Information processing system and information processing method

Abstract

An information processing system (1) according to the present invention is characterized by: training an emissions calculation model (39) by machine learning; training an emission right unit cost calculation model (40) by machine learning; obtaining a future emissions estimate by entering a future value related to the activities of a business operator into the machine-trained emissions calculation model; obtaining a future emissions unit cost by entering a future value related to the market environment into the machine-trained emission right unit cost model; calculating a compensation for an emissions limit margin value or an emissions limit shortfall value according to the magnitude relationship between an emissions limit value and the emissions estimate; and displaying the calculated compensation.

Description

Information processing system and information processing method

[0001] The present invention relates to an information processing system and an information processing method for evaluating the profitability of a business's carbon offset activities.

[0002] Currently, there are concerns about climate change and the impact on ecosystems caused by the expansion of social activities on a global scale. Efforts to reduce the environmental impact of social activities and make society sustainable, such as achieving carbon neutrality and sustainability, are becoming essential requirements not only for national and local governments but also for individual businesses.

[0003] To demonstrate the environmental impact of their business activities, businesses must continually respond to trends in ESG (Environment, Social, and Governance) investment and the disclosure requirements of the Task Force on Climate-related Financial Disclosures. Furthermore, to achieve carbon neutrality, businesses must understand the greenhouse gas portfolios associated with their business activities and develop scenarios for long-term transitions.

[0004] The path to carbon neutrality requires not only the reduction of greenhouse gas emissions but also the procurement of energy through renewable or generated energy sources. Furthermore, it is also essential to utilize carbon offsetting activities such as green certificates, afforestation, and the trading of carbon credits in exchange for greenhouse gas emission rights.

[0005] Patent Document 1 is CO-- ２ The summary of the device is that "the calculation unit calculates the CO of each power plant of each power generation company. ２ CO by collection company ２ CO after recovery ２ Calculate the emissions and CO ２ The total amount of emissions (after collection) is CO ２ CO held by collection company ２ It is determined whether the total amount is smaller than the allowable emission amount determined by the emission rights. If the total amount is larger than the allowable emission amount, the user inputs CO ２ Accept the change in the recovery rate and ２CO of each power plant to be recovered ２ CO after recovery ２ On the other hand, if the total amount is less than the allowable amount of emissions, CO ２ Disposal costs and recovered CO ２ The profits obtained from the secondary use of CO ２ CO held by collection company ２ CO of Emissions Rights ２ Emissions and CO2 emissions ２ CO per emission ２ The selling price of the emission rights will be determined.

[0006] Japanese Patent Application Laid-Open No. 2006-350401

[0007] However, the device disclosed in Patent Document 1 is intended to be implemented in a power plant. ２ Emissions and CO ２ The various data formats that must be prepared to calculate the selling price of emissions credits are fixed, and there is room for improvement in making the system easier to use for general businesses other than power generation companies.

[0008] Furthermore, in the device according to Patent Document 1, if the total amount is greater than the allowable discharge amount, ２ Accept the change of the recovery rate and CO after recovery ２ It simply recalculates the emissions, ２ The focus is not on determining the price of purchasing the emission rights. If the total amount is less than the allowable emissions, the CO ２ However, the device determines the selling price of the emission rights. ２ Disposal costs, recovered CO ２ Profits and COs obtained from secondary use of ２ CO held by collection company ２ CO of Emissions Rights ２ If you don't know the amount of emissions, ２ A separate method is required to determine the selling price of emission credits. Therefore, the present invention aims to enable general businesses to easily calculate the price of greenhouse gas emission credits, regardless of the level of greenhouse gas emissions.

