Greenhouse gas emissions management methods

Abstract

To provide a method of enabling efficient greenhouse gas emissions management, including computation of greenhouse gas emissions by a business operator.SOLUTION: A greenhouse gas emissions management method according to an embodiment of the present invention comprises receiving image data of invoice information from a business operator terminal, extracting information on emission intensity based on a learning model using the image data as input data, and recommending the information on the emission intensity.SELECTED DRAWING: Figure 11

Description

[Technical Field]

[0001] The present invention relates to a method for managing greenhouse gas emissions. [Background technology]

[0002] With regard to greenhouse gas emissions from businesses resulting from the use of fuel, electricity, etc., reporting systems targeting SCOPE 1 emissions (direct emissions by the company) and SCOPE 2 emissions (indirect emissions by the company) have become widespread, and efforts to calculate and reduce emissions in SCOPE 1 and SCOPE 2 have progressed.

[0003] Non-Patent Document 1 proposes that, with the aim of further reducing greenhouse gas emissions by businesses, SCOPE 3 emissions should be calculated as emissions other than SCOPE 1 and SCOPE 2, i.e., emissions from the supply chain of other related businesses (the entire series of processes including raw material procurement, manufacturing, logistics, sales, and disposal). [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] "Concepts for Calculating Supply Chain Emissions," Ministry of the Environment, November 2017 Summary of the Invention [Problem to be solved by the invention]

[0005] However, although the technology disclosed in Non-Patent Document 1 discloses methods for calculating greenhouse gas emissions related to SCOPE 3, it takes a great deal of time and effort for each business, particularly companies and local governments, to collect and input a huge amount of data to calculate emissions, calculate emissions, and manage the calculation results.In particular, in the area of ​​GHG emissions management, the scope of emissions to be calculated is expanding, the data that forms the basis for emissions calculations varies widely for each scope, and data management methods differ for each business, so improvements in business efficiency through the introduction of advanced technology are lagging behind.

[0006] Therefore, the present invention aims to provide a method for efficiently managing GHG emissions by utilizing advanced technology in the area of ​​GHG emissions management, such as the calculation of greenhouse gas emissions by businesses. [Means for solving the problem]

[0007] A method for managing greenhouse gas emissions according to one embodiment of the present invention receives image data of invoice information from a business operator terminal, and using the image data as input data, extracts information related to emission intensity units based on a learning model, and recommends information related to the emission intensity units. [Effects of the Invention]

[0008] According to the present invention, it is possible to provide a method for efficiently realizing management such as calculation of greenhouse gas emissions by a business operator. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating a greenhouse gas emission management system according to a first embodiment of the present invention. [Figure 2] FIG. 2 is a functional block diagram of a management terminal constituting the greenhouse gas emission management system. [Figure 3] FIG. 2 is a functional block diagram of a business operator terminal that constitutes the greenhouse gas emission management system. [Figure 4] FIG. 3 is a diagram illustrating details of business operator data according to the first embodiment of the present invention. [Figure 5] FIG. 3 is a diagram illustrating details of invoice information according to the first embodiment of the present invention. [Figure 6] FIG. 3 is a diagram illustrating an example of transaction information according to the first embodiment of the present invention. [Figure 7] FIG. 10 is a diagram illustrating another example of transaction information according to the first embodiment of the present invention. [Figure 8] FIG. 3 is a flowchart showing an example of a process for calculating greenhouse gas emissions according to the first embodiment of the present invention. [Figure 9] FIG. 3 is a flowchart illustrating an example of a process for predicting causes of changes in greenhouse gas emissions according to the first embodiment of the present invention. [Figure 10] FIG. 3 is a flowchart showing an example of a transaction process of greenhouse gas emissions according to the first embodiment of the present invention. [Figure 11] FIG. 3 is a flowchart showing an example of an automatic emission unit assignment process for calculating greenhouse gas emissions according to the first embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0010] The contents of an embodiment of the present invention will be listed and explained below. A greenhouse gas emission management system (hereinafter simply referred to as "system") according to an embodiment of the present invention has the following configuration. [Item 1] 1. A method for managing greenhouse gas emissions, comprising: Accepting image data of invoice information from business terminals, extracting information about emission intensity units based on a learning model using the image data as input data; Recommendation methods for information on said emission intensity. [Item 2] 2. The management method according to item 1, wherein the learning model is generated by inputting information about emission intensity units associated with past input data that is similar to the input data. [Item 3] 3. The management method according to item 2, wherein information on emission intensity units that is associated with past input data similar to the input data is recommended based on the learning model. [Item 4] 2. The management method according to item 1, wherein the input data and the past input data each include text information recognized based on image data.

