Measure specifying program, measure specifying method, and information processor

Abstract

To improve accuracy of measure analysis.SOLUTION: An information processor creates a digital twin which reproduces a state of a water area in the real world in a virtual space. The information processor executes simulation in the created digital twin. The information processor specifies a measure to be applied out of a plurality of measure candidates on the basis of a result of the executed simulation, and displays information regarding the specified measure on a display screen.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a program, a method, and an information processing device. [Background technology]

[0002] Policy analysis is required in a variety of situations, and simulations are used to analyze policies and identify optimal policies. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2023-182560 Summary of the Invention [Problem to be solved by the invention]

[0004] However, because conventional techniques perform simulations using preset calculation formulas, they can only obtain simulation results under specific conditions, and it is difficult to say that they can provide highly reproducible simulation data. Also, while it is possible to use the results of the simulation to formulate measures, because they depend on the results of the simulation, it is not possible to formulate measures that will be highly effective when introduced.

[0005] In one aspect, an object of the present invention is to provide a program, a method, and an information processing device that can improve the accuracy of policy analysis. [Means for solving the problem]

[0006] In the first proposal, the policy identification program is characterized by causing a computer to execute the following process: generate a digital twin that reproduces the state of a real-world water body in a virtual space; perform a simulation on the generated digital twin; identify a policy to be applied from among multiple candidate policies based on the results of the simulation; and display information about the identified policy on a display screen. [Effects of the Invention]

[0007] According to one embodiment, the accuracy of policy analysis can be improved. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram illustrating an example of the overall configuration of an optimal measure specifying system according to the first embodiment. [Figure 2] Figure 2 is a diagram explaining the definition of the ocean. [Figure 3] FIG. 3 is a functional block diagram of the information processing apparatus according to the first embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of the selection of development candidates. [Figure 5] FIG. 5 is a diagram illustrating an example of a construction pattern. [Figure 6] FIG. 6 is a diagram for explaining the estimation of the marine environment. [Figure 7] FIG. 7 is a diagram illustrating an example of the results of a simulation of the amount of algae. [Figure 8] FIG. 8 is a diagram illustrating a method for estimating factors. [Figure 9] FIG. 9 is a diagram illustrating a method for estimating factors. [Figure 10] FIG. 10 is a diagram illustrating the results of factor estimation. [Figure 11] FIG. 11 is a diagram illustrating the scoring results. [Figure 12] FIG. 12 is a diagram illustrating a display example. [Figure 13] FIG. 13 is a flowchart showing the overall processing flow. [Figure 14] FIG. 14 is a flowchart showing the flow of the simulation process. [Figure 15] FIG. 15 is a flowchart showing the flow of the policy evaluation process. [Figure 16] FIG. 16 is a diagram illustrating the flow of processing for identifying a policy to be applied. [Figure 17] FIG. 17 is a diagram illustrating another example of policy evaluation. [Figure 18] FIG. 18 is a diagram illustrating an example of a hardware configuration. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, embodiments of a program, a method, and an information processing device disclosed in the present application will be described in detail with reference to the accompanying drawings. However, the present invention is not limited to these embodiments. [Example]

[0010] (Overall composition) Fig. 1 is a diagram illustrating an example of the overall configuration of an optimal measure specifying system according to Example 1. As shown in Fig. 1, the optimal measure specifying system includes a plurality of sensors and an information processing device 10, and is a system that performs simulations related to, for example, the growth of seaweed and the creation of seaweed beds.

[0011] Each of the multiple sensors is a type of sensor used to collect sea area data (ocean data), such as an underwater drone equipped with a camera that captures images of underwater conditions, a light intensity sensor that measures underwater light intensity, a temperature sensor that measures water temperature, and a concentration sensor that measures salinity concentration.

[0012] The information processing device 10 acquires data on actual ocean areas, accurately measures the volume of seaweed and the density of seaweed beds where the seaweed grows, and performs modeling to reproduce ocean conditions. The information processing device 10 is an example of a computer that simulates the carbon dioxide (CO2) absorption effect and environmental impact of environmental measures such as the restoration and creation of seaweed beds, and formulates plans based on the simulation results.

[0013] The information processing device 10 generates a digital twin that reproduces the state of a water body in the real world in a virtual space, and performs a simulation using the generated digital twin. Next, the information processing device 10 identifies a policy to be applied from among multiple policy candidate candidates based on the results of the simulation. The information processing device 10 then displays information about the identified policy on a display screen.

[0014] More specifically, the information processing device 10 generates a digital twin that reproduces the state of a real-world ocean area in a virtual space, and in the generated digital twin, uses the state of the ocean area to perform a simulation of the state of the target to which the policy will be applied.The information processing device 10 then identifies the target policy from among multiple candidate policies based on the results of the simulation, and displays information about the identified target policy on the display screen.

[0015] For example, the information processing device 10 acquires ocean area data from multiple sensors, regional and ocean shapes from satellite data, and weather information from a weather server, and uses this information to generate a digital twin that reproduces virtual ocean conditions.

[0016] The information processing device 10 then runs a seaweed growth simulation on the digital twin for each of a plurality of seaweed cultivation patterns, which are examples of measures. Furthermore, the information processing device 10 runs an environmental simulation for the seaweed cultivation on the digital twin. The information processing device 10 then evaluates each of the seaweed cultivation patterns using the results of the environmental simulation, and presents the optimal seaweed cultivation pattern to the user.