[0009] The information processing system of the present invention includes an emission learning unit that machine-learns an emission calculation model, which takes values ​​related to the activities of an enterprise as input and outputs the amount of greenhouse gas emissions emitted by the enterprise, using past values ​​related to the activities of the enterprise and the past values ​​of the amount of greenhouse gas emissions emitted by the enterprise as learning data; an emission credit unit price learning unit that machine-learns an emission credit unit price calculation model, which takes values ​​related to the environment of a market in which greenhouse gas emission credits are traded as input and outputs the unit price of greenhouse gas emission credits in the market, using past values ​​related to the environment of the market and the past values ​​of the unit price of the emission credit; a processing condition setting unit that receives future values ​​related to the activities of the enterprise, future values ​​related to the environment of the market, and emission limit values ​​corresponding to the greenhouse gas emission credits held by the enterprise; and inputs the received future values ​​related to the activities of the enterprise into the machine-learned emission calculation model. an emission estimation unit that inputs the received future values ​​related to the market environment into the machine-learned emission credit unit price calculation model and acquires an estimated future emission credit unit price from the machine-learned emission credit unit price calculation model, and a calculation unit that, if the emission limit value is greater than the estimated emission value, calculates a consideration for the emission limit margin value by multiplying the estimated emission credit unit price by an emission limit margin value obtained by subtracting the estimated emission value from the emission limit value, and, if the emission limit value is smaller than the estimated emission value, calculates a consideration for the emission limit shortfall by multiplying the estimated emission credit unit price by an emission limit shortfall value obtained by subtracting the emission limit value from the estimated emission value, and displays the calculated consideration. Other means will be described in the description of the embodiment of the invention.

[0010] According to the present invention, a general business operator can easily calculate the price of greenhouse gas emission rights, regardless of the level of greenhouse gas emissions.

[0011] Fig. 1 is a diagram showing a configuration of an information processing system. Fig. 2 is a flowchart of an emission amount calculation model learning processing procedure. Fig. 3 is a flowchart of an emission credit unit price calculation model learning processing procedure. Fig. 4 is a flowchart of a consideration calculation processing procedure. Fig. 5 is an example of an emission amount estimation explanatory variable reception screen. Fig. 6 is an example of an emission credit unit price estimation explanatory variable reception screen.

[0012] Hereinafter, an embodiment of the present invention (hereinafter referred to as "the present embodiment") will be described in detail with reference to the drawings etc. In this embodiment, a general business entity carries out carbon offset activities. "General" means that it does not matter what type of business or size of company. A business entity can offset its own CO2 emissions by ２ It would be convenient if information on one's own daily business activities could be used directly to calculate (estimate) emissions. ２ It would be convenient if you could use the environmental information on the market that you can obtain directly to calculate (estimate) the unit price of emission rights. ２ is an example of a greenhouse gas.

[0013] (Configuration of Information Processing System) FIG. 1 is a diagram showing the configuration of an information processing system 1. The information processing system 1 is a general-purpose computer. The information processing system 1 includes a central control unit 11, an input device 12 such as a keyboard or a mouse, an output device 13 such as a display, a main memory device 14, an auxiliary memory device 15, and a communication device 16. The auxiliary memory device 15 stores an emissions database 31, an emission credit unit price database 35, an emissions calculation model 39, and an emission credit unit price calculation model 40. The auxiliary memory device 15 may be configured independent of the information processing system 1.

[0014] The emissions database 31 includes net emissions information 32, business activity emissions information 33, and renewable energy generation emission reduction information 34. The business activity emissions information 33 includes direct emissions information 33a, indirect emissions information 33b, and other party emissions information 33c (described in detail below). The emission credit unit price database 35 includes emission credit unit price information 36, weather information 37, and business activity information 38 (described in detail below).

[0015] The processing condition setting unit 21, emission amount learning unit 22, emission amount estimation unit 23, emission allowance unit price learning unit 24, emission allowance unit price estimation unit 25, and calculation unit 26 stored in the main memory device 14 are programs. In the following description, when the subject is described as "XX unit," this means that "the central control device 11 reads each program from the auxiliary memory device 15 to the main memory device 14, thereby realizing the function previously described in that program."

[0016] (Emissions database 31) Business activity emissions information 33 stores business activity emissions in chronological order. Business activity emissions are greenhouse gas emissions resulting from a company's past business activities. In this embodiment, the business activity emissions information 33 stores business activity emissions of the entire supply chain in the company's business after classifying them into three types.

[0017] The direct emission information 33a stores direct emission amounts in chronological order. Direct emission amounts are caused by greenhouse gas emissions by the business itself. For example, greenhouse gas emissions emitted by the business itself in the fuel combustion process or product manufacturing process are direct emission amounts.