[0011] First Embodiment Hereinafter, a system according to an embodiment of the present invention will be described with reference to the drawings.

[0012] FIG. 1 is a diagram illustrating a greenhouse gas emission management system according to a first embodiment of the present invention.

[0013] As shown in FIG. 1, in an emission amount management system 1 according to this embodiment, an administrator terminal 100 and a plurality of business operator terminals 200A and 200B are connected to each other via a communication network NW.

[0014] For example, the management terminal 100 receives basic information about the business and input information (eg, image data of bill information) for calculating greenhouse gas (eg, CO2) emissions from the business terminals 200A and 200B.

[0015] The management terminal 100 also analyzes the received image data of the invoice information using machine learning, extracts necessary items of the invoice information contained in the image data, and calculates the greenhouse gas emissions. The management terminal 100 also analyzes the calculated changes in greenhouse gas emissions over time (e.g., increases or decreases) using machine learning, and predicts the causes of the changes.

[0016] Furthermore, the management terminal 100 has a wallet and is connected to the public blockchain network NW. Based on the information on greenhouse gas emissions for each predetermined period, the management terminal 100 generates a single hash value using SHA256 or another hash function and records it as transaction information on the blockchain network. On the blockchain network, a block is generated based on the transaction information, the hash value recorded in the previous block, and a nonce value mined by the node. This block is recorded after the previous block, forming a blockchain. The hash generation and / or recording of transaction information on the blockchain can also be performed via another terminal rather than the management terminal 100. In this case, the management terminal 100 transmits the greenhouse gas emissions calculated in the matching process to the other terminal. Furthermore, the management terminal 100 can record information on greenhouse gas emissions as a smart contract on the blockchain network. Using a smart contract, contracts for emissions trading with other businesses can be automatically generated, approved, and executed based on the emissions information without the involvement of a third party. Furthermore, the smart contract allows each business operator to refer to transaction information without going through the management terminal, improving service convenience and reducing operational costs. Furthermore, the management terminal 100 is connected to a large-scale language model (LLM, e.g., GPT) 300 via the network, which accepts predetermined questions from other terminals via an API (Application Programming Interface), analyzes the questions, and generates answers.

[0017] As described above, a public blockchain allows transactions to be approved by an unspecified number of nodes and miners, rather than by a specific administrator, and therefore can ensure higher data tamper resistance and fault tolerance than a private blockchain, thereby ensuring the security of transactions. Therefore, a public blockchain is preferable as the destination for recording energy transactions in this embodiment. Typical public blockchains include Bitcoin and Ethereum, and Ethereum, for example, has higher tamper resistance and reliability than other public blockchains.

[0018] Furthermore, the management terminal 100 can associate information regarding greenhouse gas emissions using an identifier or the like and record it on the blockchain network as a non-fungible token (hereinafter, "NFT"). NFTs are tokens issued, for example, in the "ERC721" standard of Etherium, a blockchain network platform, and are a unit of data recorded on the blockchain network, and are non-fungible in nature. NFTs are recorded on the blockchain together with smart contracts and are traceable, making it possible to prove transaction information, including details and history of business operators managing greenhouse gas emissions.

