containing the corresponding transaction identification code exists are not matched with any invoice records. (3) The step of identifying the transaction code that appears most frequently among the purchase records that are matched with any invoice record, The pairing means pairs each of the plurality of invoice records with each of the plurality of purchase records that includes the identified transaction code, and each of the invoice records, The first invoice record can be paired with any purchase record, and the invoice amount associated with that invoice record matches the purchase amount associated with the paired purchase record. A second claim record that could be paired with any purchase record, but the claim amount associated with the claim record does not match the purchase amount associated with the paired purchase record, and classifying into a third claim record that could not be paired with any purchase record, and a step in which the total purchase amount calculation means calculates a total purchase amount that is the total value of the purchase amounts associated with the purchase records paired with any claim record, and a step in which the amount difference specifying means specifies the amount difference between the claim amount associated with the second claim record and the purchase amount associated with the paired purchase record, and calculates the total amount of the amount differences, and a step in which the unaccounted amount specifying means specifies the claim amount associated with the third claim record as an unaccounted amount, and calculates the total amount of the unaccounted amounts, and a claim reconciliation method including a step in which the difference calculation means calculates the difference between (1) the total amount of the total purchase amount, the total amount of the amount differences, and the total amount of the unaccounted amounts, and (2) the total claim amount.

[0010] [4] The claim reconciliation method according to [3], wherein the one or more predetermined items are a set of quantity, unit price, and amount.

[0011] [5] The claim reconciliation method according to any one of [1] to [4], wherein the transaction code is assigned for each branch of the payment destination.

[0012] [6] The claim reconciliation method according to any one of [1] to [5], wherein the transaction code is assigned for each branch of the claim destination.

[0013] [7] The claim reconciliation method according to any one of [2] to [6], wherein the transaction identification code is an order number, a purchase order number, an invoice number, or a delivery note number.

[0014] [8] An invoice reading means reads invoice data containing multiple billing records and information on the total billing amount, associates the billing amount with each of the multiple billing records, and identifies the total billing amount. A purchase data reading means for reading purchase data that includes multiple purchase records, each of which is associated with a purchase amount, wherein each of the purchase records includes a purchase data reading means that includes a transaction code, Transaction identification means for identifying the purchase record corresponding to any of the aforementioned multiple invoice records, and for identifying the transaction code included in the identified purchase record, Each of the aforementioned multiple invoice records is paired with each of the aforementioned multiple purchase records that contains the identified transaction code, and each of the aforementioned invoice records is then paired with, The first invoice record can be paired with any purchase record, and the invoice amount associated with that invoice record matches the purchase amount associated with the paired purchase record. A second invoice record was paired with one of the purchase records, but the invoice amount associated with that invoice record did not match the purchase amount associated with the paired purchase record. A third invoice record that has not been paired with any purchase record, and a pairing method for classifying them as follows: A means for calculating the total purchase amount, which is the sum of the purchase amounts associated with purchase records paired with any invoice record, Regarding the second invoice record, a means for identifying the difference in amount between the invoice amount associated with the second invoice record and the purchase amount associated with the paired purchase record, and for calculating the total amount of the difference, With respect to the third billing record, the means for identifying the billing amount associated with the third billing record as an unaccounted amount and calculating the total amount of the unaccounted amount, An invoice matching system comprising: (1) the sum of the total purchase amount, the sum of the amount differences, and the sum of the unrecorded amounts, and (2) the total invoice amount, and a difference calculation means for calculating the difference between these two amounts.

[0015] [9] An invoice reading means reads invoice data including multiple invoice records and information on the total invoice amount, associates the invoice amount with each of the multiple invoice records, and identifies the total invoice amount, wherein each of the invoice records includes one or more predetermined items and a transaction identification code. A purchase data reading means for reading purchase data that includes multiple purchase records, each of which is associated with a purchase amount, wherein each of the purchase records includes one or more predetermined items, a transaction code, and a transaction identification code, (1) Identify one or more purchase records that include one or more specified items that match one or more specified items included in any of the multiple invoice records, (2) For each purchase code that shares a transaction code with one or more identified purchase records, the transaction identification code contained in each purchase record is matched with the invoice record containing the corresponding transaction identification code. Purchase records for which no invoice record containing the corresponding transaction identification code exists are not matched with any invoice records. (3) A transaction identification means that identifies the transaction code that appears most frequently among the transaction codes included in the purchase record that is matched with any invoice record, Each of the aforementioned multiple invoice records is paired with each of the aforementioned multiple purchase records that contains the identified transaction code, and each of the aforementioned invoice records is then paired with, The first invoice record can be paired with any purchase record, and the invoice amount associated with that invoice record matches the purchase amount associated with the paired purchase record. A second invoice record was paired with one of the purchase records, but the invoice amount associated with that invoice record did not match the purchase amount associated with the paired purchase record. A third invoice record that has not been paired with any purchase record, and a pairing method for classifying them as follows: A means for calculating the total purchase amount, which is the sum of the purchase amounts associated with purchase records paired with any invoice record, Regarding the second invoice record, a means for identifying the difference in amount between the invoice amount associated with the second invoice record and the purchase amount associated with the paired purchase record, and for calculating the total amount of the difference, With respect to the third billing record, the means for identifying the billing amount associated with the third billing record as an unaccounted amount and calculating the total amount of the unaccounted amount, An invoice matching system comprising: (1) the sum of the total purchase amount, the sum of the amount differences, and the sum of the unrecorded amounts, and (2) the total invoice amount, and a difference calculation means for calculating the difference between these two amounts.

[0016]

[10] One or more computers, An invoice reading means reads invoice data containing multiple billing records and information on the total billing amount, associates the billing amount with each of the multiple billing records, and identifies the total billing amount. A purchase data reading means for reading purchase data that includes multiple purchase records, each of which is associated with a purchase amount, wherein each of the purchase records includes a purchase data reading means that includes a transaction code, Transaction identification means for identifying the purchase record corresponding to any of the aforementioned multiple invoice records, and for identifying the transaction code included in the identified purchase record, Each of the aforementioned multiple invoice records is paired with each of the aforementioned multiple purchase records that contains the identified transaction code, and each of the aforementioned invoice records is then paired with, The first invoice record can be paired with any purchase record, and the invoice amount associated with that invoice record matches the purchase amount associated with the paired purchase record. A second invoice record was paired with one of the purchase records, but the invoice amount associated with that invoice record did not match the purchase amount associated with the paired purchase record. A third invoice record that has not been paired with any purchase record, and a pairing method for classifying them as follows: A means for calculating the total purchase amount, which is the sum of the purchase amounts associated with purchase records paired with any invoice record, Regarding the second invoice record, a means for identifying the difference in amount between the invoice amount associated with the second invoice record and the purchase amount associated with the paired purchase record, and for calculating the total amount of the difference, With respect to the third billing record, the means for identifying the billing amount associated with the third billing record as an unaccounted amount and calculating the total amount of the unaccounted amount, An invoice matching program that functions as a difference calculation means for calculating the difference between (1) the total amount of purchases, the total amount of discrepancies, and the total amount of unrecorded items, and (2) the total invoice amount.

[0017]

[11] One or more computers, An invoice reading means reads invoice data including multiple invoice records and information on the total invoice amount, associates the invoice amount with each of the multiple invoice records, and identifies the total invoice amount, wherein each of the invoice records includes one or more predetermined items and a transaction identification code. A purchase data reading means for reading purchase data that includes multiple purchase records, each of which is associated with a purchase amount, wherein each of the purchase records includes one or more predetermined items, a transaction code, and a transaction identification code, (1) Identify one or more purchase records that include one or more specified items that match one or more specified items included in any of the multiple invoice records, (2) For each purchase code that shares a transaction code with one or more identified purchase records, the transaction identification code contained in each purchase record is matched with the invoice record containing the corresponding transaction identification code. Purchase records for which no invoice record containing the corresponding transaction identification code exists are not matched with any invoice records. (3) A transaction identification means that identifies the transaction code that appears most frequently among the transaction codes included in the purchase record that is matched with any invoice record, Each of the aforementioned multiple invoice records is paired with each of the aforementioned multiple purchase records that contains the identified transaction code, and each of the aforementioned invoice records is then paired with, The first invoice record can be paired with any purchase record, and the invoice amount associated with that invoice record matches the purchase amount associated with the paired purchase record. A second invoice record was paired with one of the purchase records, but the invoice amount associated with that invoice record did not match the purchase amount associated with the paired purchase record. A third invoice record that has not been paired with any purchase record, and a pairing method for classifying them as follows: A means for calculating the total purchase amount, which is the sum of the purchase amounts associated with purchase records paired with any invoice record, Regarding the second invoice record, a means for identifying the difference in amount between the invoice amount associated with the second invoice record and the purchase amount associated with the paired purchase record, and for calculating the total amount of the difference, With respect to the third billing record, the means for identifying the billing amount associated with the third billing record as an unaccounted amount and calculating the total amount of the unaccounted amount, An invoice matching program that functions as a difference calculation means for calculating the difference between (1) the total amount of purchases, the total amount of discrepancies, and the total amount of unrecorded items, and (2) the total invoice amount.