[0018] The indirect emission information 33b stores indirect emissions in chronological order. Indirect emissions are caused by a business's consumption of energy supplied by another party. For example, greenhouse gas emissions caused by a business's consumption of energy (electricity, heat, steam, etc.) supplied by another party to the business are indirect emissions. In this case, it is the other party (the energy supplier) that actually emitted the greenhouse gases. The business that used the energy is considered to have emitted greenhouse gases in the calculation of business activity emissions.

[0019] The third-party emission information 33c stores third-party emissions in chronological order. Third-party emissions result from greenhouse gas emissions by third parties (other than energy suppliers) in the supply chain to which the business belongs. For example, greenhouse gas emissions by parts manufacturers upstream in the supply chain and assembly subcontractors downstream are third-party emissions. In this case, it is natural that the greenhouse gases were actually emitted by third parties. The business representing the supply chain is assumed to have emitted greenhouse gases in the calculation of business activity emissions.

[0020] The renewable energy generation emission reduction information 34 stores the renewable energy generation emission reduction amount in chronological order. The renewable energy generation emission reduction amount is the greenhouse gas emissions that have been reduced by a business operator in the past by implementing renewable energy and / or energy generation. For example, a business operator that generates renewable energy and / or generated energy using solar power generation equipment or wind power generation equipment is evaluated as having reduced the amount of greenhouse gases that would have been emitted if the business operator had used thermal power generation equipment. This evaluated amount is the renewable energy generation emission reduction amount.

[0021] The net emission information 32 stores net emissions in chronological order. Net emissions are the net amount of greenhouse gas emissions resulting from the past activities of a business. Net emissions are basically the value obtained by subtracting the amount of emissions reduced by renewable energy generation from the amount of emissions from business activities, but this is often not so simple (see below for details).

[0022] (Emissions credit unit price database 35) Greenhouse gas emissions credits are traded for a fee in the market. While the emissions database 31 is a historical database relating to individual businesses, the emissions credit unit price database 35 is a historical database relating to such a market, and is more macroscopic.

[0023] The emission credit unit price information 36 stores emission credit unit prices in chronological order. The emission credit unit price is the price (yen / kg) per unit quantity of greenhouse gas emission credits in the market in the past.

[0024] The weather information 37 stores weather observation values ​​in chronological order. The weather observation values ​​are past observation values ​​of the natural environment of the area where the market is located. The weather observation values ​​here include at least one of temperature, humidity, wind speed, wind direction, sunshine duration, and sunshine angle.

[0025] The business activity information 38 stores macroeconomic indicators in chronological order. Macroeconomic indicators are indicators that directly or indirectly indicate the past level of economic activity in the region to which the market belongs, such as national (prefectural and city) income, savings rate, employed population, consumer population, population flow, industrial production, number of housing starts, and volume of transported goods. Macroeconomic indicators can also be considered as statistical values ​​of the economic environment of the region to which the market belongs.

[0026] (Emission Calculation Model 39) The emission calculation model 39 is, for example, the following function F １ It is. ｔ =F １ (p ｔ , q ｔ , r ｔ , s ｔ ) where G ｔ is the net emission at time t. ｔ is the direct emissions at time t. ｔ is the indirect emissions at time t. ｔ is the amount of emissions from other parties at time t. ｔ is the amount of emission reduction due to renewable energy generation at time t.

[0027] Function F １ may be any function such as a linear function, a higher-order function, a logarithmic function, an exponential function, a trigonometric function, or a composite or combination function of these functions. １ has multiple parameters (coefficients, etc.) that determine its position and shape in multidimensional space.

[0028] Furthermore, the function F １ is the vector "(p ｔ , q ｔ , r ｔ , s ｔ )" to get the scalar "G ｔIn this case, a propagation parameter is defined for each node in the intermediate layer, which determines how much information is to be propagated to which node in the next intermediate layer.