[0017] In this way, the information processing device 10 can perform simulations of various measures that take into account changes in the environment and sea area on a digital twin that reproduces an actual sea area, thereby improving the accuracy of policy analysis.

[0018] (Terminology explanation) The definition and terminology of the ocean used in this embodiment will be explained. Fig. 2 is a diagram for explaining the definition of the ocean. Fig. 2 shows the ocean, and defines an area with seawater as a water body, an area on land with no seawater as a coast, an area above the sea surface as the ocean, an area with seawater from the sea surface to the ground with sand and rocks as the undersea, and an area with no seawater but ground or crust with sand and rocks as the seabed.

[0019] As examples of terms used in this embodiment, "space" refers to three-dimensional shapes and water depths, and "artificial objects" refers to buildings, sailing ships, etc. Furthermore, "measures" refer to patterns for areas in the ocean where seaweed beds should be created, and parameters within the measures include the species of seaweed and the time of year.

[0020] As examples of oceanographic data used in this embodiment, the environment refers to temperature, brightness, atmospheric pressure, water pressure, weight, pH, and precipitation, the organism refers to species of organisms and biomass, and the substance refers to nutrients, rock element content, gas concentration, etc. These measured oceanographic data are measured as time-series data.

[0021] (Functional configuration) 3 is a functional block diagram illustrating a functional configuration of the information processing device 10 according to Example 1. As shown in FIG.

[0022] The communication unit 11 is a processing unit that controls communication with other devices, and is realized by, for example, a communication interface, etc. For example, the communication unit 11 acquires oceanographic data from various sensors in time series.

[0023] The display unit 12 is a processing unit that displays and outputs various information, and is realized by, for example, a display, a touch panel, etc. For example, the display unit 12 outputs information about the optimal measure identified by the control unit 20, which will be described later.

[0024] The storage unit 13 is a processing unit that stores various data and programs executed by the control unit 20, and is realized by, for example, a memory or a hard disk. For example, the storage unit 13 stores the development patterns described below, programs and parameters used for the digital twin, ocean data acquired by the communication unit 11, machine learning models used by the control unit 20, and the like.

[0025] The control unit 20 is a processing unit that controls the entire information processing device 10, and is realized by, for example, a processor. This control unit 20 has a data integration infrastructure unit 30, an ocean model construction unit 40, and a policy determination unit 50. The data integration infrastructure unit 30, the ocean model construction unit 40, and the policy determination unit 50 are realized by electronic circuits executed by the processor, processes executed by the processor, etc.

[0026] The data integration platform unit 30 is a processing unit that generates a digital twin that reproduces the state of real-world ocean areas in a virtual space. For example, the data integration platform unit 30 adjusts time series data such as oceanographic data acquired from various sensors and weather information acquired from an external server, and uses these to generate a digital twin that is time-synchronized with the current ocean area. In addition, the data integration platform unit 30 can maintain a digital twin that tracks the ever-changing state of real-world ocean areas by continuously changing and modifying the digital twin using the time series data.

[0027] The marine model construction unit 40 has a creation pattern generation unit 41, a marine environment estimation unit 42, an algae amount simulation unit 43, and a CO2 amount estimation unit 44, and is a processing unit that uses the state of the sea area on the digital twin to perform a simulation of seaweed growth and seaweed bed creation.

[0028] The creation pattern generation unit 41 is a processing unit that generates creation patterns, which are measures. Specifically, the creation pattern generation unit 41 generates multiple creation patterns by receiving from the user or automatically selecting the location and area of ​​the seaweed bed to be created, the type of seaweed to be created, and the creation period.

[0029] Figure 4 is a diagram illustrating an example of selecting a development candidate. As shown in Figure 4, when the development pattern generation unit 41 receives a region selection from the user, it displays map data of the sea area including the candidate site to be set in that region. Then, when the development pattern generation unit 41 receives a selection of the development position and development range on the map data, it divides the water area range within the selected range into meshes and automatically sets the divided mesh areas as development candidate positions in a brute-force manner.

[0030] Here, the generated creation patterns will be explained. FIG. 5 is a diagram illustrating an example of a creation pattern. As shown in FIG. 5, the creation pattern generation unit 41 generates "sea area to be created, creation range, seaweed species, creation time" as a creation pattern. "Sea area to be created" indicates the sea area where seaweed will be created, "creation range" indicates the range where seaweed will be created, "seaweed species" indicates the type of seaweed to be created, and "creation time" indicates the time when it will be created. The example in FIG. 5 shows that a measure has been generated as "creation pattern A" to create "eelgrass" in "sea area A" within an area of ​​"1 hectare" in "March."

[0031] The marine environment estimation unit 42 is a processing unit that uses a digital twin to estimate various environmental information to be input into an algae growth model used in algae growth simulations. Specifically, the marine environment estimation unit 42 estimates future ocean conditions (temperature, brightness, air pressure, water pressure, weight, pH, precipitation) at the end of a pre-specified development period, based on current ocean conditions (seabed surface sediment and water depth corresponding to the location of the candidate seaweed bed). More specifically, the marine environment estimation unit 42 estimates time-series data on light intensity, water temperature, and nutrient concentration from the present to the future (for example, five years from now).