[0019] FIG. 2 is a functional block diagram of a management terminal that constitutes the emission management system.

[0020] The communication unit 110 is a communication interface for communicating with an external terminal via a network NW, and communication is performed according to a communication protocol such as TCP / IP (Transmission Control Protocol / Internet Protocol).

[0021] The storage unit 120 stores input data, programs for executing various control processes and functions in the control unit 130, and is composed of RAM (Random Access Memory), ROM (Read Only Memory), and the like. The storage unit 120 also has an operator data storage unit 121 that stores various data related to operators, and an AI model storage unit 122 that stores learning data and learning models obtained by AI (artificial intelligence) learning the learning data. A database (not shown) that stores various data may be constructed outside the storage unit 120 or the management terminal 100.

[0022] The control unit 130 controls the overall operation of the management terminal 100 by executing programs stored in the storage unit 120, and is composed of a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), etc. The functions of the control unit 130 include an information receiving unit 131 that receives information from an external terminal such as the business operator terminal 200, an image analysis unit 132 that analyzes image data such as invoice information received from the business operator terminal and calculates greenhouse gas emissions, a cause analysis unit 133 that analyzes the image data and analyzes the cause of time-series changes in the greenhouse gas emissions calculated based on information included in the extracted invoice information, a transaction processing unit 134 that compiles information on greenhouse gas emissions for a predetermined period, generates a hash value, and records the hash value as transaction information on the blockchain network, and a report generation unit 135 that generates and transmits report data to the business operator every predetermined period to output the results of an analysis of greenhouse gas emissions and the cause of changes in emissions.

[0023] Although not shown, the control unit 130 also has an image generation unit that generates screen information to be displayed via a user interface of an external terminal such as the business operator terminal 200. For example, the control unit 130 generates information to be displayed on the user interface by using image and text data stored in the storage unit 120 as material and arranging various images and text in predetermined areas of the user interface based on predetermined layout rules. Processing related to the image generation unit can also be executed by a GPU (Graphics Processing Unit).

[0024] The management terminal 100 also has a wallet (not shown) necessary for recording transaction information in the blockchain network. Note that this wallet may also be located outside the management terminal 100.

[0025] FIG. 3 is a functional block diagram of the business operator terminal that constitutes the emission management system.

[0026] The operator terminal 200 includes a communication unit 210 , a display operation unit 220 , a storage unit 230 , and a control unit 240 .

[0027] The communication unit 210 is a communication interface for communicating with the management terminal 100 via the network NW, and communication is carried out according to a communication protocol such as TCP / IP.

[0028] The display operation unit 220 is a user interface used by the business operator to input instructions and display text, images, etc. in accordance with input data from the control unit 240, and is configured with a display, keyboard, and mouse when the business operator terminal 200 is configured as a personal computer, and is configured with a touch panel, etc. when the business operator terminal 200 is configured as a smartphone or tablet terminal. This display operation unit 220 is started up by a control program stored in the storage unit 230 and executed by the business operator terminal 200, which is a computer (electronic calculator).

[0029] The storage unit 230 stores input data, programs for executing various control processes and functions within the control unit 240, and is composed of RAM, ROM, etc. The storage unit 230 also temporarily stores the contents of communications with the management terminal 100.

[0030] The control unit 240 controls the overall operation of the operator terminal 200 by executing the programs stored in the storage unit 230, and is configured from a CPU, a GPU, and the like.

[0031] FIG. 4 is a diagram illustrating details of business operator data according to the first embodiment of the present invention.