[0029] The discharge amount learning unit 22 uses "p ｔ , q ｔ , r ｔ , s ｔ " and "G ｔ The above-mentioned parameters or propagation parameters can be optimized by performing machine learning (including simpler multiple regression analysis, etc.) on the emission calculation model 39 using supervised learning data, which is a set of combinations of ".

[0030] (Necessity of the emission calculation model 39) ｔ , q ｔ , r ｔ , s ｔ If the units of " and " are standardized as "kg", for example, the function F １ becomes extremely simple, consisting only of addition and subtraction. ｔ = p ｔ +q ｔ +r ｔ -s ｔ " holds true.

[0031] However, general businesses use units other than "kg" to measure "p" ｔ , q ｔ , r ｔ , s ｔ ". For example, transportation companies often manage figures equivalent to direct emissions using vehicle distance (km), fuel consumption (kl), fuel purchase price (yen), and transported cargo weight (tons). Depending on the industry, production volume (units, tons), factory operation rate (%), etc. may be managed as figures equivalent to direct emissions. The same applies to indirect emissions, third-party emissions, and emission reductions from renewable energy generation.

[0032] Therefore, an emission calculation model 39 is required that can be used in units that are used daily by businesses. ｔ , q ｔ, r ｔ , s ｔ Although each of these is named as "XX emissions" for the sake of convenience, it is a broad concept that can ultimately be converted into the weight (kg) of greenhouse gases.

[0033] For example, if the business is a department store, any information such as the number of customers, the temperature in the store, the amount of merchandise sold, and the amount of merchandise sold can be stored in the “p ｔ , q ｔ , r ｔ The number of solar panels installed on the roof of the store, the operation rate, the hours of sunlight on the roof of the store, etc. can be used to calculate "s ｔ "It could become.

[0034] (Emissions credit unit price calculation model 40) The emissions credit unit price calculation model 40 uses, for example, the following function F ２ It is. W ｔ =F ２ (x ｔ , y ｔ ) where W ｔ is the unit price of emission rights at time t. ｔ is the meteorological observation value at time t. ｔ is the macroeconomic indicator at time t.

[0035] Function F ２ may be any function such as a linear function, a higher-order function, a logarithmic function, an exponential function, a trigonometric function, or a composite or combination function of these functions. ２ has multiple parameters (coefficients, etc.) that determine its position and shape in multidimensional space.

[0036] Furthermore, the function F ２ is the vector "(x ｔ , y ｔ )" to get the scalar "W ｔ In this case, a propagation parameter is defined for each node in the intermediate layer, which determines how much information is to be propagated to which node in the next intermediate layer.

[0037] The emission credit unit price learning unit 24 uses "x ｔ , y ｔ" and "W" as the objective variable ｔ By performing machine learning (including simpler multiple regression analysis, etc.) on the emission credit unit price calculation model 40 using supervised learning data, which is a set of combinations of "

[0038] The emission rights unit price calculation model 40 is ｔ , y ｔ For example, if the operator is a railway operator, the temperature along the railway line, the population along the railway line, the railway transport population including competitors, etc. can be used as "x ｔ , y ｔ "It could become.

[0039] (Summary) As is clear from the above, the emission amount learning unit 22 uses values ​​related to the activities of a business as input and machine-learns an emission amount calculation model 39 that outputs the amount of greenhouse gases emitted by the business using past values ​​related to the activities of the business and past values ​​of the amount of greenhouse gases emitted by the business as learning data. Here, the "values ​​related to the activities of a business" correspond to direct emissions, indirect emissions, emissions from others, and emission reductions due to renewable energy generation. The "emission amount of greenhouse gases emitted by a business" corresponds to net emissions.

[0040] The emission credit unit price learning unit 24 uses values ​​related to the market environment in which greenhouse gas emission credits are traded as input and outputs the unit price of greenhouse gas emission credits in the market, and performs machine learning of the emission credit unit price calculation model 40 using past values ​​related to the market environment and past values ​​of the emission credit unit price as learning data. Here, the "values ​​related to the market environment" correspond to meteorological observation values ​​and macroeconomic indicators.

