Determination method, determination program, and determination device

Abstract

To determine whether a separation of estimated results and actual results is not caused by a measure response model.SOLUTION: A determination method causes a computer to perform processing to: set, based upon a measure taking zone, a measure influence region in which a measure response model having a response of an agent defined to a measure taking zone for which taking of measures is set in an area virtualized as a simulation object exerts an influence; and determine that a separation between a simulation result and an actual measurement result is not caused by the measure response model when the difference between the simulation result in application of the measure taking model and the actual measurement result in taking of measures corresponding to the measure taking zone satisfies a specific condition.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present invention relates to a determination method, a determination program, and a determination apparatus. [Background technology]

[0002] Simulation techniques are known for modeling social phenomena and analyzing or predicting their behavior. In such simulation techniques, the following types of policy response models may be applied from the perspective of solving social problems.

[0003] For example, in a traffic simulation, areas where measures such as road pricing, lane control, and school zones are implemented are defined, and a policy response model that specifies the response to those areas is applied to the agent's behavior. [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2021-125033 [Overview of the project] [Problems that the invention aims to solve]

[0005] However, when a discrepancy occurs between the predicted results of the above simulation and reality, it is difficult to determine whether the cause lies in the above policy response model or elsewhere.

[0006] In one aspect, the present invention aims to provide a determination method, determination program, and determination device that can determine whether or not the cause of the discrepancy between the budget and actual results lies outside of the policy response model. [Means for solving the problem]

[0007] In a determination method relating to one aspect, a policy influence area is defined based on the policy implementation zone, where a policy response model, which specifies the agent's response to a policy implementation zone where a policy is set to be implemented within a virtualized area used as a simulation target, has an influence. If, in a divided area of ​​the area, the area is divided into a divided area outside the policy influence area, and the difference between the simulation result when the policy response model is applied and the actual measurement result when the policy is implemented corresponding to the policy implementation zone satisfies certain conditions, the computer executes a process to determine that the cause of the discrepancy between the simulation result and the actual measurement result is something other than the policy response model. [Effects of the Invention]

[0008] This allows us to determine whether the discrepancy between the budget and actual results is due to factors other than the policy response model. [Brief explanation of the drawing]

[0009] [Figure 1] Figure 1 is a block diagram showing an example of the functional configuration of a server device. [Figure 2] Figure 2 is a schematic diagram illustrating the factors contributing to the discrepancy in traffic volume. [Figure 3] Figure 3 is a schematic diagram illustrating an example of the application of the policy response model. [Figure 4] Figure 4 is a schematic diagram showing an example of actual traffic flow. [Figure 5] Figure 5 shows an example of the area of ​​influence of the policy. [Figure 6] Figure 6 shows an example of a divided region. [Figure 7] Figure 7 is a schematic diagram showing an example of traffic volume without any intervention measures. [Figure 8] Figure 8 is a schematic diagram showing an example of traffic volume with a policy in place. [Figure 9] Figure 9 is a schematic diagram illustrating an example of the difference in traffic volume with and without the implementation of a policy. [Figure 10] Figure 10 is a schematic diagram showing an example of actual traffic volume. [Figure 11] Figure 11 is a schematic diagram illustrating an example of the difference between planned and actual traffic volume. [Figure 12] Figure 12 shows an example of a factor causing the discrepancy. [Figure 13] Figure 13 is a schematic diagram showing an example of actual traffic volume. [Figure 14] Figure 14 is a schematic diagram illustrating an example of the difference between planned and actual traffic volume. [Figure 15] Figure 15 shows another example of a factor causing the discrepancy. [Figure 16] Figure 16 is a schematic diagram showing an example of actual traffic volume. [Figure 17] Figure 17 is a schematic diagram illustrating an example of the difference between planned and actual traffic volume. [Figure 18] Figure 18 is a flowchart (1) showing the procedure for the judgment process. [Figure 19] Figure 19 is a flowchart (2) showing the procedure for the judgment process. [Figure 20] Figure 20 shows an example of a hardware configuration. [Modes for carrying out the invention]

[0010] The following describes embodiments of the determination method, determination program, and determination apparatus according to the present application with reference to the attached drawings. Each embodiment is merely an example or aspect, and such examples do not limit the range of numerical values ​​or functions, or the usage scenarios. Furthermore, each embodiment can be appropriately combined as long as the processing content is not contradictory. [Examples]

[0011] <Overall Structure> Figure 1 is a block diagram showing an example of the functional configuration of the server device 10. The server device 10 shown in Figure 1 provides a simulation function that models social phenomena and analyzes or predicts their behavior and interactions.