[0032] The business operator data 1000 shown in FIG. 4 stores various data related to the business operator acquired from the business operator via the business operator terminal 200. For ease of explanation, FIG. 4 shows an example of one business operator (a business operator identified by business operator ID "10001"), but information on multiple businesses can be stored. The various data related to the business operator can include, for example, basic information about the business operator (e.g., the business operator's corporate name, user name, address, industry, contact information, email address, business name, affiliated company name, and business operator names related in the supply chain), input information (e.g., image data of invoice information), analysis information (e.g., information regarding invoices extracted from image data, greenhouse gas emissions, and predicted causes of changes in greenhouse gas emissions), customer information (e.g., customer ID, blockchain address, and the like), and offset report information (e.g., TXID, NFTID, and the like).

[0033] FIG. 8 is a flowchart showing an example of the process of calculating greenhouse gas emissions according to the first embodiment of the present invention.

[0034] First, in the process of step S101, the information acquisition unit 131 of the control unit 130 of the management terminal 100 acquires image data including invoice information collected by the business operator from the business operator terminal 200 via the network NW. The business operator uploads invoices, receipts, slips, etc. (in this embodiment, these are collectively referred to as "invoices") to the management terminal 100 via the business operator terminal 200 in file formats such as PDF, Excel, JPG, etc. (in this embodiment, these are collectively referred to as "image data"). The image data acquired by the information acquisition unit 131 is stored as input information in the business operator data storage unit 121 of the memory unit 120.

[0035] Next, in step S102, the image analysis unit 132 of the control unit 130 of the management terminal 100 analyzes the image data acquired in the previous step using machine learning. Here, the image analysis uses a technique known as OCR, and the image analysis unit 132 of the control unit 130 of the management terminal 100 recognizes text from the image data using a learning model that has been generated in advance by learning from image data of multiple invoices in various formats and that is stored in the AI ​​model storage unit 122 of the memory unit 120, and extracts items contained in the invoice information as structured character string data. Here, the image analysis can also use an image analysis engine (such as an OCR engine) provided by a business operator other than the management terminal 100, which is linked via an API.

[0036] Image analysis is performed by recognizing and extracting text from image data containing invoice information, as shown in FIG. 5. As shown in FIG. 5, invoice information includes various items, such as the electricity bill breakdown, the amount (yen) for each breakdown, the contracted power (kW), the amount of electricity used (kWh) for each breakdown, the total amount (yen), and the date (year and month). While this example illustrates an electricity bill breakdown, it may also be an invoice for other energy usage, including electricity, gas, and fuel. It may also be a receipt for travel expenses, a receipt for employee commuting expenses, an invoice for a transaction with a freight company, or a breakdown of an invoice for a transaction with a waste disposal company. The image analysis unit 132 analyzes the image data of this invoice information to extract, as text, the amount information, activity level information, and other information contained in the invoice information. The extracted invoice information is stored as analysis information in the business operator data storage unit 121 of the memory unit 120. In this way, image analysis using machine learning allows businesses to obtain the vast amount of information necessary to calculate greenhouse gas emissions as image data without having to manually enter invoice information, and highly accurate image recognition makes it possible to accurately extract the information necessary to calculate greenhouse gas emissions, thereby making it possible to calculate greenhouse gas emissions more efficient and with greater accuracy.

[0037] Next, in step S103, the image analysis unit 132 of the control unit 130 calculates greenhouse gas emissions based on the invoice information extracted from the image data or based on information on energy usage obtained from the LLM 300. Here, greenhouse gas emissions are classified into SCOPE 1, SCOPE 2, and SCOPE 3, with SCOPE 1 being direct greenhouse gas emissions by the business itself (e.g., emissions associated with fuel combustion and industrial processes), SCOPE 2 being indirect emissions associated with the use of electricity, heat, gas, etc. supplied to the business by other companies, and SCOPE 3 being the calculation standard for emissions throughout an organization's supply chain issued by the GHG Protocol, and referring to emissions throughout a business's supply chain (the entire series of processes, including raw material procurement, manufacturing, logistics, sales, and disposal). SCOPE 3 is further classified into 15 categories (1) purchased products / services, 2) capital goods, 3) fuel and energy-related activities not included in SCOPE 1 and SCOPE 2, 4) transportation and distribution (upstream), 5) waste generated from operations, 6) business travel, 7) employee commuting, 8) leased assets (upstream), 9) transportation and distribution (downstream), 10) processing of sold products, 11) use of sold products, 12) disposal of sold products, 13) leased assets (downstream), 14) franchises, and 15) investments. Here, greenhouse gases include carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), sulfur hexafluoride (SF6), and nitrogen trifluoride (NF3), but this embodiment will be described using CO2 as an example.