[0041] (Processing Procedures) The processing procedures will be explained below. There are three processing procedures: an emission amount calculation model learning processing procedure, an emission credit unit price calculation model learning processing procedure, and a consideration calculation processing procedure. Before the consideration calculation processing procedure is executed, the emission amount calculation model learning processing procedure and the emission credit unit price calculation model learning processing procedure are executed.

[0042] 2 is a flowchart of the emission calculation model learning process. In step S101, the emission learning unit 22 of the information processing system 1 acquires the net emission amount information 32. Specifically, the emission amount learning unit 22 acquires the net emission amount information 32 from the emission amount database 31.

[0043] In step S102, the emission amount learning unit 22 acquires the business activity emission amount information 33. Specifically, the emission amount learning unit 22 acquires the direct emission amount information 33a, the indirect emission amount information 33b, and the other party emission amount information 33c from the emission amount database 31.

[0044] In step S103, the emission amount learning unit 22 acquires the renewable energy generation emission reduction amount information 34. Specifically, the emission amount learning unit 22 acquires the renewable energy generation emission reduction amount information 34 from the emission amount database 31.

[0045] In step S104, the emission amount learning unit 22 performs machine learning on the emission amount calculation model 39. Specifically, first, the emission amount learning unit 22 acquires the emission amount calculation model 39 before machine learning from the auxiliary storage device 15. Second, the emission amount learning unit 22 uses the information acquired in steps S101 to S103 to acquire the emission amount calculation model 39 before machine learning. ｔ , q ｔ , r ｔ , s ｔ " and "G ｔ " combinations.

[0046] Third, the emission amount learning unit 22 uses the supervised learning data generated in "second" of step S104 to machine-learn the emission amount calculation model 39 acquired in "first" of step S104 (optimize parameters). Fourth, the emission amount learning unit 22 stores the machine-learned emission amount calculation model 39 in the auxiliary storage device 15. Thereafter, the emission amount learning process procedure ends.

[0047] 3 is a flowchart of the emission credit unit price calculation model learning process. In step S201, the emission credit unit price learning unit 24 of the information processing system 1 acquires the emission credit unit price information 36. Specifically, the emission credit unit price learning unit 24 acquires the emission credit unit price information 36 from the emission credit unit price database 35.

[0048] In step S202, the emission credit unit price learning unit 24 acquires the weather information 37. Specifically, the emission credit unit price learning unit 24 acquires the weather information 37 from the emission credit unit price database 35.

[0049] In step S203, the emission credit unit price learning unit 24 acquires the business activity information 38. Specifically, the emission credit unit price learning unit 24 acquires the business activity information 38 from the emission credit unit price database 35.

[0050] In step S204, the emission credit unit price learning unit 24 performs machine learning on the emission credit unit price calculation model 40. Specifically, first, the emission credit unit price learning unit 24 acquires the emission credit unit price calculation model 40 before machine learning from the auxiliary storage device 15. Second, the emission credit unit price learning unit 24 uses the information acquired in steps S201 to S203 to learn the explanatory variable "x ｔ , y ｔ " and "W" as the objective variable ｔ " combinations.

[0051] Third, the emission credit unit price learning unit 24 uses the supervised learning data generated in "second" of step S204 to machine-learn the emission credit unit price calculation model 40 acquired in "first" of step S204 (optimize parameters). Fourth, the emission credit unit price learning unit 24 stores the machine-learned emission credit unit price calculation model 40 in the auxiliary storage device 15. Thereafter, the emission credit unit price calculation model learning process procedure ends.

[0052] 4 is a flowchart of the compensation calculation process. In step S301, the processing condition setting unit 21 of the information processing system 1 accepts the values ​​of the explanatory variables at the target time point. Specifically, first, the processing condition setting unit 21 displays an explanatory variable acceptance screen 41 (FIG. 5) for estimating emission amount on the output device 13.

[0053] Second, the processing condition setting unit 21 accepts input (selection) of explanatory variable values ​​at a future target time point by a user (such as a business operator) via the input device 12. Although we are still in the middle of Fig. 4, the explanation will now move on to Fig. 5.