[0032] The marine environment estimation unit 42 can also realize future ocean conditions in the digital twin by estimating future ocean conditions using a machine learning model on the digital twin. FIG. 6 is a diagram illustrating marine environment estimation. As shown in FIG. 6(a), the marine environment estimation unit 42 can input current light intensity and temperature data into an estimation model on the digital twin to predict future water temperature data. As shown in FIG. 6(b), the marine environment estimation unit 42 can input light intensity and temperature data for a predetermined period, such as one month or one year, into the estimation model to estimate changes in water temperature data up to one year from now. Note that the estimation model shown here is merely an example, and the input and output can be changed depending on training methods, etc.

[0033] The algae mass simulation unit 43 is a processing unit that simulates the growth of seaweed in each creation pattern on a digital twin using the environmental information estimated by the marine environment estimation unit 42. For example, the algae mass simulation unit 43 inputs future ocean conditions (temperature, brightness, air pressure, water pressure, weight, pH, and precipitation) into a physical model (algae growth model) that models the growth of seaweed, and predicts the amount of algae biomass for a future time after the creation period.

[0034] In this way, the algae quantity simulation unit 43 performs an algae growth simulation for each cultivation pattern, and obtains information such as whether the algae to be cultivated will grow, how much the cultivation area of ​​the algae to be cultivated will expand or contract, and whether the size of the algae to be cultivated will increase or decrease. Note that an existing model can be used as the algae growth model, and there are no limitations on the model to be used.

[0035] The CO2 amount estimation unit 44 is a processing unit that estimates the amount of CO2 absorption for each artificial seaweed bed from the amount of algae biomass at a future time simulated by the algae amount simulation unit 43. For example, the CO2 amount estimation unit 44 estimates the amount of CO2 absorption by the artificial seaweed bed using the area of ​​the seaweed bed multiplied by the absorption coefficient per unit area.

[0036] The seaweed bed area is the area created when each creation pattern is generated or the area created by the seaweed at a future time included in the simulation results, and the absorption coefficient is a constant set for each type of seaweed. The calculation amount of CO2 absorption is not limited to "seaweed bed area x absorption coefficient per unit area," and other calculation formulas can be used, and different calculation formulas can be used for each type of seaweed.

[0037] FIG. 7 is a diagram illustrating an example of the results of a simulation of algae mass. As shown in FIG. 7, the algae mass simulation unit 43 and the CO2 amount estimation unit 44 calculate "growth, CO2 absorption, and creation cost" for each creation pattern through processing. Here, "growth" is information set by the algae mass simulation unit 43 and indicates whether the target seaweed will grow from the present to the end of the creation period, with "1" set if it will grow and "0" set if it will not grow. "CO2 absorption" indicates the CO2 absorption calculated by the CO2 amount estimation unit 44. "Creation cost" is information calculated by the algae mass simulation unit 43 or information set by the user, and is a cost that summarizes the expenses required for the measures (creation pattern) and the number of personnel required to implement the measures.

[0038] The policy determination unit 50 has a factor estimation unit 51, a surrounding environment evaluation unit 52, a policy evaluation unit 53, and a result output unit 54, and is a processing unit that executes evaluation of effective policies.

[0039] The factor estimation unit 51 is a processing unit that uses a classification model to extract common correlations between the simulation results (growth of seaweed beds = increased CO2 absorption) and each parameter that constitutes the measure, and identifies the causal relationships between the parameters using the extracted results and a causal discovery model.

[0040] Here, we briefly explain classification models. Figures 8 and 9 are diagrams explaining the factor estimation method. The classification model shown in Figure 8 is a model that has learned a combination of hypotheses and importance. Generally, deep learning improves accuracy by refining a single model by stacking multiple layers of neural networks that mimic the structure of the neural circuits in the human brain, resulting in a complex model that is incomprehensible to humans. On the other hand, as shown in Figure 8, a classification model is a highly accurate classification model that extracts a large number of hypotheses by combining data items, which are examples of attribute values, and adjusts the importance of the hypotheses (knowledge chunks (hereinafter sometimes simply referred to as "chunks")). A knowledge chunk is a simple model that humans can understand, describing hypotheses that may be valid as input / output relationships in a logical expression.

[0041] Specifically, the factor estimation unit 51 treats all combination patterns of data items in the input data as hypotheses (chunks) and determines the importance of each hypothesis based on the hit rate of the label for that hypothesis.The factor estimation unit 51 then constructs a model based on the extracted multiple knowledge chunks and labels (objective variables).At this time, the factor estimation unit 51 controls the importance to be lowered when the items that make up a knowledge chunk overlap frequently with the items that make up other knowledge chunks.

[0042] A specific example will be explained using Figure 9. Here, we consider an example where we want to determine customers who will purchase a certain product or service. Customer data contains various items (attribute values), such as "gender," "driver's license status," "marriage status," "age," and "annual income." All combinations of these items are considered as hypotheses, and the importance of each hypothesis is considered. For example, there are 10 customers in the data who fit the hypothesis combining the items "male," "owner," and "married." If 9 out of these 10 customers have purchased a product, etc., the hypothesis with a high hit rate is "People who are male, own, or married purchase," and this is extracted as a knowledge chunk. Note that here, as an example, the label, or the objective variable, is a binary representation of whether or not the product was purchased.