[0038] Greenhouse gas emissions are calculated by defining a business's electricity consumption, cargo transport volume, waste disposal volume, and various transaction amounts as activity volume, and multiplying this activity volume by the CO2 emissions per kWh of electricity used, CO2 emissions per ton of cargo transported, and CO2 emissions per ton of waste incinerated as emissions intensity. Greenhouse gas emissions are calculated for the above-mentioned SCOPE 1, SCOPE 2, and SCOPE 3 (SCOPE 3 is divided into 15 categories), and the total emissions are calculated as supply chain emissions.

[0039] In this embodiment, the image analysis unit 132 extracts related invoice information by SCOPE 1, SCOPE 2, and SCOPE 3 (and further by category for SCOPE 3), and calculates emissions based on the invoice information, for example, the amount of electricity used (kWh), using the above calculation method. The calculated emissions are stored as analysis information in the business operator data storage unit 121 of the memory unit 120.

[0040] Next, in the processing of step S104, the report generation unit 135 of the control unit 130 generates a visualized report showing a breakdown of emissions over time by SCOPE (and further by category for SCOPE 3) based on the information regarding the calculated emissions.

[0041] FIG. 9 is a flowchart showing an example of the process of predicting the causes of changes in greenhouse gas emissions according to the first embodiment of the present invention.

[0042] First, in the process of step S201, the cause analysis unit 133 of the control unit 130 of the management terminal 100 refers to information about the greenhouse gas emissions of the business operator calculated in step S103 of FIG. 8. Here, for the greenhouse gas emissions, the emissions by SCOPE (and further by category for SCOPE 3) are referenced. Furthermore, the cause analysis unit 133 can check changes (increases or decreases) in the greenhouse gas emissions by referring to the past emission data of the same business operator. As described above, the emissions are stored as analysis information in the business operator data storage unit 121 of the memory unit 120.

[0043] Next, in the processing of step S202, the cause analysis unit 133 analyzes and predicts the cause of changes in emission amounts by machine learning based on the emission amount information referenced above. Here, when analyzing the cause, the cause analysis unit 133 of the control unit 130 of the management terminal 100 predicts the cause of changes in emission amounts by SCOPE (and further by category for SCOPE 3) by using the emission amount information referenced above, factors that affect changes (increases or decreases) in emission amounts, and a learning model that has been generated in advance by learning data related to factors that affect changes (increases or decreases) in emission amounts that is stored in the AI ​​model storage unit 122 of the memory unit 120.

[0044] Factors that affect changes (increases or decreases) in output include, for example, weather, temperature, product demand and / or factory operation, store or factory business hours, equipment or facility changes, software measures, energy-saving activities, fuel conversions, energy menu changes, changes in business trips or commuting, and the amount of electricity generated by privately-owned power plants. Each of these factors affects emissions in one of the following scopes. For example, weather affects precipitation, wind volume, hours of sunshine, and temperature. Precipitation affects the amount of hydroelectric power generation, wind volume affects the amount of wind power generation, hours of sunshine affect the amount of solar power generation, and temperature affects air conditioning. Furthermore, the amount of electricity generated affects the amount of privately-owned power generation, which affects CO2 emissions from electricity, thereby affecting emissions in SCOPE 2. Meanwhile, air conditioning affects gas usage, which affects CO2 emissions from gas combustion, thereby affecting emissions in SCOPE 1. Energy-saving activities, factory operation due to product demand, and business hours affect electricity usage, which in turn affects SCOPE 2. In addition, EMS, replacement of refrigeration equipment, introduction of energy-saving equipment, and automobile usage also affect electricity usage and thus affect SCOPE 2. In addition, automobile usage, fuel efficiency, boiler usage, and boiler efficiency affect fuel usage, which affects CO2 emissions from fuel, thereby affecting SCOPE 1.