[0054] (Explanatory Variable Reception Screen for Emissions Estimation) FIG. 5 is an example of the explanatory variable reception screen 41 for emission estimation. The explanatory variable reception screen 41 for emission estimation displays direct emissions, indirect emissions, third-party emissions, and renewable energy generation emission reductions in chronological order. These are obtained from the direct emissions information 33a, indirect emissions information 33b, third-party emissions information 33c, and renewable energy generation emission reductions information 34, respectively. "#" abbreviates and indicates a different numerical value. For example, the user may select a different past date value for each explanatory variable (thick solid line), or may select a common past date for all explanatory variables (thick dashed line). Note that past dates are for explanatory purposes, and weekdays / holidays and market opening / closing times are ignored (similar to FIG. 6 ).

[0055] Now, let us assume that the user has selected the value of the explanatory variable within the bold dashed line. This indicates, for example, the following: - The user wants to know the emissions for the following year, for example, on January 7, 2024. - As a prerequisite, the user wants to input the explanatory variables for January 7, 2021 into the trained emissions calculation model 39. For the explanation, we will return to Figure 4.

[0056] Thirdly, in step S301 to which the processing condition setting unit 21 returns, the processing condition setting unit 21 displays the explanatory variable reception screen 42 for estimating the emission credit unit price (FIG. 6) on the output device 13.

[0057] Fourth, the processing condition setting unit 21 accepts input (selection) of explanatory variables at a future target time point by a user (such as a business operator) via the input device 12. Although we are still in the middle of Fig. 4, the explanation will now move on to Fig. 6.

[0058] (Screen for accepting explanatory variables for estimating the unit price of emission credits) Figure 6 is an example of the screen 42 for accepting explanatory variables for estimating the unit price of emission credits. The screen 42 for accepting explanatory variables for estimating the unit price of emission credits displays meteorological observation values ​​and macroeconomic indicators in chronological order. These are obtained from the weather information 37 and business activity information 38, respectively. "#" indicates a different numerical value in an abbreviated form. For example, the user may select values ​​from different past dates for each explanatory variable (bold solid line), or may select values ​​from a common past date for all variables together (bold dashed line).

[0059] Now, let's assume that the user has selected the values ​​of the explanatory variables within the thick solid lines. This indicates, for example, the following: - The user wants to know the unit price of emissions credits for a certain day in early January 2024 of the following year. - As a prerequisite, the user wants to input weather observation values ​​for January 2, 2021 and macroeconomic indicators for January 3, 2021 into the trained emission credit unit price calculation model 40.

[0060] The above is an example in which the values ​​of explanatory variables for a future day are selected from past actual values. However, the processing condition setting unit 21 may also accept the user's selection of a future period spanning multiple days using a bold dashed line. Furthermore, at this time, the processing condition setting unit 21 may also accept the user's modification of past actual values. In either case, the processing condition setting unit 21 obtains the length of the target period in the future and the values ​​of each explanatory variable (average values ​​per day if the period spans multiple days). Return to FIG. 4 for further explanation.

[0061] In step S302, the processing condition setting unit 21 accepts an emission limit value. Specifically, the processing condition setting unit 21 accepts an emission limit value input by the user via the input device 12. The emission limit value (kg) corresponds to the greenhouse gas emission credits held (purchased) by the business operator. The business operator can emit greenhouse gases below the emission limit value at no additional cost.

[0062] In step S303, the emission amount estimation unit 23 of the information processing system 1 calculates an emission amount estimation value. Specifically, first, the emission amount estimation unit 23 acquires the machine-learned emission amount calculation model 39 from the auxiliary storage device 15. Second, the emission amount estimation unit 23 inputs the values ​​of the explanatory variables input in "second" of step 301 to the machine-learned emission amount calculation model 39, and acquires the emission amount estimation value output by the machine-learned emission amount calculation model 39 as the target variable.

[0063] In step S304, the emission credit unit price estimation unit 25 of the information processing system 1 calculates an estimated emission credit unit price. Specifically, first, the emission credit unit price estimation unit 25 acquires the machine-learned emission credit unit price calculation model 40 from the auxiliary storage device 15. Second, the emission credit unit price estimation unit 25 inputs the values ​​of the explanatory variables input in "fourth" of step 301 to the machine-learned emission credit unit price calculation model 40, and acquires the estimated emission credit unit price output as a target variable by the machine-learned emission credit unit price calculation model 40.