[0043] On the other hand, there are 100 customers in the data who fit the hypothesis that combines the items "male" and "ownership." If only 60 of these 100 people have purchased a product, the hit rate for purchase will be 60%, which is less than the threshold (e.g., 80), so the hypothesis with a low hit rate, "People who are male and own purchase," will not be extracted as a knowledge chunk.

[0044] Furthermore, there are 20 customers in the data who fit the hypothesis that combines the items "male," "does not own," and "unmarried." If 18 of these 20 have not purchased a product, the hit rate for non-purchasers is 90%, which is above a threshold (e.g., 80), so the hypothesis with a high hit rate, "People who are male, do not own, and are unmarried have not purchased," is extracted as a knowledge chunk.

[0045] In this way, the factor estimation unit 51 extracts tens or hundreds of millions of knowledge chunks that support purchase and knowledge chunks that support non-purchase, and performs model learning. The model learned in this way lists combinations of features as hypotheses (chunks), and each hypothesis is assigned an importance, which is an example of likelihood indicating plausibility. The sum of the importance of hypotheses that appear in the input data becomes a score, and if the score is equal to or greater than a threshold, it is output as a positive example.

[0046] In other words, the score is an index that indicates the likelihood of a state, and is the sum of the importance values ​​of chunks (hypotheses) generated by each model that satisfy all of the associated features. For example, suppose chunk A is associated with "Importance: 20, Features (A1, A2)," chunk B with "Importance: 5, Features (B1)," and chunk C with "Importance: 10, Features (C1, C2)," and the data item to be evaluated contains (A1, A2, B1, C1). In this case, all of the features of chunk A and chunk B appear, so the score is "20 + 5 = 25." The features here refer to user behavior, etc.

[0047] As described above, the factor estimation unit 51 identifies a causal relationship from the hypotheses extracted in this manner using a causal discovery model. Specifically, the information processing device 10 identifies a causal relationship between a combination (hypothesis) of explanatory variables and a target variable. For example, the factor estimation unit 51 uses a technique for analyzing which factor is the cause of a result by analyzing the mutual influence of two factors when they change, thereby comprehensively checking combinations of feature quantities and extracting causal relationships for each condition group.

[0048] More specifically, the factor estimation unit 51 generates a causal relationship between each item constituting a measure and the objective variable by extracting the degree of influence of each combination (hypothesis) on each condition set as the objective variable. For example, the factor estimation unit 51 sets, among the extracted combinations, a "combination that has a large influence on the objective variable" as a grouping rule for the original data. Then, by analyzing the causal relationships within a specific group, the factor estimation unit 51 individually extracts causal relationships that would be invisible when viewed as a whole because multiple causal relationships are mixed and cancel each other out.

[0049] FIG. 10 is a diagram illustrating the results of factor estimation. When the above technology is applied to this embodiment, as shown in FIG. 10, a hypothesis, which is a combination of features, is a combination of each item constituting a measure, such as "algae species, construction time," "algae species, construction time, growth," "constructed sea area, algae species," and "constructed sea area, construction range, algae species." The factor estimation unit 51 then identifies a causal relationship between each combination of "algae species, construction time," "algae species, construction time, growth," "constructed sea area, algae species," and "constructed sea area, construction range, algae species," and the objective variable "CO2 absorption amount." The example in FIG. 10 shows that "constructed sea area, algae species, construction time" has a high correlation with "CO2 absorption amount."

[0050] The factor estimation unit 51 then displays the identified causal relationships on the display unit 12 or the like. The causal relationships identified in this way are used to determine measures and to determine whether to change the sea area. For example, if none of the combinations has an effect on the objective variable "CO2 absorption amount," it may be possible to change the sea area. Also, if it is found that the "sea area" has a significant effect on the objective variable "CO2 absorption amount," it may be possible to narrow down the measures to that "sea area" and reconsider them.

[0051] Returning to Figure 3, the surrounding environment assessment unit 52 is a processing unit that assesses the environmental value of the sea area surrounding the seaweed bed creation location for each creation pattern (measure) through simulation using a digital twin. Specifically, the surrounding environment assessment unit 52 performs an environmental simulation from the time of creation to the time of completion of creation (future point in time) for promising measures that are expected to increase CO2 absorption, where the "CO2 absorption amount" is equal to or greater than the threshold value according to the CO2 amount estimation unit 44, and assigns an environmental value score.

[0052] For example, the surrounding environment assessment unit 52 uses the digital twin to perform an environmental simulation to calculate a score for the current state index of the marine health index, and identifies the environmental value of the sea area surrounding the seaweed bed creation location at a future time when each measure (creation pattern) is implemented. More specifically, the surrounding environment assessment unit 52 evaluates each of the following items as environmental value: food, water quality, coastal protection, tourism, economy, marine products, image of the place, and biodiversity.

[0053] Here, "food" refers to the degree of increase in fish catch in the target sea area, "water quality" refers to the degree of improvement in water quality in the target sea area, "coastal protection" refers to the degree of protection efforts in the target sea area, "tourism" refers to the degree to which the target sea area is evaluated as a tourist destination, "marine products" refers to the degree of increase in marine products in the target sea area, "place image" refers to the degree to which the favorable public image of the target sea area increases, and "biodiversity" refers to the degree of increase in biological species in the target sea area.