[0012] One example of such a simulation is a micro-traffic simulation that simulates the behavior and interactions of vehicle agents placed in a virtual space where a road network within an area of ​​arbitrary size is virtualized. However, this is not the only example.

[0013] This is merely an example and does not preclude the implementation of other simulations besides micro-traffic simulations, such as urban planning, social simulations, traffic simulations, data assimilation, and social digital twins.

[0014] Server device 10 is an example of a computer that provides the above-described simulation function. In one embodiment, server device 10 can be implemented as a server that provides the above-described simulation function on-premises. Alternatively, server device 10 can also be implemented as a PaaS (Platform as a Service) or SaaS (Software as a Service) application to provide the above-described simulation function as a cloud service.

[0015] As shown in Figure 1, the server device 10 can be connected to the sensors 20 and client terminals 30 via a network NW so as to be able to communicate with them. For example, the network NW can be any type of communication network, such as the Internet or a LAN (Local Area Network), whether wired or wireless. Although Figure 1 shows an example where one sensor 20 and one client terminal 30 are connected to one server device 10, any number of sensors 20 and client terminals 30 may be connected.

[0016] The sensors 20 can be used to sense the input data for the above simulation. For example, in the case of a micro-traffic simulation, examples of sensors 20 include imaging sensors and position detection sensors.

[0017] As just one example, Origin-Destination (OD) data, used in a route selection model that defines one of the agent's behaviors, can be generated from images sensed by imaging sensors placed on a road network. In this case, individual vehicles can be identified from images sensed by the imaging sensors using license plates, etc., and by tracking the vehicles chronologically across images, the trip defined by the origin O and destination D can be estimated. OD data is generated by aggregating the number of trips for each such combination of origin-destination OD. Here, an imaging sensor is given as just one example of the sensors 20, but other sensors may be used. For example, OD data can also be generated from time-series data of location information detected by position detection sensors such as GPS (Global Positioning System) receivers mounted on in-vehicle devices such as digital tachographs and drive recorders.

[0018] The client terminal 30 corresponds to an example of a computer that receives the above-described simulation function. For example, the client terminal 30 may be implemented as a desktop or laptop personal computer. This is merely an example, and the client terminal 30 may be any computer, such as a mobile terminal or a wearable device.

[0019] Figure 1 shows an example of the above simulation function being provided in a client-server system, but this is merely one example, and the above simulation function may also be provided as a standalone service.

[0020] <One aspect of the problem> As explained in the background technology section above, when a discrepancy occurs between the predicted results of the above simulation and reality, it is difficult to determine whether the cause lies in the policy response model, which defines the response to the policy implementation area, or in some other way.

[0021] In other words, investigating the causes of the discrepancies between forecasts and actual results is crucial from the perspective of adjusting and evolving the various models used in the above simulations to better reflect social conditions. In particular, "policy response models" for policies related to human decision-making and behavior are prone to errors and can be a major cause of the discrepancies between forecasts and actual results. Therefore, the technical significance of being able to determine whether the cause of the discrepancies lies in the policy response models or elsewhere is considerable.

[0022] Figure 2 is a schematic diagram illustrating the factors contributing to the discrepancy in traffic volume. As shown in Figure 2, during the simulation, OD estimation is performed on sensing data stored in storage 2, etc., to generate OD data (S1A). When this generated OD data is input to the simulator, the behavior of agents placed on the road network being simulated is simulated according to various models that define the behavior of the agents (S2A).

[0023] For example, examples of agent models may include the policy response model mentioned above, as well as route selection models that select the route a vehicle agent will take, and driving behavior models that reproduce the driving behavior of a vehicle agent.

[0024] Here, as an example of the above measures, we will consider area pricing, also known as road pricing. In this case, a charging zone is set within the area being simulated. Under such a setting of charging zones, if the route selected by the route selection model includes a charging zone, a policy response model is applied that avoids passing through the charging zone according to the probability corresponding to the amount charged in the charging zone. Here, road pricing is given as an example of a policy, but other measures such as lane control or school zone setting may also be implemented.

[0025] When this policy response model is applied and the simulation is run, traffic flow data with the policy in place is obtained. The "traffic flow data" referred to here can, for example, be time-series data of the agent's location.

[0026] Subsequently, based on traffic flow data with implemented measures, the traffic volume of agents passing through each zone, which is a division of the area to be simulated, is aggregated (S3A). As an example, time-series data W1 of the predicted traffic volume for the zone is obtained, as shown in graph G1 in Figure 2.