[0064] In step S305, the calculation unit 26 of the information processing system 1 determines whether the emission limit value is greater than the emission estimated value. Specifically, if the emission limit value accepted in step S302 is greater than the emission estimated value calculated in step S303 (step S305 "Yes"), the calculation unit 26 proceeds to step S306. Otherwise (step S305 "No"), the calculation unit 26 proceeds to step S308. Note that if "emission limit value -- = emission estimated value", the calculation unit 26 displays "the compensation is zero" on the output device 13 and then terminates the compensation calculation processing procedure.

[0065] In step S306, the calculation unit 26 calculates an emission limit margin value. Specifically, the calculation unit 26 subtracts the emission estimate value calculated in step S303 from the emission limit value accepted in step S302, and sets the result as the emission limit margin value.

[0066] In step S307, the calculation unit 26 calculates the value of the emission limit surplus value. Specifically, the calculation unit 26 multiplies the emission limit surplus value calculated in step S306 by the estimated emission credit unit price calculated in step S304, and determines the result as the value of the emission limit surplus value. The value of the emission limit surplus value is the income (sales amount) for the business operator. Then, the process proceeds to step S310.

[0067] In step S308, the calculation unit 26 calculates the emission limit shortfall value. Specifically, the calculation unit 26 subtracts the emission limit value received in step S302 from the emission amount estimated value calculated in step S303, and sets the result as the emission limit shortfall value.

[0068] In step S309, the calculation unit 26 calculates the compensation for the emission limit shortfall. Specifically, the calculation unit 26 multiplies the emission limit shortfall calculated in step S308 by the estimated emission credit unit price calculated in step S304, and determines the compensation as the result. The compensation here is an expenditure (purchase price) for the business operator. Then, the process proceeds to step S310.

[0069] In step S310, the calculation unit 26 displays the consideration. Specifically, the calculation unit 26 displays the consideration calculated in step S307 or S309 on the output device 13, along with a distinction between the selling price and the purchasing price. The calculation unit 26 may output the consideration to any other device via the communication device 16. Thereafter, the consideration calculation process procedure ends.

[0070] (Effects of this embodiment) (1) The information processing system enables a general business to calculate the compensation for the greenhouse gas emission limit surplus value and the compensation for the emission limit shortfall value. (2) The information processing system can display the compensation for the purchase or sale of emission rights. (3) The information processing system can learn an emission calculation model using information that a business manages on a daily basis. (4) The information processing system can learn an emission right unit price calculation model using information that a business manages on a daily basis.

[0071] The present invention is not limited to the present embodiment, and various modifications are included. For example, the present embodiment has been described in detail to clearly explain the present invention, and is not necessarily limited to an embodiment including all of the described configurations. Furthermore, it is possible to replace part of the configuration of the present embodiment with another configuration, and it is also possible to add another configuration to the configuration of the present embodiment. Furthermore, it is possible to add, delete, or replace part of the configuration of the present embodiment with another configuration.

[0072] REFERENCE SIGNS LIST 1 Information processing system 11 Central control unit 12 Input device 13 Output device 14 Main memory device 15 Auxiliary memory device 16 Communication device 21 Processing condition setting unit 22 Emission amount learning unit 23 Emission amount estimation unit 24 Emission allowance unit price learning unit 25 Emission allowance unit price estimation unit 26 Calculation unit 31 Emission amount database 32 Net emission amount information 33 Business activity emission amount information 33a Direct emission amount information 33b Indirect emission amount information 33c Other party emission amount information 35 Emission allowance unit price database 36 Emission allowance unit price information 37 Weather information 38 Business activity information 39 Emission amount calculation model 40 Emission allowance unit price calculation model 41 Explanatory variable acceptance screen for emission amount estimation 42 Explanatory variable acceptance screen for emission allowance unit price estimation