[0018]

[12] A method for identifying a customer or billing party from invoice data, The database stores the correspondence between transaction codes and the trading partner or billing address. The invoice reading method includes the step of reading invoice data containing multiple invoice records, A purchase data reading means reads purchase data which includes multiple purchase records, wherein each of the purchase records includes a transaction code. A method for identifying a transaction means, comprising the steps of (1) identifying a purchase record that corresponds to one of the plurality of invoice records, (2) identifying a transaction code included in the identified purchase record, and (3) referring to the database to identify a customer or billing party from the identified transaction code.

[0019]

[13] A system that identifies a business partner or billing address from invoice data, The database stores the correspondence between transaction codes and the trading partner or billing address. An invoice reading method that reads invoice data containing multiple billing records, A purchase data reading means for reading purchase data that includes multiple purchase records, wherein each of the purchase records includes a purchase data reading means that includes a transaction code, A system comprising: (1) identifying a purchase record that corresponds to any of the multiple invoice records; (2) identifying a transaction code included in the identified purchase record; and (3) a transaction identification means that, by referring to the database, identifies a trading partner or billing party from the identified transaction code.

[0020]

[14] A program that identifies a customer or billing address from invoice data, The database stores the correspondence between transaction codes and the trading partner or billing address. One or more computers, An invoice reading method that reads invoice data containing multiple billing records, A purchase data reading means for reading purchase data that includes multiple purchase records, wherein each of the purchase records includes a purchase data reading means that includes a transaction code, A program that functions as a transaction identification means: (1) identifying a purchase record that corresponds to any of the multiple invoice records; (2) identifying a transaction code included in the identified purchase record; and (3) referring to the database to identify a trading partner or billing party from the identified transaction code.

[0021]

[15] A method for identifying a customer or billing party from invoice data, The database stores the correspondence between transaction codes and the trading partner or billing address. The invoice reading means includes the step of reading invoice data containing multiple invoice records, wherein each of the invoice records includes one or more predetermined items and a transaction identification code. A purchase data reading means reads purchase data including multiple purchase records, wherein each of the purchase records includes one or more predetermined items, a transaction code, and a transaction identification code. Transaction identification method, (1) Identify one or more purchase records that include one or more specified items that match one or more specified items included in any of the multiple invoice records, (2) For each purchase code that shares a transaction code with one or more identified purchase records, the transaction identification code contained in each purchase record is matched with the invoice record containing the corresponding transaction identification code. Purchase records for which no invoice record containing the corresponding transaction identification code exists are not matched with any invoice records. (3) Identify the transaction code that appears most frequently in the purchase records that are matched with any invoice record, (4) A method comprising the step of referring to the database to identify a customer or billing party from the identified transaction code.

[0022]

[16] A system that identifies a business partner or billing address from invoice data, The database stores the correspondence between transaction codes and the trading partner or billing address. An invoice reading means for reading invoice data containing multiple billing records, wherein each of the billing records includes one or more predetermined items and a transaction identification code, A purchase data reading means for reading purchase data that includes multiple purchase records, wherein each of the purchase records includes one or more predetermined items, a transaction code, and a transaction identification code, (1) Identify one or more purchase records that include one or more specified items that match one or more specified items included in any of the multiple invoice records, (2) For each purchase code that shares a transaction code with one or more identified purchase records, the transaction identification code contained in each purchase record is matched with the invoice record containing the corresponding transaction identification code. Purchase records for which no invoice record containing the corresponding transaction identification code exists are not matched with any invoice records. (3) Identify the transaction code that appears most frequently in the purchase records that are matched with any invoice record, (4) A system comprising a transaction identification means that identifies a trading partner or billing party from the identified transaction code by referring to the database.

[0023]

[17] A program that identifies a customer or billing address from invoice data, The database stores the correspondence between transaction codes and the trading partner or billing address. One or more computers, An invoice reading means for reading invoice data containing multiple billing records, wherein each of the billing records includes one or more predetermined items and a transaction identification code, A purchase data reading means for reading purchase data that includes multiple purchase records, wherein each of the purchase records includes one or more predetermined items, a transaction code, and a transaction identification code, (1) Identify one or more purchase records that include one or more specified items that match one or more specified items included in any of the multiple invoice records, (2) For each purchase code that shares a transaction code with one or more identified purchase records, the transaction identification code contained in each purchase record is matched with the invoice record containing the corresponding transaction identification code. Purchase records for which no invoice record containing the corresponding transaction identification code exists are not matched with any invoice records. (3) Identify the transaction code that appears most frequently in the purchase records that are matched with any invoice record, (4) A program that functions as a means for identifying a trading partner or billing party from the identified transaction code by referring to the database. [Brief explanation of the drawing]

[0024] [Figure 1] A schematic diagram illustrating an example of an invoice. [Figure 2] A schematic diagram illustrating an example of procurement data. [Figure 3] A block diagram showing the schematic configuration of the invoice matching system according to the first embodiment. [Figure 4] A flowchart showing an example of the processing operation of the invoice matching system according to the first embodiment. [Figure 5] A schematic diagram showing the invoice loading screen. [Figure 6] A schematic diagram showing the loading result display screen. [Figure 7A]A schematic diagram showing a portion of the pairing result display screen. [Figure 7B] A schematic diagram showing another part of the pairing results display screen. [Figure 7C] A schematic diagram showing another part of the pairing results display screen. [Figure 7D] A schematic diagram showing another part of the pairing results display screen. [Figure 8A] A schematic diagram showing a portion of the screen used to identify discrepancies in the amount. [Figure 8B] A schematic diagram showing another part of the screen that identifies differences in the amount. [Figure 8C] A schematic diagram showing another part of the screen that identifies differences in the amount. [Figure 8D] A schematic diagram showing a portion of the pairing result display screen. [Figure 8E] A schematic diagram showing another part of the pairing results display screen. [Figure 9A] A schematic diagram showing a portion of the screen used to identify discrepancies in the amount. [Figure 9B] A schematic diagram showing another part of the screen that identifies differences in the amount. [Figure 9C] A schematic diagram showing another part of the screen that identifies differences in the amount. [Figure 9D] A schematic diagram showing a portion of the pairing result display screen. [Figure 9E] A schematic diagram showing another part of the pairing results display screen. [Figure 10A] A schematic diagram showing a portion of the screen for identifying unaccounted amounts. [Figure 10B] A schematic diagram showing a portion of the screen for identifying unaccounted amounts. [Figure 10C] A schematic diagram showing a portion of the pairing result display screen. [Figure 10D] A schematic diagram showing another part of the pairing results display screen. [Figure 11A] A schematic diagram illustrating an example of an invoice. [Figure 11B] A schematic diagram illustrating an example of an invoice. [Figure 12] A schematic diagram illustrating an example of procurement data. [Figure 13] A block diagram showing the schematic configuration of the invoice matching system according to the second embodiment. [Figure 14] A schematic diagram illustrating the structure of the transaction database. [Figure 15] A flowchart showing an example of the processing operation of the invoice matching system according to the second embodiment. [Figure 16A] A schematic diagram illustrating an example of an invoice. [Figure 16B] A schematic diagram illustrating an example of an invoice. [Figure 17] A schematic diagram illustrating an example of procurement data. [Figure 18] A schematic diagram illustrating the structure of the transaction database. [Figure 19] A diagram showing an example of a billing record included in an invoice. [Figure 20] A schematic diagram illustrating an example of procurement data. [Figure 21] A flowchart illustrating the supplier identification method according to the fourth embodiment. [Figure 22] A diagram illustrating the fourth embodiment. [Modes for carrying out the invention]

[0025] The embodiments of the present invention will be described in detail below with reference to the drawings. First, we will describe the problems that the inventors of this application have found regarding butt joint operations.