[0054] The measure evaluation unit 53 is a processing unit that calculates scores for the amount of CO2 absorbed, the cost of development, and the environmental value for each measure (development pattern), evaluates each measure, and identifies the optimal measure. Specifically, the measure evaluation unit 53 calculates scores for the simulation results (see, for example, FIG. 7) by the algae quantity simulation unit 43.

[0055] For example, the policy evaluation unit 53 normalizes the estimated CO2 absorption amount as a score of the CO2 absorption amount so that the maximum value of the CO2 absorption amount for each policy to be compared becomes a constant value. For example, the maximum score is defined as 500 points.

[0056] The measure evaluation unit 53 also calculates the cost of creating a seaweed bed by multiplying the distance from the coast by the area of ​​the seaweed bed by the type of seaweed and the reciprocal of a coefficient set according to the installation method, and normalizes the calculation result so that the maximum creation cost for each measure being compared is a constant value. Note that the higher the creation cost, the smaller the score value.

[0057] Furthermore, the policy evaluation unit 53 uses, as the score of the environmental value, the "score for the current status index of the marine health index" calculated by the surrounding environment evaluation unit 52. Note that the items "possibility of artisanal fisheries," which are difficult to calculate, and "carbon stocks," which are originally included in the score, are excluded from the score.

[0058] For example, the policy evaluation unit 53 calculates and scores the environmental value points for policies "A, D, E, G" among the policies "A, B, C, D, etc." shown in Figure 7, for which the "CO2 absorption amount" is the threshold value according to the CO2 amount estimation unit 44.

[0059] FIG. 11 is a diagram illustrating the scoring results. As shown in FIG. 11, the policy evaluation unit 53 calculates, for each policy, "CO2 absorption amount, development cost, environmental value (food, water quality, coastal protection, tourism, economy, marine products, image of the place, biodiversity), total." The higher the value of each evaluation item, the higher the evaluation. In the example of FIG. 11, policy "A" is evaluated as "CO2 absorption amount = 100, development cost = 100, environmental value (food = 2, water quality = 5, coastal protection = 1, tourism = 0, economy = 2, marine products = 4, image of the place = 8, biodiversity = 10)," with the total evaluation value being "232."

[0060] The policy evaluation unit 53 identifies policy "E" as the optimal policy, which has the highest score among the total score of policy "A" (232), policy "D" (335), policy "E" (364), and policy "G" (186) shown in Figure 11.

[0061] The result output unit 54 is a processing unit that outputs the results evaluated by the policy evaluation unit 53 to the display unit 12. For example, the result output unit 54 can output all evaluated policies, or can output only the identified optimal policy. The result output unit 54 can also display the output results of the causal relationship model, etc.

[0062] Fig. 12 is a diagram illustrating a display example. As shown in Fig. 12, the result output unit 54 displays, as the result of the policy simulation, the evaluation results of promising policies that are expected to increase CO2 absorption, evaluated by the policy evaluation unit 53. The result output unit 54 also outputs the evaluation results of the optimal policy "E" by associating the development area of ​​the optimal policy "E" with map data of the sea area designated as the policy target. Note that the display example is merely an example, and the display items, etc. can be changed as desired.

[0063] (Overall processing flow) Fig. 13 is a flowchart showing the overall processing flow. As shown in Fig. 13, when the information processing device 10 starts processing in response to a user's instruction operation or the like, it sets various conditions related to the creation of seaweed beds and generates creation patterns (measures) (S101). Meanwhile, the information processing device 10 acquires ocean data (S102) and generates a digital twin that reproduces the ocean in the real world (S103).

[0064] Then, the information processing device 10 executes a seaweed growth simulation for each measure (S104), and estimates the amount of CO2 absorption using the simulation results (S105).

[0065] Thereafter, the information processing device 10 performs factor estimation using a classification model, a causal discovery model, etc., and performs an environmental evaluation simulation on the digital twin to evaluate the environmental information (S106).The information processing device 10 then performs an evaluation of each measure and determines the optimal evaluation (S107).

[0066] (Simulation process flow) 14 is a flowchart showing the flow of the simulation process, which is a detailed explanation of S101 to S105 in FIG.

[0067] As shown in FIG. 14, the information processing device 10 receives the selection of the sea area and region for creating a seaweed bed, the mesh precision, and the creation period (S201), and determines the seaweed bed creation pattern as a measure (S202).

[0068] The information processing device 10 then executes steps S203 to S207 for all of the creation patterns. Specifically, the information processing device 10 loads an algae growth model and parameters for the specified sea area (S204), and estimates the amount of light, water temperature, and nutrients from the present until the end of the creation period (S205). The information processing device 10 then estimates the amount of biomass from the algae growth model for each creation pattern (S206), and estimates the amount of CO2 absorption from the amount of biomass (S207).

[0069] Thereafter, when the information processing device 10 completes the estimation of the CO2 absorption amount for all the development patterns (S203: Yes), it executes the process of identifying the optimal measure shown in FIG. 15 (S208).

[0070] (Flow of policy evaluation process) 15 is a flowchart showing the flow of the policy evaluation process, which is a detailed explanation of S107 in FIG.