[0027] On the other hand, when the measures are actually implemented, analysis processing is performed on the sensing data sensed by the sensors 20 (S1B), thereby obtaining actual traffic flow data with the measures in place. For example, as an example of the above analysis processing, image processing may be performed to identify individual vehicles from images sensed by the imaging sensors using license plates, etc., and to track the vehicles in chronological order between images. At this time, the vehicle's position information can be determined in chronological order from the coordinates on the area where the imaging sensors are placed and the coordinates on the images captured by the imaging sensors. As a result, chronological data of the vehicle's position information can be obtained for each vehicle.

[0028] Based on the actual traffic flow data with the implemented measures in place, the traffic volume of vehicles passing through each zone, which is a division of the area to be simulated, is aggregated (S3B). As an example, time-series data W2 of the actual traffic volume for the zone is obtained, as shown in graph G2 in Figure 2.

[0029] Discrepancies can occur between these simulations and reality, as shown by W1 and W2 in Figure 2. Such discrepancies may be caused by the policy response model or by other factors.

[0030] For example, in addition to the policy response model, there may be a wide range of other causes for the discrepancy between planned and actual results mentioned above, such as (1) errors in the driving behavior model, (2) errors in OD estimation, (3) differences in the assumed total traffic volume, or (4) irregularities.

[0031] Of these, if (1) above is the cause of the discrepancy, the driving behavior model may further include models such as a vehicle-following model, a lane-changing model, an overtaking model, an intersection behavior model, and a lane-merging model. In this case, further detailed investigation of the cause and tuning of the model causing the discrepancy are required. Also, if (3) above is the cause of the discrepancy, since traffic volume that was not anticipated during the simulation is actually occurring, tuning of the simulation's input data is required.

[0032] As described above, numerous factors can contribute to the discrepancy between planned and actual results, making it difficult to pinpoint the exact cause. Therefore, determining whether the discrepancy lies in the policy response model (which is more prone to errors than other models) or elsewhere is of great technical significance for investigating the cause of the discrepancy and, consequently, implementing countermeasures.

[0033] <One aspect of a problem-solving approach> Therefore, as a related function of the simulation function according to this embodiment, a determination function is provided to determine whether or not the cause of the discrepancy between the budget and actual results lies outside the policy response model. In such a determination function, the policy impact area affected by the policy response model is set based on the policy implementation zone in the area to be simulated, and if a discrepancy between the budget and actual results exists outside the policy impact area, it is determined that the cause of the discrepancy is outside the policy response model.

[0034] One reason why this judgment logic is valid is that if the discrepancy between the budget and actual results is caused by the policy response model, the error in the budget and actual results is likely to be distributed near the policy implementation zone.

[0035] Figure 3 is a schematic diagram illustrating an example of applying a policy response model. Figure 3 shows four agents, A1 to A4, as examples of agents whose routes selected by the route selection model include policy implementation zones, such as charging zones. Furthermore, Figure 3 schematically shows the agent routes when a simulation is performed according to a policy response model in which the probability of avoiding passage through charging zones is set to 75%. When such a policy response model is applied to the simulation, as shown in Figure 3, agents A1, A3, and A4, which correspond to 75% of the total, avoid passing through policy implementation zones, while agent A2, which corresponds to 25% of the total, passes through policy implementation zones.

[0036] Figure 4 is a schematic diagram showing an example of actual traffic flow. Figure 4 schematically shows the traffic flow measured when the policy implementation zones corresponding to the policy implementation zones shown in Figure 3 are actually established. Furthermore, Figure 4 shows four vehicles, C1 to C4, as an example of actual traffic flow. As shown in Figure 4, vehicle C4, which corresponds to 25% of the total, avoids passing through the policy implementation zone, while vehicles C1 to C3, which correspond to 75% of the total, pass through the policy implementation zone.

[0037] As shown in the examples in Figures 3 and 4, there is a difference between the 75% probability that traffic in the policy implementation zone will be avoided according to the policy response model, and the 25% probability that traffic in the policy implementation zone will actually be avoided. When such a discrepancy occurs due to an error in the policy response model, since the policy response model defines the behavior in relation to the policy implementation area, it is highly likely that the error between the forecast and actual results will be distributed near the policy implementation zone. Conversely, if the discrepancy exists outside the vicinity of the policy implementation zone, it can be determined that the cause lies elsewhere than the policy response model.

[0038] As merely one example, the judgment function in this embodiment can also set the above-mentioned policy impact area based on the difference between the simulation results when the policy response model is not applied and the simulation results when the policy response model is applied.