[0026] (First Embodiment) Figure 1 is a schematic diagram illustrating an example of an invoice. The invoice is sent by the supplier and may be in electronic format (e.g., a PDF file with or without text data), but it is primarily assumed to be a paper invoice.

[0027] An invoice contains multiple invoice records A1 through A6. Each of these invoice records consists of fields such as date, product name, order number, unit price, quantity, and amount. However, since invoices are sent by suppliers, their format varies depending on the supplier. That is, an invoice record may be missing some of these fields, or it may include other fields. Ideally, all fields included in an invoice record should have the correct values ​​set (filled in), but sometimes the value of some fields (such as the order number) may be missing, as in invoice record A3, and in some cases, the value of a field may be incorrect.

[0028] Furthermore, invoices include information on the total amount due. This information usually includes the amount excluding tax, the amount including consumption tax and tax (total amount), but it may also include only the amount excluding tax or the amount including tax. In addition to the total amount due information, invoices may also include information such as the name of the invoicer (supplier), the date of issue, the bank account details, the account holder's name, and the payment due date. This information may be included, for example, on the back of the invoice.

[0029] Figure 2 schematically shows an example of purchase data. This purchase data is managed internally by the company and is in electronic format such as CSV. The purchase data includes multiple purchase records B1 to B6. Each of the purchase records B1 to B6 is associated with items such as customer code, supplier name, date, product name, order number, unit price, quantity, and purchase amount. However, a purchase record may be missing some of these items or may include other items. Since the purchase data is managed internally by the company, it is in a specific format.

[0030] For the sake of simplicity, we will assume that the supplier name (customer code) in the purchase record is the same for all. If multiple supplier names (customer codes) are present, the process will need to be performed separately for each supplier name. This point will be discussed in the second embodiment.

[0031] Purchase data and invoices are supposed to correspond. Therefore, each of the invoice records A1-A6 should correspond to (be paired with) one of the purchase records B1-B6. And the invoice amount in the invoice record should match the purchase amount in the corresponding purchase record. However, for some reason, the amounts may not match, and there may be invoice records or purchase records that do not form a pair.

[0032] For example, the invoice amount (600 yen) in invoice record A4 in Figure 1 does not match the purchase amount (750 yen) in the corresponding purchase record B4 in Figure 2. The purchase record corresponding to invoice record A6 in Figure 1 does not exist in the purchase data in Figure 2. The invoice record corresponding to purchase record B6 in Figure 2 does not exist in the invoice in Figure 1.

[0033] The matching process involves detecting discrepancies in amounts or missing invoice or purchase records and identifying the cause. When there are a large number of invoice records or purchase records in the purchase data, the matching process becomes extremely burdensome. To alleviate this workload, the inventors of this application invented an invoice matching system. A specific example is described below.

[0034] Figure 3 is a block diagram illustrating the schematic configuration of the invoice matching system according to the first embodiment. The invoice matching system comprises an invoice reading means 1, a purchase data reading means 2, a pairing means 3, a purchase total amount calculation means 4, an amount difference identification means 5, an unrecorded amount identification means 6, a difference calculation means 7, and a display control means 8.

[0035] These may consist of a single device (computer) or be distributed across two or more devices (computers). For example, a computer may be configured to function as each of the above means by installing a predetermined program on it. As another example, the invoice matching system may consist of a network-connected server and terminals, with user-involved functions implemented on the terminals (e.g., via a web browser) and other functions implemented on the server.

[0036] The invoice reading means 1 reads invoice data (electronic invoices, or digitized paper invoices) and links the invoice amount to each of the multiple invoice records. Typically, the invoice data is an electronic file (PDF file) obtained by scanning a paper invoice. The invoice reading means 1 then digitizes the read invoice data using OCR functionality and uses pre-trained artificial intelligence to link each invoice record to the invoice amount contained in that record. Since invoices differ depending on the supplier, it is desirable that the AI ​​be trained for each supplier. The invoice reading means 1 also reads the invoice data and identifies the total invoice amount.

[0037] In the example invoice shown in Figure 1, the invoice amount of "1,000 yen" is associated with invoice record A1, and the invoice amount of "1,500 yen" is associated with invoice record A2. Furthermore, by reading the mirror of the invoice, or by summing the invoice amounts associated with each invoice record, the total invoice amount of "5,800 yen" is determined.

[0038] However, the linking here may be incomplete. For example, errors in OCR or artificial intelligence may result in incorrect billing amounts being linked to billing records, or billing amounts not being linked at all. The total billing amount may also be incorrect or unavoidable.

[0039] Furthermore, the invoice reading means 1 may read the invoice data and identify the supplier (in the example in Figure 1, xx Trading Co., Ltd.). The invoice reading means 1 may have previously associated and stored the supplier and customer code, and may further identify the customer code from the identified supplier name.

[0040] The invoice reading means 1 may associate some or all of the values ​​entered in fields such as date, product name, order number, unit price, quantity, and amount with each of the multiple invoice records. The associated values ​​can be used in the pairing process described later.

[0041] The purchase data reading means 2 reads the purchase data. As illustrated in Figure 2, the purchase data includes multiple purchase records, each of which is associated with a customer code, supplier name, date, product name, order number, unit price, quantity, purchase amount, etc.

[0042] Pairing method 3 pairs each invoice record with each purchase record by identifying the purchase record that corresponds to each invoice record. Pairing may be performed using artificial intelligence, for example. Since invoices differ depending on the supplier, it is desirable that the AI ​​learns for each supplier. Pairing may also be performed by referencing the fields associated with the invoice record and the fields included in the purchase record. Since the fields associated with the invoice record differ depending on the supplier, it is desirable that the fields referenced be set for each supplier.

[0043] As a result of pairing, most invoice records are paired with one of the purchase records, and the invoice amount in the invoice record matches the purchase amount in the purchase record. Such pairs of invoice and purchase records are called "amount matching pairs." For example, invoice record A1 in Figure 1 and purchase record B1 in Figure 2 are amount matching pairs.

[0044] Note that a matching amount pair may be one invoice record paired with two or more purchase records, where the total invoice amount associated with the invoice record matches the total purchase amount associated with the purchase record. Conversely, a matching amount pair may be two or more invoice records paired with one purchase record, where the total invoice amount associated with the invoice record matches the total purchase amount associated with the purchase record.

[0045] On the other hand, while some invoice records are paired with purchase records, the invoice amount in the invoice record and the purchase amount in the purchase record may not match. Such pairs of invoice and purchase records are called "amount mismatch pairs." For example, invoice record A4 in Figure 1 and purchase record B4 in Figure 2 are amount mismatch pairs.

[0046] Furthermore, a mismatched amount pair may occur when one invoice record is paired with two or more purchase records, and the total invoice amount associated with the invoice record does not match the total purchase amount associated with the purchase records. Conversely, a mismatched amount pair may occur when one purchase record is paired with two or more invoice records, and the total invoice amount associated with the invoice record does not match the total purchase amount associated with the purchase records.

[0047] Possible causes for such discrepancies in amounts include errors in the invoice, errors in the OCR function or artificial intelligence during invoice scanning, etc.

[0048] Furthermore, for various reasons (primarily due to not being accounted for), some invoice records may not be able to be paired with any purchase records (for example, invoice record A6 in Figure 1). Additionally, for various reasons (primarily due to not being invoiced), some purchase records may not be able to be paired with any invoice records (for example, purchase record B6 in Figure 2).

[0049] Pairing method 3 can be said to classify multiple invoice records included in the invoice data into three categories: invoice records whose amounts match the paired purchase records (first invoice records), invoice records whose amounts do not match the paired purchase records (second invoice records), and invoice records that do not pair with any purchase records (third invoice records). Furthermore, pairing method 3 can be said to identify invoice records that are not paired with any purchase records as unaccounted records. In addition, pairing method 3 can be said to identify purchase records that are not paired with any invoice records as uninvoiced records.

[0050] The purchase total amount calculation means 4 calculates the sum of the purchase amounts associated with the purchase records paired with any of the invoice records (hereinafter referred to as "purchase total amount").

[0051] The method for identifying discrepancies in amounts 5 identifies the difference in amounts between the invoice amount associated with the invoice record and the purchase amount associated with the paired purchase record for the invoice record in the invoice record in the discrepancy in amounts pair.