[0071] As shown in Figure 15, when the information processing device 10 has finished estimating the amount of CO2 absorption for all creation patterns (S301), it inputs the results of the seaweed bed creation simulation (creation pattern, algae species, period) into a causal discovery model (S302), and extracts promising measures that are likely to absorb CO2 from the results of the seaweed bed creation simulation (S303).

[0072] Then, the information processing device 10 executes the processes from S304 to S307 for all promising measures. Specifically, the information processing device 10 executes an environmental simulation at the time of development (S305), scores the environmental value (S306), and estimates the development cost from the range of the development site, etc. (S307). The cost may be estimated when the development pattern is generated, or can be calculated at any timing.

[0073] The information processing device 10 then scores the CO2 absorption amount, creation cost, and environmental value of all measures (S308), and displays each score, the cause and effect of the creation pattern, algae species, and period, and recommends the measures with the highest scores to the user (S309). As a result, the user makes a decision as to whether or not to implement the recommended measures (S310).

[0074] (effect) As described above, the information processing device 10 generates a digital twin that reproduces real-world ocean conditions in a virtual space, performs a simulation of seaweed growth and seaweed bed development using the ocean conditions in the generated digital twin, and identifies the optimal measure from among multiple candidate measures based on the results of the simulation. As a result, the information processing device 10 is not limited to specific conditions, but can reproduce and simulate various conditions in the digital twin, thereby improving the accuracy of measure analysis.

[0075] Furthermore, the information processing device 10 can identify which factors affect the amount of CO2 absorption, and therefore can encourage the user to create efficient measures. Furthermore, the information processing device 10 can identify optimal measures from multiple perspectives, not just CO2 absorption, and can therefore recommend measures for each priority item, such as environmental priorities or cost priorities, and can propose optimal measures according to the user's requests.

[0076] Furthermore, the information processing device 10 can reproduce not only actual oceanographic data but also predicted future oceanographic data in a digital twin, thereby improving the accuracy of measure selection. Furthermore, the information processing device 10 can reproduce conditions that are not currently anticipated in a digital twin, making it possible to identify optimal measures under various conditions and to identify optimal measures that correspond to various patterns, such as situations anticipated by the user and unexpected situations. [Example]

[0077] Here, the processing flow of the information processing device 10 according to Example 2 will be described. In the marine digital twin, the information processing device 10 can identify measures to be applied to objects located in an area by analyzing the ocean state over time in the area where the objects to which the measures are applied are located. In addition, various measures using marine data can be targeted as candidate measures.

[0078] Fig. 16 is a diagram illustrating the flow of a process for identifying a target measure. As shown in Fig. 16, first, the information processing device 10 generates an ocean digital twin. For example, the information processing device 10 uses ocean data measured by multiple sensors present in the ocean to generate a digital twin that reproduces ocean conditions in the real world in a virtual space. More specifically, the control unit 20 of the information processing device 10 uses the ocean data to generate a digital twin in which the real world and the virtual space are time-synchronized and that reproduces ocean conditions for each ocean location (S401).

[0079] Next, the information processing device 10 accepts a plurality of candidate measures and an area to which the measures will be applied. For example, the control unit 20 of the information processing device 10 accepts information on types of seaweed as a plurality of candidate measures. Furthermore, for example, the control unit 20 of the information processing device 10 accepts an area to develop a seaweed bed as an area to which the measures will be applied (S402).

[0080] The information processing device 10 then performs a simulation on the ocean digital twin. For example, the information processing device 10 performs a simulation of ocean conditions in an area where an object to which a measure is to be applied is located on the digital twin in which the ocean conditions are reproduced. More specifically, the control unit 20 of the information processing device 10 performs a time-series simulation of ocean conditions in an area where a seaweed bed is to be developed (S403).

[0081] Next, the information processing device 10 predicts the state of the target object for each of the multiple candidate measures. For example, the information processing device 10 predicts the growth state of seaweed for each of the multiple seaweed types using the simulated ocean conditions. More specifically, the control unit 20 of the information processing device 10 acquires a machine learning model trained, for example, using the ocean conditions as training data and the growth state of seaweed as ground truth data. Then, the control unit 20 of the information processing device 10 predicts the growth state of seaweed for each seaweed type by, for example, inputting the simulated ocean conditions into the machine learning model trained for each seaweed type (S404).

[0082] The information processing device 10 then identifies a measure to be applied from among multiple candidate measures. The information processing device 10, for example, identifies a measure to be applied to an object located in an area to which the measure will be applied. The information processing device 10, for example, identifies a measure in which the predicted state of the object satisfies a predetermined condition as the measure to be applied. More specifically, the control unit 20 of the information processing device 10, for example, identifies a type of seaweed that will form a seaweed bed of a predetermined area or more after growth in an area where the seaweed bed is to be developed. The control unit 20 of the information processing device 10 then, for example, sets the identified type of seaweed as the type of seaweed with the best growth condition (S405).

[0083] Next, the information processing device 10 displays information about the measures to be applied on the display screen. For example, the control unit 20 of the information processing device 10 displays information about the type of seaweed that is growing best in the area where the seaweed bed is to be created on the display screen (S406). As a result, the information processing device 10 is not limited to specific conditions, and can reproduce and simulate various conditions on the ocean digital twin, making it possible to identify the measures to be applied from among multiple candidate measures. [Example]

[0084] Although the embodiments of the present invention have been described above, the present invention may be embodied in various different forms other than the above-described embodiments.