Claims

1. An emission learning unit that machine-learns an emission calculation model, which takes values ​​related to an enterprise's activities as input and outputs the amount of greenhouse gas emissions emitted by the enterprise, using past values ​​related to the enterprise's activities and the past values ​​of the amount of greenhouse gas emissions emitted by the enterprise as learning data; an emission credit unit price learning unit that machine-learns an emission credit unit price calculation model, which takes values ​​related to the environment of a market in which greenhouse gas emission credits are traded as input and outputs the unit price of greenhouse gas emission credits in the market, using past values ​​related to the environment of the market and the past values ​​of the unit price of the emission credits as learning data; a processing condition setting unit that accepts future values ​​related to the enterprise's activities, future values ​​related to the environment of the market, and emission limit values ​​corresponding to the greenhouse gas emission credits held by the enterprise; an emission estimation unit that inputs the future values ​​related to the enterprise's activities that have been accepted into the machine-learned emission calculation model, and obtains future emission estimates from the machine-learned emission calculation model; an emission rights unit price estimation unit that inputs future values ​​related to the accepted market environment into the machine-learned emission rights unit price calculation model and obtains an estimated value of the future emission rights unit price from the machine-learned emission rights unit price calculation model; and a calculation unit that, when the emission limit value is greater than the estimated emission value, calculates a consideration for the emission limit margin value by multiplying the emission rights unit price estimated value by an emission limit margin value obtained by subtracting the estimated emission value from the emission limit value, and, when the emission limit value is smaller than the estimated emission value, calculates a consideration for the emission limit shortfall value by multiplying the emission rights unit price estimated value by an emission limit shortfall value obtained by subtracting the emission limit value from the estimated emission value, and displays the calculated consideration.

2. The information processing system according to claim 1, wherein the calculation unit displays the value of the emission limit surplus value as a selling price, and displays the value of the emission limit shortfall value as a purchasing price.

3. The information processing system of claim 1, wherein the values ​​relating to the activities of the business operator include at least one of the following: direct emissions resulting from the business operator's greenhouse gas emissions; indirect emissions resulting from the business operator's consumption of energy supplied by others; third-party emissions resulting from the business operator's greenhouse gas emissions by others in the supply chain to which the business operator belongs; and greenhouse gas emissions reduced by the business operator's implementation of renewable energy and / or energy creation.

4. The information processing system according to claim 1, wherein the market environment value includes at least one of meteorological observation values ​​of the region to which the market belongs and macroeconomic indicators of the region to which the market belongs.

5. An emission learning unit of an information processing system: takes values ​​related to an enterprise's activities as input, and machine-learns an emission calculation model that outputs the amount of greenhouse gas emissions emitted by the enterprise, using past values ​​related to the enterprise's activities and past values ​​of the amount of greenhouse gas emissions emitted by the enterprise as learning data; an emission rights unit price learning unit of the information processing system: takes values ​​related to the environment of a market in which greenhouse gas emission rights are traded as input, and machine-learns an emission rights unit price calculation model that outputs the unit price of greenhouse gas emission rights in the market, using past values ​​related to the market environment and past values ​​of the emission rights unit price as learning data; a processing condition setting unit of the information processing system: accepts future values ​​related to the enterprise's activities, future values ​​related to the market environment, and an emission limit value corresponding to the greenhouse gas emission rights held by the enterprise; an emission estimation unit of the information processing system: inputs the future values ​​related to the enterprise's activities that have been accepted into the machine-learned emission calculation model, and obtains future emission estimates from the machine-learned emission calculation model; an emission rights unit price estimation unit of the information processing system: an emission credit unit price calculation model that has undergone machine learning to input future values ​​related to the accepted market environment, and obtain an estimated future emission credit unit price from the emission credit unit price calculation model that has undergone machine learning; a calculation unit of the information processing system calculates a consideration for the emission credit limit margin value by multiplying an emission credit unit price estimated value by an emission credit margin value obtained by subtracting the estimated emission credit from the emission credit limit value, when the emission credit limit value is greater than the estimated emission credit value; and calculates a consideration for the emission credit limit deficit value by multiplying an emission credit unit price estimated value by an emission credit margin value obtained by subtracting the estimated emission credit from the estimated emission credit value, when the emission credit limit value is smaller than the estimated emission credit value; and displays the calculated consideration.

Images