[0052] For example, the method for identifying discrepancies in amounts 5 may calculate the discrepancy in amounts from the difference between the invoice amount linked to the invoice record and the purchase amount linked to the paired purchase record.

[0053] Alternatively, the amount difference identification means 5 may identify the difference by accepting the amount difference setting from the user through manual setting. More specifically, considering that the invoice reading means 1 may make errors in reading the invoice amount, the user visually reads the amount (invoice amount) written on the invoice and manually sets that amount. The amount difference identification means 5 may then calculate the amount difference from the difference between the manually set invoice amount and the purchase amount.

[0054] If there are multiple pairs with mismatched amounts, the amount discrepancy identification method 5 calculates the total amount of the discrepancies. If there are no errors in the invoice and its reading, there are no discrepancies, and the total amount will be 0 yen.

[0055] The unrecorded amount identification method 6 identifies the unrecorded amount for invoice records that are not paired with any purchase records.

[0056] For example, the unrecorded amount identification method 6 may identify the unrecorded amount from the invoice amount linked to the invoice record.

[0057] Alternatively, the unaccounted amount identification means 6 may be identified by accepting the setting of the unaccounted amount from the user through manual setting. More specifically, considering that the invoice reading means 1 may make errors in reading the invoice amount, the user visually reads the amount (invoice amount) written on the invoice and manually sets that amount.

[0058] If there are multiple invoice records that do not match any purchase records, the unaccounted amount identification method 6 calculates the total amount of unaccounted amounts. If there are no unaccounted invoice records, the total amount of unaccounted amounts will be 0 yen.

[0059] The difference calculation method 7 calculates the difference between the sum of the total purchase amount, the sum of the amount differences, and the sum of the unrecorded amounts, and the total invoiced amount. If there are no errors in the invoice and its reading, and there are no unrecorded invoice records, the sum of the amount differences and the sum of the unrecorded amounts will both be 0 yen, so the difference between the total purchase amount and the total invoiced amount will be 0 yen. Otherwise, by identifying all amount differences and calculating the sum of the amount differences, and by identifying all unrecorded invoice records and calculating the sum of the unrecorded amounts, the above difference will also be 0 yen (this point will be explained later with an example).

[0060] The display control means 8 displays various screens on a display (not shown) so that they can be viewed by the user.

[0061] Figure 4 is a flowchart showing an example of the processing operation of the invoice matching system according to the first embodiment. The process will be explained in detail below, along with examples of screens displayed on the display. Note that Figure 4 is not intended to limit the processing operation; the order of processing may be rearranged, some processing may be omitted, or additional processing may be performed.

[0062] To load a new invoice, the display control means 8 displays an invoice loading screen on the display as shown in Figure 5. The user then specifies the invoice data by selecting the "Select File" button or by dragging and dropping a file. As a result, the invoice loading means 1 loads the invoice data (step S1 in Figure 4).

[0063] The display control means 8 then displays a reading result display screen on the display as shown in Figure 6. The reading result display screen includes the invoice and the customer name and total invoice amount identified by the invoice reading means 1 using OCR function and artificial intelligence, etc. In addition, if customer names and customer codes are pre-associated in the invoice matching system, the screen may also include the customer code associated with the identified customer name.

[0064] The total billing amount may be incorrectly identified, or the invoice reading device 1 may not be able to identify it (in which case the total billing amount field will be blank). Therefore, the invoice reading device 1 may accept the correct value of the total billing amount from the user. The same applies to the customer name and customer code.

[0065] Next, in order to load new purchase data, the display control means 8 displays a purchase data loading screen on the display, similar to that shown in Figure 5. The user then specifies the purchase data. As a result, the purchase data loading means 2 loads the purchase data (step S2 in Figure 4).

[0066] Then, the pairing means 3 pairs the invoice records in the invoice data with the purchase records in the purchase data (step S3). Then, the total purchase amount calculation means 4 calculates the total purchase amount, which is the sum of the purchase amounts associated with the purchase records that have been paired with any of the invoice records (step S4). The display control means 8 displays a pairing result display screen on the display showing the pairing results.

[0067] Figure 7A schematically shows a part of the pairing result display screen. The base of this screen is the invoice read in step S1 (the one shown in Figure 1 is an example). The display control means 8 then displays a mark (classification information) indicating the pairing result, associated with each of the invoice records A1 to A6. In the example in Figure 7A, ◎ means that the amount matches, △ means that the amount does not match, and no mark means that it was not paired with any purchase record (a mark may be placed instead of no mark).

[0068] For example, invoice records A1 to A3 are a matching amount pair, invoice records A4 and A5 are a mismatched amount pair, and invoice record A6 is not paired with any purchase record. In this way, invoice records with mismatched amounts or those that are not paired can be easily identified from the screen shown in Figure 7A.

[0069] Figure 7B schematically shows another part of the pairing result display screen. The total invoice amount is shown in Figure 6 and is the amount read by the invoice reading means 1 (or the amount set by the user afterwards). The total purchase amount is calculated by the total purchase amount calculation means 4 in step S4 of Figure 4. The total difference in amounts and the total unrecorded amount are 0 yen at the time only pairing is performed (i.e., before the identification of the difference in amounts and the identification of the unrecorded amount described later). The difference is calculated by the difference calculation means 7 as follows: Total invoice amount - (Total purchase amount + Total difference in amounts + Total unrecorded amount) = 5,800 - (5,250 + 0 + 0) = 550 yen.

[0070] If the pairing results in a difference of 0 yen, it means that the total invoice amount and the total purchase amount match, and there are no problems with the invoice. In this way, the user can complete the matching process simply by importing the invoice data and purchase data.

[0071] On the other hand, if the difference is not 0 yen after pairing, it means there is some kind of problem (for example, an error in the invoice amount, an error in OCR or artificial intelligence, unbilled, or unaccounted). Even in that case, the user can easily find out the reason why the difference is not 0 yen by following the steps below.

[0072] Figure 7C schematically shows another part of the pairing results display screen. This figure shows a list of unbilled purchase records that were not paired with any invoice records as a result of the pairing process. In the example shown in this figure, purchase record B6 is marked as unbilled among the purchase data exemplified in Figure 2. Thus, unbilled (i.e., not billed in the corresponding invoice) purchase records can be easily identified from the screen in Figure 7C.

[0073] Figure 7D schematically shows another part of the pairing results screen. The base of this screen is the purchase data read in step S2 (as shown in Figure 2). The display control means 8 then displays a mark (classification information) indicating the pairing result, associated with each of the purchase records B1 to B6. The meaning of each mark is the same as in Figure 7A.

[0074] For example, purchase records B1 to B3 are a matching amount pair, purchase records B4 and B5 are a mismatched amount pair, and purchase record B6 is not paired with any invoice record. In this way, purchase records with mismatched amounts or those that are not paired can be easily identified from the screen shown in Figure 7D.

[0075] As shown in Figures 7C and 7D, purchase records B1 to B5 are paired with one of the invoice records A1 to A6 in the invoice data exemplified in Figure 1. Therefore, the total purchase amount shown in Figure 7B is 5,250 yen, which is the sum of the purchase amounts linked to purchase records B1 to B5.

[0076] The screens shown in Figures 7A to 7D may be displayed on a single screen, or they may be displayed on multiple separate screens.

[0077] Next, the user identifies any discrepancies in amounts and any unaccounted amounts. For example, in Figure 7A, the user selects billing record A4 as a mismatched amount pair. In response, the display control means 8 displays a screen identifying the mismatched amount on the display.

[0078] Figure 8A schematically shows a part of the screen for identifying discrepancies in amounts. In this screen, the display of the △ mark (e.g., color) changes from Figure 7A, and the details of purchase record B4, which is paired with invoice record A4, are displayed overlaid.

[0079] Figure 8B schematically shows another part of the screen for identifying discrepancies in amounts. This screen is displayed on the display by the display control means 8 based on the selection of the △ mark associated with billing record A4.

[0080] In addition to the details of purchase record B4 paired with invoice record A4, a "billing amount" field is provided. The purchase amount in purchase record B4 is 750 yen. However, according to the invoice (Figure 1, Figure 8A), the billing amount in the corresponding invoice record A4 is 600 yen. Therefore, the user visually reads this "600 yen" and sets it in the "billing amount" field (Figure 8C). Accordingly, the amount difference identification means 5 identifies -150 yen as the difference between the billing amount and the purchase amount (step S5 in Figure 4).