[0085] For example, the information processing device 10 can verify ocean-related policies by simulating changes in the ocean's environment and the growth of organisms in the ocean digital twin. Furthermore, for example, the information processing device 10 can calculate the amount of carbon dioxide absorption to verify blue carbon-related policies. Furthermore, for example, the information processing device 10 can incorporate biological knowledge into the ocean digital twin to perform a time-series simulation of the growth of marine organisms. This allows for a simulation in digital space of the transition in carbon dioxide absorption when conservation measures for a certain marine ecosystem are taken, including the growth status of marine organisms, to verify in advance the effectiveness of global warming countermeasures.

[0086] (Numbers, etc.) The numerical values, example measures, scores, etc. used in the above examples are merely examples and can be changed as desired. Furthermore, the process flow described in each flowchart can also be changed as appropriate within a consistent range.

[0087] (Digital Twin) In the above embodiment, an example of predicting oceanographic data using a machine learning model on a digital twin has been described, but the present invention is not limited to this. For example, the information processing device 10 can directly set parameters (such as light intensity, water temperature, and nutrients) that indicate the state that the user expects or the state that the user wants to simulate in the digital twin, reproduce that state in the digital twin, and then perform a simulation.

[0088] (Another example of measures) In the above embodiment, an example of simulating seaweed growth patterns as a measure has been described, but the present invention is not limited to this and various measures using ocean data can be targeted. For example, the information processing device 10 can also set maintenance items for artificial objects such as offshore wind turbines used for power generation, buildings, and structures as measures.

[0089] Fig. 17 is a diagram illustrating another example of policy evaluation. For example, the information processing device 10 can simulate, using a digital twin, the state of an offshore wind turbine X years from now when maintenance according to policy A is performed on the offshore wind turbine (see (a) of Fig. 17), the state of the offshore wind turbine X years from now when maintenance according to policy B is performed on the offshore wind turbine (see (b) of Fig. 17), and the state of the offshore wind turbine X years from now when maintenance according to policy C is performed on the offshore wind turbine (see (c) of Fig. 17), and evaluate the results.

[0090] (Target of measures) In the above embodiment, the explanation was given using an example of a sea area, but the present invention is not limited to this and can be applied to freshwater bodies of water as well. Furthermore, the same processing can be performed on freshwater algae by using the growth model of freshwater algae, not just seaweed.

[0091] (system) The information including the processing procedures, control procedures, specific names, various data and parameters shown in the above documents and drawings may be changed arbitrarily unless otherwise specified.

[0092] Furthermore, the specific form of distribution and integration of the components of each device is not limited to that shown in the figure. For example, the ocean model construction unit 40 and the policy assessment unit 50 may be integrated. In other words, all or part of the components may be functionally or physically distributed or integrated in any unit depending on various loads, usage conditions, etc. Furthermore, all or any part of the processing functions of each device may be realized by a CPU and a program analyzed and executed by the CPU, or may be realized as hardware using wired logic.

[0093] Furthermore, all or any part of the processing functions performed by each device may be realized by a CPU and a program analyzed and executed by the CPU, or may be realized as hardware using wired logic.

[0094] (Hardware) Fig. 18 is a diagram illustrating an example of a hardware configuration. Here, an information processing device 10 will be described as an example. As shown in Fig. 18, the information processing device 10 includes a communication device 10a, an HDD (Hard Disk Drive) 10b, a memory 10c, and a processor 10d. The components illustrated in Fig. 18 are connected to each other via a bus or the like.

[0095] The communication device 10a is a network interface card or the like, and communicates with other devices. The HDD 10b stores programs and DBs that operate the functions shown in FIG.

[0096] The processor 10d reads out a program that executes the same processing as each processing unit shown in Fig. 3 from the HDD 10b, etc., and expands it into the memory 10c, thereby operating a process that executes each function described in Fig. 3, etc. For example, this process executes the same functions as each processing unit possessed by the information processing device 10. Specifically, the processor 10d reads out a program that has the same functions as the data integration infrastructure unit 30, the ocean model construction unit 40, the policy determination unit 50, etc., from the HDD 10b, etc. Then, the processor 10d executes a process that executes the same processing as the data integration infrastructure unit 30, the ocean model construction unit 40, the policy determination unit 50, etc.

[0097] In this way, the information processing device 10 operates as an information processing device that executes a policy identification method by reading and executing a program. The information processing device 10 can also realize functions similar to those of the above-described embodiment by reading the program from a recording medium using a medium reading device and executing the read program. Note that the program in these other embodiments is not limited to being executed by the information processing device 10. For example, the above-described embodiment may also be applied in the same way to cases where another computer or server executes the program, or where these execute the program in cooperation with each other.