[0045] In addition, the factor of product sales volume affects categories 1, 9, 10, 11, and 12 in SCOPE 3, capital investment affects category 2, renewable energy ratio and amount of energy procured affects category 3, number of transports and changes in transport routes affects categories 4 and 9, product loss rate affects category 5, business trips and number of people commuting to the office affects category 6, number of people commuting to the office and number of employees commuting to the office affects category 7, electricity consumption affects category 8, reduction in processing through product improvements affects category 10, improvements to energy-efficient products affects category 11, increased recycling rate affects category 12, electricity used in tenant offices affects category 13, franchise emissions affects category 14, and emissions from investment targets affects category 15.

[0046] In this way, by using machine learning to learn which factors affect which scope or category, and then obtaining information on emissions and information on each factor from businesses, it is possible to predict the causes of changes in emissions. Here, by predicting the causes of emissions using machine learning, it is possible to efficiently and accurately predict the factors that affect changes in greenhouse gas emissions for each business and for each scope.

[0047] Next, in the processing of step S203, the report generation unit 135 of the control unit 130 generates a visualized report on the causes of changes in emissions by SCOPE (and further by category for SCOPE 3) based on the information regarding the predicted causes of changes in emissions analyzed above.

[0048] FIG. 10 is a flowchart showing an example of a transaction process of greenhouse gas emissions according to the first embodiment of the present invention.

[0049] First, in the process of step S301, the transaction processing unit 134 of the control unit 130 of the management terminal 100 refers to the business operator data stored in the business operator data storage unit 121 of the memory unit 120. Here, the business operator data to be referred to includes business operator analysis information (emissions of greenhouse gases for each SCOPE), etc.

[0050] Next, in step S302, the transaction processing unit 134 generates a hash value based on the business data referenced in step S301. That is, the transaction processing unit 134 generates a row of hash values ​​for greenhouse gas emissions over a predetermined period using a hash function and records the hash value as transaction information in the public blockchain. On the blockchain network, this block is generated based on the transaction information, the hash value recorded in the previous block, and the nonce value mined by the node, and is recorded following the previous block, forming a blockchain. Here, in this example, to reduce the cost of blockchain recording, it is assumed that the block is recorded in layer 2 (e.g., a side chain) separate from the main blockchain (so-called layer 1).

[0051] Furthermore, the transaction processing unit 134 can assign and manage NFTIDs by linking them to the blockchain records of the greenhouse gas emissions of the business. More specifically, as shown in Fig. 4, the business's customer ID can be assigned as customer information to the business data 1000, the blockchain address to be referenced can be stored, and the NFTID and TXID can be assigned as offset report information.

[0052] As shown in Figure 6, each NFT ID is associated with a blockchain address on the blockchain network, and the management terminal 100 manages NFT IDs and customer IDs. For example, information about greenhouse gas emissions for a business associated with customer ID "2" can be retrieved by referencing the blockchain address for each NFT ID, such as NFT IDs "13" and "14," as shown in Figure 7. Figure 7 shows information about an offset report associated with NFT ID "14." A TXID is assigned to the offset report, and the offset report includes information about CO2 emissions by scope, the target year and month, and the report's publication date. In addition to the CO2 emissions for the target year and month in this example, recent CO2 emissions, reduced CO2 emissions, and offset CO2 emissions can also be converted into NFTs. By managing CO2 emissions in this way, businesses can trade NFT-converted certificates while ensuring tamper-resistance and transaction reliability, and can also verify emissions to third parties.