[0081] Based on the above, the display control means 8 updates the screen shown in Figure 7A as shown in Figure 8D. Specifically, it sets the mark associated with billing record A4 to ○, indicating that the amount mismatch was manually set, and adds the difference of -150 yen.

[0082] Furthermore, the difference calculation means 7 calculates the total invoice amount - (total purchase amount + total difference in amount + total unrecorded amount) = 5,800 - (5,250 + (-150) + 0) = 700 yen (step S7). In this way, the difference calculation means 7 updates the difference each time a difference in amount for a single invoice record is identified. Then, the display control means 8 updates the screen shown in Figure 7B as shown in Figure 8E. That is, the total difference in amount is set to -150 yen, and the difference is set to 700 yen.

[0083] Next, as can be seen from the screen in Figure 8D, invoice record A5 also has an amount mismatch. Therefore, the user selects invoice record A5 as an amount mismatch pair in Figure 8D. Accordingly, the display control means 8 displays the amount mismatch identification screen on the display.

[0084] Figure 9A schematically shows a part of the screen for identifying discrepancies in amounts. In this screen, the display method (e.g., color) of the △ mark changes from that in Figure 8D, and the details of purchase record B5, which is paired with invoice record A5, are displayed overlaid.

[0085] Figure 9B schematically shows another part of the screen for identifying discrepancies in amounts. This screen is displayed on the display by the display control means 8 based on the selection of the △ mark associated with billing record A5.

[0086] In addition to the details of purchase record B5 paired with invoice record A5, a "billing amount" field is provided. The purchase amount in purchase record B5 is 800 yen. However, according to the invoice (Figure 1, Figure 9A), the billing amount in the corresponding invoice record A5 is 1,000 yen. Therefore, the user visually reads this "1,000 yen" and sets it in the "billing amount" field (Figure 9C). Accordingly, the amount difference identification means 5 identifies 200 yen as the difference between the billing amount and the purchase amount (step S5 in Figure 4).

[0087] Furthermore, the method for identifying differences in amounts 5 adds up the -150 yen identified in Figure 8C and the 200 yen identified in Figure 9C as the total of the differences in amounts, resulting in a total value of 50 yen.

[0088] Based on the above, the display control means 8 updates the screen shown in Figure 8D as shown in Figure 9D. Specifically, it sets the mark associated with billing record A5 to ○, indicating that the amount mismatch was manually set, and adds the difference of 200 yen.

[0089] Furthermore, the difference calculation means 7 calculates the total invoice amount - (total purchase amount + total difference in amount + total unrecorded amount) = 5,800 - (5,250 + 50) + 0) = 500 yen (step S7). Then, the display control means 8 updates the screen shown in Figure 8E as shown in Figure 9E. That is, the total difference in amount is set to 50 yen, and the difference is set to 500 yen.

[0090] Next, as can be seen from the screen in Figure 9D, invoice record A6 is an unaccounted record that is not paired with any purchase records. Therefore, the user selects invoice record A6 as an unaccounted record on the screen in Figure 9D. In response, the display control means 8 displays the unaccounted amount identification screen on the display.

[0091] Figure 10A schematically shows a part of the screen for identifying unaccounted amounts. This screen is displayed on the display by the display control means 8 based on the selection of invoice record A6. According to the invoice (Figures 1 and 9D), the invoice amount for invoice record A6 is 500 yen. The user then visually reads this "500 yen" and sets it in the "Invoice Amount" field (Figure 10B). Accordingly, the unaccounted amount identification means 6 identifies 500 yen as the unaccounted invoice amount (step S6 in Figure 4).

[0092] Furthermore, the method for identifying unrecorded amounts 6 calculates the total amount of unrecorded amounts if there are other unrecorded invoice amounts (in this example, there are no other unrecorded invoice amounts).

[0093] Based on the above, the display control means 8 updates the screen shown in Figure 9D as shown in Figure 10C. Specifically, it sets the mark associated with invoice record A6 to ■ (classification information), which means that it is not yet accounted for, and adds the unaccounted amount of 500 yen.

[0094] Furthermore, the difference calculation means 7 calculates the total invoice amount - (total purchase amount + total difference in amount + total unrecorded amount) = 5,800 - (5,250 + 50) + 500) = 0 yen (step S7). In this way, the difference calculation means 7 updates the difference each time an unrecorded amount for a single invoice record is identified. Then, the display control means 8 updates the screen shown in Figure 9E as shown in Figure 10D. That is, the total unrecorded amount is set to 500 yen, and the difference is set to 0 yen.

[0095] As described above, when the difference becomes 0 yen, it means that all the reasons why the total invoice amount and the total purchase amount do not match have been identified.

[0096] As described above, this embodiment automatically pairs invoice records in invoice data with purchase records in purchase data. It then identifies invoice records with mismatched amounts, unrecorded invoice records, and uninvoiced purchase records, and displays each of them clearly on the display. Therefore, even if the total invoice amount and the total purchase amount do not match (the difference is not 0 yen), the user can quickly understand the cause, and the matching process is simplified.

[0097] (Second Embodiment) The first embodiment described above assumed that there was only one supplier in the purchase data (Figure 2). However, in reality, there may be multiple suppliers (which can also be called trading partners or billing parties), and the purchase data may include multiple suppliers. One example of a case where there are multiple suppliers is the need to distinguish between the head office and each branch office. Therefore, the second embodiment assumes a case where there are multiple suppliers (particularly a case where it is difficult to identify the supplier from the invoice). Note that in the following, the points common to the first embodiment will be omitted or simplified in the explanation.

[0098] Figures 11A and 11B schematically illustrate examples of invoices. In Figure 11A, the supplier is the Tokyo branch of xx Trading Co., Ltd., and in Figure 11B, the supplier is the Fukuoka branch of xx Trading Co., Ltd., and the two are distinguishable. The supplier does not necessarily have to be explicitly stated on the cover of the invoice. However, the order number in each invoice record is unique (i.e., an order number included in one invoice will not be included in other invoices). The invoice in Figure 11A includes invoice records C1 to C3, and the invoice in Figure 11B includes invoice records D1 to D3.

[0099] Figure 12 schematically shows an example of purchase data. This purchase data includes purchase records E1 to E6, each containing a customer code. Unlike Figure 2, the purchase data includes multiple customer codes (here, "3333" and "5555"). Each customer code corresponds to one supplier (business partner) and can also be called a transaction code. As shown in Figure 14 later, customer code "3333" corresponds to the Tokyo branch of xx Trading Co., Ltd., and customer code "5555" corresponds to the Fukuoka branch of xx Trading Co., Ltd. In this way, customer codes are assigned to each branch of the supplier.

[0100] Figure 13 is a block diagram illustrating the schematic configuration of an invoice matching system according to the second embodiment. The invoice matching system in Figure 13 further comprises a storage means 11 and a transaction identification unit 12. The storage means 11 stores a transaction database as shown in Figure 14. The transaction database shows the correspondence between supplier (business partner) names and customer codes (transaction codes). The transaction identification unit 12 identifies the supplier in the invoice. Specifically, the transaction identification means 12 identifies the supplier based on the billing record in the invoice and the purchase record in the purchase data. Considering the identified supplier, the pairing means 3 performs pairing.

[0101] Figure 15 is a flowchart showing an example of the processing operation of the invoice matching system according to the second embodiment.

[0102] Similar to the first embodiment, the invoice reading means 1 reads the invoice data (step S1). Here, we assume that the invoice data shown in Figure 11B has been read. However, the supplier (Fukuoka branch of xx Trading Co., Ltd. in Figure 11B) does not have to be included in the cover sheet, or even if it is included, it does not have to be read correctly.

[0103] The purchase data reading means 2 reads the purchase data (Figure 12) (step S2). Then, the transaction identification means 12 identifies the supplier (step S11). As an identification method, if it can be read from the cover sheet of the invoice data, that can be used, but another method will be described.

[0104] The transaction identification means 12 identifies a purchase record that corresponds to one of the invoice records. More specifically, the transaction identification means 12 identifies a purchase record that contains an order number that corresponds to (matches) an order number contained in one of the invoice records. Then, the transaction identification means 12 identifies the customer code (transaction code) contained in the identified purchase record. Furthermore, the transaction identification means 12 may identify the supplier by referring to the transaction database.