[0098] This program may be distributed via a network such as the Internet. Alternatively, this program may be recorded on a computer-readable recording medium such as a hard disk, a flexible disk (FD), a CD-ROM, a magneto-optical disk (MO), or a digital versatile disk (DVD), and may be read out from the recording medium and executed by a computer. [Explanation of symbols]

[0099] 10. Information processing equipment 11 Communications Department 12 Display section 13 Storage section 20 Control Unit 30 Data Integration Platform Department 40 Ocean Model Construction Department 41 Construction pattern generation unit 42 Marine Environment Estimation Department 43 Algae quantity simulation section 44 CO2 amount estimation section 50 Policy Judgment Department 51 Factor Estimation Section 52 Surrounding Environment Assessment Department 53 Policy Evaluation Department 54 Result output section

Claims

1. On the computer, We create a digital twin that recreates the state of real-world water bodies in a virtual space. A simulation is performed on the generated digital twin, Identifying a target measure to be applied from among a plurality of candidate measures based on the results of the simulation that was carried out; Displaying information about the identified measures on a display screen. A measure specifying program characterized by executing a process.

2. The generating process includes: A digital twin is generated that reproduces the real-world ocean conditions in a virtual space. The identifying process includes: In the digital twin in which the oceanographic conditions are reproduced, the oceanographic conditions are analyzed over time in the area in which the object to which the policy is to be applied is located, thereby identifying the policy to be applied to the object located in the area as the policy to be applied.

2. The measure specifying program according to claim 1.

3. The generating process includes: A digital twin is generated that reproduces the real-world ocean conditions in a virtual space. The process to be performed is In the digital twin in which the oceanographic conditions are reproduced, a simulation of the oceanographic conditions in the area in which the object to which the measures are applied is located is carried out; The identifying process includes: identifying a measure to be applied to an object located in the area based on the results of the simulation of the ocean state; 2. The measure specifying program according to claim 1.

4. The process to be performed is In the digital twin in which the oceanographic state is reproduced, an estimation model is used to simulate the oceanographic state in an area where the object to which the measure is applied is located; and The identifying process includes: predicting the state of the object for each of a plurality of candidate measures based on the results of the simulation of the ocean state, and identifying a measure that satisfies a predetermined condition when the predicted state of the object is to be applied to the object located in the area; 4. The measure specifying program according to claim 3.

5. The generating process includes: Using a digital twin in which the virtual space and the real world are time-synchronized, light intensity, water temperature, and nutrient concentration are estimated from the current ocean state as a pre-specified future ocean state; The process to be performed is performing the simulation using the light intensity, the water temperature, and the nutrient concentration estimated as the future ocean state; 5. The measure specifying program according to claim 4.

6. The generating process includes: inputting information about current ocean conditions into a machine learning model that has been trained on the digital twin to obtain information about the future ocean conditions, thereby realizing the future ocean conditions in the digital twin; The identifying process includes: identifying a measure to be applied to the object from among the plurality of candidate measures as the measure to be applied based on the result of the simulation of the future ocean state; 6. The measure specifying program according to claim 5.

7. The generating process includes: A digital twin is generated that reproduces the real-world ocean conditions in a virtual space. The process to be performed is Using the digital twin, a simulation is performed of the growth of seaweed cultivated in the ocean area reproduced in the digital twin.

2. The measure specifying program according to claim 1.

8. The above measures are: Each creation pattern defines the sea area where seaweed is created, the range of seaweed creation, the algae species to be created, and the creation period for seaweed creation. The generating process includes: Generate the digital twin that reproduces the developed sea area specified in each development pattern, The process to be performed is Using the digital twin, the growth of the algae species in each of the development patterns is simulated.

8. The measure specifying program according to claim 7, wherein:

9. The process to be performed is Using the digital twin, simulate the growth of the algae species in each of the development patterns; Using the results of the simulation, estimate the amount of carbon dioxide absorbed by the algae species in each of the development patterns.

9. The measure specifying program according to claim 8.

10. The process to be performed is Calculating the environmental value when each of the development patterns is executed using the digital twin; assessing each of the seaweed growth, the environmental value, and the cost of the development pattern; The identifying process includes: Identifying the development pattern with the best evaluation results; 9. The measure specifying program according to claim 8.

11. The process to be performed is The carbon dioxide absorption amount of each of the development patterns is normalized so that the maximum value of the carbon dioxide absorption amount of each of the development patterns to be compared becomes a constant value, and is scored; Each item included in the marine health index is scored as the environmental value of each of the development patterns, The cost of each of the construction patterns is scored using coefficients set according to the distance from the coast, the area of ​​the seaweed bed, the type of seaweed, and the installation method. The identifying process includes: Identifying the development pattern with the highest total score for the carbon dioxide absorption amount, the score for the environmental value, and the score for the cost as the optimal measure; 10. The measure specifying program according to claim 9.

12. The process to be performed is Using a trained causal discovery model, from the items and combinations of items defined in each of the development patterns, identify the items and combinations of items that have a causal relationship with the amount of carbon dioxide absorption; 10. The measure specifying program according to claim 9.

13. The computer We create a digital twin that recreates the state of real-world water bodies in a virtual space. A simulation is performed on the generated digital twin, Identifying a target measure to be applied from among a plurality of candidate measures based on the results of the simulation that was carried out; Displaying information about the identified measures on a display screen. A method for identifying a measure, comprising: executing a process.

14. We create a digital twin that recreates the state of real-world water bodies in a virtual space. A simulation is performed on the generated digital twin, Identifying a target measure to be applied from among a plurality of candidate measures based on the results of the simulation that was carried out; Displaying information about the identified measures on a display screen. An information processing device comprising a control unit.

Images