[0053] FIG. 11 is a flowchart showing an example of an automatic emission unit assignment process for calculating greenhouse gas emissions according to the first embodiment of the present invention.

[0054] First, as a preprocessing step for this process, image data of multiple invoices and / or receipts (on which energy usage amounts are recorded) is used as input data, natural language analysis is performed via image analysis based on the input data, and information on the emission intensity units corresponding to the recognized image data and / or text data is input as correct answer data to generate a machine learning model. The learning model can be stored in the AI ​​model storage unit 122 of the memory unit 120 of the management terminal 100 and used for machine learning processing in the management terminal 100, or the learning model can be generated in the LLM 300, and the management terminal 100 and the LLM 300 can work together to perform machine learning processing.

[0055] In the process of step S401, the information acquisition unit 131 of the control unit 130 of the management terminal 100 acquires image data including invoice information and / or receipts collected by the business operator from the business operator terminal 200 via the network NW. The business operator uploads invoices, receipts, slips, etc. (in this embodiment, these are collectively referred to as "invoices") to the management terminal 100 via the business operator terminal 200 in file formats such as PDF, Excel, JPG, etc. (in this example, these are collectively referred to as "image data"). The image data acquired by the information acquisition unit 131 is stored as input information in the business operator data storage unit 121 of the memory unit 120.

[0056] Next, in step S402, the image analysis unit 132 of the control unit 130 of the management terminal 100 analyzes the image data acquired in the previous step using machine learning. Here, the image analysis uses a technique known as OCR, and the image analysis unit 132 of the control unit 130 of the management terminal 100 recognizes text from the image data using a learning model that has been generated in advance by learning from image data of multiple invoices in various formats and that is stored in the AI ​​model storage unit 122 of the memory unit 120, and extracts items contained in the invoice information and / or receipt information as structured character string data. Here, the image analysis can also use an image analysis engine (such as an OCR engine) provided by a business operator other than the management terminal 100 that is linked via an API. Furthermore, in this step, the image analysis unit 132 refers to the learning model stored in the AI ​​model storage unit 122 of the memory unit 120 or generated via the LLM 300 for assigning the generated emission intensity information, and performs a process of searching for and extracting information relating to one or more emission intensity units that correspond to past input data and are similar to the recognized and structured text data.

[0057] Next, in step S403, the report generation unit 135 of the control unit 130 of the management terminal 100 recommends the extracted information on the emission intensity units as candidates for the emission intensity units for the input data. The business operator confirms, via the application screen or chat interface screen displayed on the business operator terminal 200, information on one or more emission intensity units corresponding to the input image data, and performs a selection and / or confirmation process to confirm the information on the emission intensity units for the input data. The confirmed information on the emission intensity units is input into the learning model as correct answer data, thereby updating the learning model.

[0058] In this way, while invoices and receipts originally did not contain information regarding emission intensity, this example makes it possible to infer emission intensity from the image data of input invoices and / or receipts, thereby making the calculation process of greenhouse gas emissions more efficient.

[0059] The above-described embodiment is merely an example for facilitating understanding of the present invention, and is not intended to limit the present invention. The present invention can be modified and improved without departing from the spirit thereof, and it goes without saying that the present invention includes equivalents thereof. [Explanation of symbols]

[0060] 100 Management Terminal 200 Operator terminal

Claims

[Claim 1] 1. A method for managing greenhouse gas emissions, comprising: The control unit of the management terminal is receiving first image data of bill information from the business operator terminal; using image data of multiple invoices and / or receipts containing energy usage amounts, which are different from the invoice information, as input data, performing natural language analysis via image analysis based on the input data, and inputting information on emission intensity units as correct answer data corresponding to the image data and / or text data recognized by the image analysis and the natural language analysis, thereby generating a machine learning model; extracting information about an emission intensity based on the machine learning model using the first image data as input data; Recommendation methods for information on said emission intensity.

Images