[0105] As a specific example, the transaction identification means 12 identifies the purchase record E1 in Figure 12 as a purchase record containing an order number that matches the order number "3001" contained in the invoice record D1 in Figure 11B. The transaction identification means 12 then identifies the customer code (transaction code) "5555" contained in the identified purchase record E1. Furthermore, the transaction identification means 12 refers to the transaction database (Figure 14) and identifies that the supplier is the Fukuoka branch of xx Trading Co., Ltd.

[0106] Next, the pairing means 3 pairs the invoice record in the invoice data with the purchase record in the purchase data (step S3). In this embodiment, the pairing is performed with the purchase record that includes the customer code identified in step S11.

[0107] In the example above, each of the invoice records D1 to D3 shown in Figure 11B is paired with each of the purchase records E1, E4, and E5 from the multiple purchase records shown in Figure 12 that contain customer code "5555". In other words, purchase records E2, E3, and E6 are not paired with invoice records D1 to D3.

[0108] As described above, pairing can be performed even when there are multiple suppliers. The subsequent processing operations are the same as in the first embodiment.

[0109] (Third embodiment) The first embodiment described above assumed that there was only one supplier in the purchase data (Figure 2). The second embodiment described above assumed that there were multiple suppliers. In contrast, there may be multiple billing parties (which can also be called trading partners or suppliers), and the purchase data may also include multiple billing parties. One example of a case where there are multiple billing parties is the need to distinguish between the head office and each branch office. Therefore, the third embodiment assumes a case where there are multiple billing parties (particularly a case where it is difficult to identify the billing party from the invoice). Note that in the following, the common points with the first and second embodiments will be omitted or simplified in the explanation.

[0110] Figures 16A and 16B schematically show examples of invoices. In Figure 16A, the billing address is the Osaka branch of xxx Co., Ltd., and in Figure 16B, the billing address is the Nagoya branch of xxx Co., Ltd., and the two are distinguishable. The billing address does not need to be explicitly stated on the cover of the invoice. However, the order number in each invoice record is unique (i.e., an order number included in one invoice will not be included in other invoices). The invoice in Figure 16A includes invoice records F1 to F3, and the invoice in Figure 16B includes invoice records G1 to G3.

[0111] Figure 17 schematically shows an example of purchase data. This purchase data includes purchase records H1 to H6, each containing a billing code. Each billing code corresponds to one billing party (transaction source) and can also be called a transaction code. As shown in Figure 18, which will be described later, billing code "7777" corresponds to the Osaka branch of xxx Co., Ltd., and billing code "9999" corresponds to the Nagoya branch of xxx Co., Ltd. In this way, billing codes are assigned to each branch of the billing party.

[0112] The schematic configuration of the invoice matching system according to the third embodiment is the same as that shown in Figure 13. However, the storage means 11 stores a transaction database as shown in Figure 18. The transaction database shows the correspondence between the name of the billing party (transaction source) and the billing party code (transaction code). The transaction identification unit 12 identifies the billing party of the invoice. In detail, the transaction identification means 12 identifies the billing party based on the billing record in the invoice and the purchase record in the purchase data. Taking the identified billing party into consideration, the pairing means 3 performs pairing.

[0113] The processing operation of the invoice matching system according to the third embodiment may be the same as that shown in Figure 15. That is, as in the first embodiment, the invoice reading means 1 reads the invoice data (step S1). Here, let's assume that the invoice data shown in Figure 16B has been read. However, the billing address (Nagoya branch of xxx Corporation in Figure 16B) does not have to be included in the cover sheet, or even if it is included, it does not have to be read correctly.

[0114] The purchase data reading means 2 reads the purchase data (Figure 17) (step S2). Then, the transaction identification means 12 identifies the supplier (step S11). As an identification method, if it can be read from the cover sheet of the invoice data, that can be used, but another method will be described.

[0115] The transaction identification means 12 identifies a purchase record that corresponds to one of the invoice records. More specifically, the transaction identification means 12 identifies a purchase record that contains an order number that corresponds to (matches) an order number contained in one of the invoice records. Then, the transaction identification means 12 identifies the customer code (transaction code) contained in the identified purchase record. Furthermore, the transaction identification means 12 may identify the customer by referring to the transaction database.

[0116] As a specific example, the transaction identification means 12 identifies the purchase record H1 in Figure 17 as a purchase record containing an order number that matches the order number "5001" contained in the invoice record G1 in Figure 16B. Then, the transaction identification means 12 identifies the customer code (transaction code) "9999" contained in the identified purchase record H1. Furthermore, the transaction identification means 12 refers to the transaction database (Figure 18) and identifies that the customer is the Nagoya branch of xxx Co., Ltd.

[0117] Next, the pairing means 3 pairs the invoice record in the invoice data with the purchase record in the purchase data (step S3). In this embodiment, the pairing is performed with the purchase record that includes the billing address code identified in step S11.

[0118] In the example above, each of the invoice records G1 to G3 shown in Figure 16B is paired with each of the purchase records H1, H4, and H5 from the multiple purchase records shown in Figure 17 that contain the customer code "9999". In other words, purchase records H2, H3, and H6 are not paired with invoice records G1 to G3.

[0119] As described above, pairing can be performed even when there are multiple billing recipients. The subsequent processing operations are the same as in the first embodiment.

[0120] Furthermore, the second and third embodiments may be combined.

[0121] (Fourth Embodiment) The second and third embodiments described above primarily assumed the use of order numbers from invoices. However, in some cases, invoice numbers may not be accurately obtained by OCR, or they may not perfectly match the order numbers in purchase records. The fourth embodiment, described below, accurately identifies suppliers even in such cases.

[0122] Figure 19 shows an example of an invoice record included in an invoice. As shown in the figure, this invoice is assumed to contain invoice records H1 to H4. Each invoice record includes, as an example of fields, date, product name, order number, unit price, quantity, and amount. Below, we will focus on the unit price, quantity, and amount combination, which are fields that are highly likely to be accurately read by OCR. On the other hand, the order number may not always be accurately read by OCR.

[0123] Figure 20 is a schematic diagram illustrating an example of purchase data. As shown in the diagram, this purchase data includes purchase records I1 to I6. Each purchase record includes, as an example of items, customer code, supplier name, date, product name, order number, unit price, quantity, and amount. Below, we will focus on the unit price, quantity, and amount combination, which is the same as the items in the invoice record. Note that the order number included in the purchase record may be an exact match with the order number included in the invoice record, or it may be a partial match (for example, the order number in the purchase record may be part of the order number in the invoice record).

[0124] The schematic configuration of the invoice matching system according to the fourth embodiment is the same as that shown in Figure 13. Figure 21 is a flowchart illustrating the supplier identification method according to the fourth embodiment. The invoice reading (step S31) and the purchase data reading (step S32) are as described above.

[0125] Next, the transaction identification means 12 identifies a purchase record containing a unit price, quantity, and amount that matches the unit price, quantity, and amount contained in any of the multiple invoice records included in the invoice data (step S33). For example, the unit price, quantity, and amount in invoice record H1 in Figure 19 are 100 yen, 1 unit, and 100 yen, respectively. The purchase records I1 and I2 in Figure 20 match this. That is, in step 33, purchase records I1 and I2 are identified.

[0126] Next, the transaction identification means 12 extracts purchase records that share the same customer code as the identified purchase record (step S34). For example, if purchase records I1 and I2 are identified in step S33, purchase records I3 and I5, which share the same customer code (100) as purchase record I1, and purchase record I4, which shares the same customer code (200) as purchase record I2, are extracted. In other words, purchase records I1 to I5 are extracted (see Figure 22(b)).

[0127] Next, the transaction identification means 12 matches each extracted purchase record with the order number contained in each purchase record and the corresponding invoice record containing the corresponding order number (step S35). For example, as shown in Figure 21, for the extracted purchase record I1, the invoice record H1 containing the order number (111) contained in it is matched with the corresponding order number (5111).

[0128] Note that while an exact match of the order number is acceptable, a match will still occur even if it is not an exact match. For example, if a part of the order number in the invoice record (e.g., 5111) matches the order number in the purchase record (e.g., 111), then both records may be matched. The reason why a part of the order number in the invoice record may match the order number in the purchase record is due to system constraints at the supplier, such as the wholesaler's order number being appended to the supplier's management code in the invoice record, or due to input errors in the system.

[0129] The above matching process is performed on purchase records I1 to I5 extracted in step S34. As a result, the matching is performed as follows (see Figure 22). Note that the order number in purchase record I2 is 211, and there is no invoice record containing the corresponding order number, so it is not matched. In other words, the matching is performed on purchase records I1, I3 to I5. Purchase record, Invoice record I1 H1 I3 H2 I4 H4 I5 H3

[0130] Next, the transaction identification means 12 identifies the customer code that appears most frequently among the customer codes (transaction codes) included in the purchase records that have been matched with any of the invoice records (step S36). In Figures 21(a) and (b), the purchase records I1, I3 to I5 have been matched with any of the invoice records. These purchase records contain three instances of customer code 100 (purchase records I1, I3, I5) and one instance of customer code 200 (purchase record I4) (see Figure 21(c)). Therefore, the most frequently appearing customer code is 100.

[0131] The transaction identification means 12 then refers to a transaction database (see Figure 14) that shows the correspondence between supplier (business partner) names and customer codes (transaction codes), and identifies the supplier (business partner) from the identified customer code (step S37). During matching, in step S3 of Figure 15, pairing is performed with the purchase record containing the identified customer code. The process thereafter is the same as in the second embodiment.

[0132] As described above, even if the order number cannot be read, the supplier can be identified from the items included in the invoice record (e.g., unit price, quantity, amount). Note that in Figure 20, similar to Figure 12, the purchase data includes a customer code. In contrast, if the purchase data includes a billing code, similar to Figure 17, the billing party can be identified from the items included in the invoice record, and the same processing as in the third embodiment can be performed.

[0133] In the embodiment described above, an order number was used, but any transaction identification code may be used. Examples of transaction identification codes include order numbers or purchase order numbers assigned by the billing party, and invoice numbers or delivery note numbers assigned by the billing party.

[0134] Any part or all of the functional components described herein may be implemented by program. Programs referred to herein may be recorded non-temporarily on a computer-readable recording medium.

[0135] Such programs may be installed on a computer (so-called native apps). In that case, the program may be downloaded to the computer via a communication line such as the internet (including wireless communication), or it may be distributed already installed on the computer.

[0136] Alternatively, the program may run on a web browser (a so-called web application). In that case, the computer may receive the program, written in a markup language file (e.g., an HTML file), from a server and execute it through the web browser.

[0137] Based on the above description, those skilled in the art may be able to conceive of additional effects and various modifications of the present invention, but the embodiments of the present invention are not limited to the individual embodiments described above. For example, inventions that take only a part of each embodiment, or inventions that combine multiple embodiments, are naturally conceivable. Various additions, modifications, and partial deletions are possible as long as they do not depart from the conceptual idea and spirit of the present invention derived from the contents of the claims and their equivalents.

[0138] For example, what is described as a single device in this specification (including what is depicted as a single device in the drawings) may be implemented by multiple devices. Conversely, what is described as multiple devices in this specification (including what is depicted as multiple devices in the drawings) may be implemented by a single device. Alternatively, some or all of the means or functions included in one device (e.g., a server) may be included in another device (e.g., a user terminal). Furthermore, a "system" may consist of a single device or of two or more devices (e.g., a server and a user terminal, or multiple user terminals).

[0139] Furthermore, not all matters described herein are mandatory requirements. In particular, matters described herein but not included in the claims can be considered optional additional matters.

[0140] Furthermore, unless otherwise specified, the term “means” in this specification and the claims means hardware (or functions realized by hardware) and does not include human beings (or human mental activity).

[0141] Furthermore, the functions realized by the components described herein may be implemented in a circuitry or processing circuitry, including general-purpose processors, application-specific processors, integrated circuits, ASICs (Application Specific Integrated Circuits), CPUs (a Central Processing Unit), conventional circuits, and / or combinations thereof, programmed to realize the functions described herein. A processor includes transistors and other circuits and is considered a circuitry or processing circuitry. A processor may be a programmed processor that executes a program stored in memory.

[0142] Furthermore, in this specification, circuitry, unit, and means are hardware programmed to perform or execute the functions described herein. Such hardware may be any hardware disclosed herein, or any hardware known to be programmed to perform or execute the functions described herein.

[0143] If the hardware is a processor that is considered to be a type of circuitry, then the circuitry, means, or unit is a combination of hardware and software used to constitute the hardware and / or processor.

[0144] It should also be noted that the applicant is only aware of the prior art inventions described in the "Prior Art Documents" section of this specification, and the present invention is not necessarily intended to solve the problems described in those prior art inventions. The problems that the present invention aims to solve should be determined by considering this specification as a whole. For example, if this specification describes that a certain effect is achieved by a particular configuration, it can also be said that the problem that is the inverse of that predetermined effect is solved. However, this does not necessarily mean that such a particular configuration is an essential requirement. [Explanation of Symbols]

[0145] 1. Invoice reading method 2. Purchase data reading means 3. Pairing method 4. Method for calculating the total purchase amount 5. Methods for identifying differences in amount 6. Methods for identifying unrecorded amounts 7. Method for calculating the difference 8 Display control means 11 Transaction Identification Methods 12 Memory means

Claims

1. The invoice reading means reads invoice data including multiple invoice records and information on the total invoice amount, associates the invoice amount with each of the multiple invoice records, and identifies the total invoice amount. The purchase data reading means includes the step of reading purchase data which includes multiple purchase records, each of which is associated with a purchase amount, and each of the purchase records includes a transaction code, The transaction identification means includes the steps of identifying the purchase record that corresponds to one of the plurality of invoice records, and identifying the transaction code included in the identified purchase record, The pairing means pairs each of the plurality of invoice records with each of the plurality of purchase records that includes the identified transaction code, and each of the invoice records, The first invoice record can be paired with any purchase record, and the invoice amount associated with that invoice record matches the purchase amount associated with the paired purchase record. A second invoice record was paired with one of the purchase records, but the invoice amount associated with that invoice record did not match the purchase amount associated with the paired purchase record. The third invoice record, which is not paired with any purchase record, is classified as follows: The method for calculating the total purchase amount includes the step of calculating the total purchase amount, which is the sum of the purchase amounts associated with purchase records paired with any invoice record, The means for identifying differences in amounts includes the steps of identifying the difference in amounts between the invoice amount associated with the second invoice record and the purchase amount associated with the paired purchase record, and calculating the total amount of the differences in amounts, The means for identifying unrecorded amounts includes the steps of identifying the invoice amount associated with the third invoice record as an unrecorded amount and calculating the total amount of the unrecorded amounts, An invoice matching method that includes the step of calculating the difference between (1) the sum of the total purchase amount, the sum of the amount differences, and the sum of the unrecorded amounts, and (2) the total invoice amount.

2. Each of the aforementioned billing records includes a transaction identification code, Each of the aforementioned purchase records includes a transaction identification code, The invoice matching method according to claim 1, wherein the transaction identification means identifies the purchase record that includes a transaction identification code corresponding to a transaction identification code included in any of the plurality of invoice records.

3. The invoice reading means reads invoice data including multiple invoice records and information on the total invoice amount, associates the invoice amount with each of the multiple invoice records, and identifies the total invoice amount, wherein each of the invoice records includes one or more predetermined items and a transaction identification code. A purchase data reading means reads purchase data, which includes multiple purchase records, each of which is associated with a purchase amount, and each of the purchase records includes one or more predetermined items, a transaction code, and a transaction identification code. Transaction identification method, (1) Identify one or more purchase records that include one or more specified items that match one or more specified items included in any of the multiple invoice records, (2) For each of the purchase codes that have a common transaction code with one or more identified purchase records, the transaction identification code contained in each purchase record is matched with the invoice record containing the corresponding transaction identification code. Purchase records for which there is no invoice record containing the corresponding transaction identification code are not matched with any invoice records. (3) The step of identifying the transaction code that appears most frequently among the purchase records that are matched with any invoice record, The pairing means pairs each of the plurality of invoice records with each of the plurality of purchase records that includes the identified transaction code, and each of the invoice records, The first invoice record can be paired with any purchase record, and the invoice amount associated with that invoice record matches the purchase amount associated with the paired purchase record. A second invoice record was paired with one of the purchase records, but the invoice amount associated with that invoice record did not match the purchase amount associated with the paired purchase record. The third invoice record, which is not paired with any purchase record, is classified as follows: The method for calculating the total purchase amount includes the step of calculating the total purchase amount, which is the sum of the purchase amounts associated with purchase records paired with any invoice record, The means for identifying differences in amounts includes the steps of identifying the difference in amounts between the invoice amount associated with the second invoice record and the purchase amount associated with the paired purchase record, and calculating the total amount of the differences in amounts, The means for identifying unrecorded amounts includes the steps of identifying the invoice amount associated with the third invoice record as an unrecorded amount and calculating the total amount of the unrecorded amounts, An invoice matching method that includes the step of calculating the difference between (1) the sum of the total purchase amount, the sum of the amount differences, and the sum of the unrecorded amounts, and (2) the total invoice amount.

4. The invoice matching method according to claim 3, wherein the one or more specified items are a set of quantity, unit price, and amount.

5. The invoice matching method according to claim 1 or 3, wherein the transaction code is assigned to each branch of the payee.

6. The invoice matching method according to claim 1 or 3, wherein the transaction code is assigned to each branch of the billing recipient.

7. The invoice matching method according to claim 2 or 3, wherein the transaction identification code is an order number, purchase order number, invoice number, or delivery note number.

8. An invoice reading means reads invoice data containing multiple billing records and information on the total billing amount, associates the billing amount with each of the multiple billing records, and identifies the total billing amount. A purchase data reading means for reading purchase data that includes multiple purchase records, each of which is associated with a purchase amount, wherein each of the purchase records includes a purchase data reading means that includes a transaction code, Transaction identification means for identifying the purchase record corresponding to any of the aforementioned multiple invoice records, and for identifying the transaction code included in the identified purchase record, Each of the aforementioned multiple invoice records is paired with each of the aforementioned multiple purchase records that contains the identified transaction code, and each of the aforementioned invoice records is then paired with, The first invoice record can be paired with any purchase record, and the invoice amount associated with that invoice record matches the purchase amount associated with the paired purchase record. A second invoice record was paired with one of the purchase records, but the invoice amount associated with that invoice record did not match the purchase amount associated with the paired purchase record. A pairing method for classifying a third invoice record that has not been paired with any purchase record, A means for calculating the total purchase amount, which is the sum of the purchase amounts associated with purchase records paired with any invoice record, Regarding the second invoice record, a means for identifying the difference in amount between the invoice amount associated with the second invoice record and the purchase amount associated with the paired purchase record, and for calculating the total amount of the difference, With respect to the third billing record, the means for identifying the billing amount associated with the third billing record as an unaccounted amount and calculating the total amount of the unaccounted amounts, An invoice matching system comprising: (1) the sum of the total purchase amount, the sum of the amounts with discrepancies, and the sum of the unrecorded amounts; and (2) the total invoice amount, and a difference calculation means for calculating the difference between these two amounts.

9. An invoice reading means reads invoice data including multiple invoice records and information on the total invoice amount, associates the invoice amount with each of the multiple invoice records, and identifies the total invoice amount, wherein each of the invoice records includes one or more predetermined items and a transaction identification code. A purchase data reading means for reading purchase data, which includes multiple purchase records, each of which is associated with a purchase amount, wherein each of the purchase records includes one or more predetermined items, a transaction code, and a transaction identification code, (1) Identify one or more purchase records that include one or more specified items that match one or more specified items included in any of the multiple invoice records, (2) For each of the purchase codes that have a common transaction code with one or more identified purchase records, the transaction identification code contained in each purchase record is matched with the invoice record containing the corresponding transaction identification code. Purchase records for which there is no invoice record containing the corresponding transaction identification code are not matched with any invoice records. (3) A transaction identification means that identifies the transaction code that appears most frequently among the transaction codes included in the purchase record that is matched with any invoice record, Each of the aforementioned multiple invoice records is paired with each of the aforementioned multiple purchase records that contains the identified transaction code, and each of the aforementioned invoice records is then paired with, The first invoice record can be paired with any purchase record, and the invoice amount associated with that invoice record matches the purchase amount associated with the paired purchase record. A second invoice record was paired with one of the purchase records, but the invoice amount associated with that invoice record did not match the purchase amount associated with the paired purchase record. A pairing method for classifying a third invoice record that has not been paired with any purchase record, A means for calculating the total purchase amount, which is the sum of the purchase amounts associated with purchase records paired with any invoice record, Regarding the second invoice record, a means for identifying the difference in amount between the invoice amount associated with the second invoice record and the purchase amount associated with the paired purchase record, and for calculating the total amount of the difference, With respect to the third billing record, the means for identifying the billing amount associated with the third billing record as an unaccounted amount and calculating the total amount of the unaccounted amounts, An invoice matching system comprising: (1) the sum of the total purchase amount, the sum of the amounts with discrepancies, and the sum of the unrecorded amounts; and (2) the total invoice amount, and a difference calculation means for calculating the difference between these two amounts.

10. One or more computers, An invoice reading means reads invoice data containing multiple billing records and information on the total billing amount, associates the billing amount with each of the multiple billing records, and identifies the total billing amount. A purchase data reading means for reading purchase data that includes multiple purchase records, each of which is associated with a purchase amount, wherein each of the purchase records includes a purchase data reading means that includes a transaction code, Transaction identification means for identifying the purchase record corresponding to any of the aforementioned multiple invoice records, and for identifying the transaction code included in the identified purchase record, Each of the aforementioned multiple invoice records is paired with each of the aforementioned multiple purchase records that contains the identified transaction code, and each of the aforementioned invoice records is then paired with, The first invoice record can be paired with any purchase record, and the invoice amount associated with that invoice record matches the purchase amount associated with the paired purchase record. A second invoice record was paired with one of the purchase records, but the invoice amount associated with that invoice record did not match the purchase amount associated with the paired purchase record. A pairing method for classifying a third invoice record that has not been paired with any purchase record, A means for calculating the total purchase amount, which is the sum of the purchase amounts associated with purchase records paired with any invoice record, Regarding the second invoice record, a means for identifying the difference in amount between the invoice amount associated with the second invoice record and the purchase amount associated with the paired purchase record, and for calculating the total amount of the difference, With respect to the third billing record, the means for identifying the billing amount associated with the third billing record as an unaccounted amount and calculating the total amount of the unaccounted amounts, An invoice matching program that functions as a difference calculation means for calculating the difference between (1) the total amount of purchases, the total amount of discrepancies, and the total amount of unrecorded items, and (2) the total invoice amount.

11. One or more computers, An invoice reading means reads invoice data including multiple invoice records and information on the total invoice amount, associates the invoice amount with each of the multiple invoice records, and identifies the total invoice amount, wherein each of the invoice records includes one or more predetermined items and a transaction identification code. A purchase data reading means for reading purchase data, which includes multiple purchase records, each of which is associated with a purchase amount, wherein each of the purchase records includes one or more predetermined items, a transaction code, and a transaction identification code, (1) Identify one or more purchase records that include one or more specified items that match one or more specified items included in any of the multiple invoice records, (2) For each of the purchase codes that have a common transaction code with one or more identified purchase records, the transaction identification code contained in each purchase record is matched with the invoice record containing the corresponding transaction identification code. Purchase records for which there is no invoice record containing the corresponding transaction identification code are not matched with any invoice records. (3) A transaction identification means for identifying the transaction code that appears most frequently among the transaction codes included in the purchase record that is matched with any invoice record, Each of the aforementioned multiple invoice records is paired with each of the aforementioned multiple purchase records that contains the identified transaction code, and each of the aforementioned invoice records is then paired with, The first invoice record can be paired with any purchase record, and the invoice amount associated with that invoice record matches the purchase amount associated with the paired purchase record. A second invoice record was paired with one of the purchase records, but the invoice amount associated with that invoice record did not match the purchase amount associated with the paired purchase record. A pairing method for classifying a third invoice record that has not been paired with any purchase record, A means for calculating the total purchase amount, which is the sum of the purchase amounts associated with purchase records paired with any invoice record, Regarding the second invoice record, a means for identifying the difference in amount between the invoice amount associated with the second invoice record and the purchase amount associated with the paired purchase record, and for calculating the total amount of the difference, With respect to the third billing record, the means for identifying the billing amount associated with the third billing record as an unaccounted amount and calculating the total amount of the unaccounted amounts, An invoice matching program that functions as a difference calculation means for calculating the difference between (1) the sum of the total purchase amount, the sum of the difference in amounts, and the sum of the unrecorded amounts, and (2) the total invoice amount.

Images