[0039] Figure 5 shows an example of the area affected by a policy. In Figure 5, the boundaries of the policy implementation zones, where the policy is implemented, are shown by thick solid lines within the divided areas of Area B. Furthermore, in Figure 5, the boundary ∂M of the divided area where the difference between the simulation results without the policy response model and the simulation results with the policy response model applied is greater than or equal to the threshold δ is shown by a thick dashed line. In addition, in Figure 5, the divided areas where the difference between the simulation results without the policy response model and the actual traffic volume during policy implementation corresponding to the policy implementation zone is greater than or equal to the threshold γ is shown by hatching, and the divided areas are shown with higher density hatching as the difference increases.

[0040] As shown in Figure 5, when the policy impact area is defined based on the difference between the simulation results when the policy response model is not applied and the simulation results when the policy response model is applied, its boundary ∂M can be defined without being limited to the policy implementation zone and its vicinity.

[0041] In other words, as shown in the example in Figure 5, the policy impact area can include not only the policy implementation zone and its vicinity, but also areas affected by agents avoiding the policy implementation zone when applying the policy response model, such as detour routes.

[0042] Under these policy impact area definitions, a segmented area outside its boundary ∂M, such as segmented area Z1 shown in Figure 5, is not near the policy implementation zone and cannot be considered a detour route, making it highly likely that the deviation is caused by something other than the policy response model.

[0043] Therefore, by setting the policy impact area based on the difference between the simulation results when the policy response model is not applied and the simulation results when the policy response model is applied, it is possible to improve the accuracy of determining whether or not the cause of the discrepancy between budget and actual results lies outside of the policy response model.

[0044] Figure 5 shows an example of setting the above policy impact area based on the difference between the simulation results when the policy response model is not applied and the simulation results when the policy response model is applied, but it is not limited to this example.

[0045] In other words, if the cause of the discrepancy between the budget and actual results lies in the policy response model, there is a high probability that the error in the budget and actual results will be distributed near the policy implementation zone. Therefore, it is possible to set a "policy impact area" within a predetermined distance from the policy implementation zone.

[0046] For example, the above determination function can set the policy influence area as the range from a representative point of the divided area corresponding to the policy implementation zone, such as the center, centroid, or vertex, up to a distance corresponding to a threshold. In addition, the above determination function can set the policy influence area as the divided area corresponding to the policy implementation zone and eight divided areas located in the vicinity of that divided area.

[0047] By defining the scope of influence of such policies, it becomes possible to improve the accuracy of determining whether the cause of the discrepancy between budget and actual results lies outside of the policy response model.

[0048] <Configuration of Server Device 10> Next, an example of the functional configuration of the server device 10 according to this embodiment will be described. Figure 1 schematically shows the blocks related to the simulation function of the server device 10. As shown in Figure 1, the server device 10 has a communication control unit 11, a storage unit 13, and a control unit 15. Note that Figure 1 only shows an excerpt of the functional units related to the above-mentioned simulation function, and the server device 10 may also be equipped with functional units other than those shown.

[0049] The communication control unit 11 is a functional unit that controls communication with other devices such as sensors 20 and client terminals 30. As just one example, the communication control unit 11 can be implemented using a network interface card such as a LAN card. In one aspect, the communication control unit 11 receives sensing data from sensors 20 and outputs the results of determining the cause of the discrepancy between planned and actual results to the client terminals 30.

[0050] The storage unit 13 is a functional unit that stores various types of data. As an example, the storage unit 13 can be implemented by internal, external, or auxiliary storage of the server device 10. For example, the storage unit 13 stores OD data 13A, traffic flow data without measures 13B, traffic flow data with measures 13C, and measured traffic flow data 13D. The descriptions of OD data 13A, traffic flow data without measures 13B, traffic flow data with measures 13C, and measured traffic flow data 13D will be explained together when reference or registration is performed.

[0051] The control unit 15 is a functional unit that performs overall control of the server device 10. For example, the control unit 15 can be implemented by a hardware processor. As shown in Figure 1, the control unit 15 has a simulation unit 16, an analysis unit 17A, a calculation unit 17B, a setting unit 17C, and a determination unit 17D. The control unit 15 may also be implemented by hardwired logic or the like.

[0052] The simulation unit 16 is a processing unit that executes the above simulation. One aspect of this is that the simulation unit 16 uses the OD data 13A stored in the storage unit 13 as input data and simulates the behavior of agents placed on the road network of the area to be simulated without applying the policy response model 16A. In this case, the policy response model 16A is not applied as a model that defines the behavior of the agents, and a route selection model or driving behavior model (not shown) is applied to the simulation. For example, the OD data 13A may be data in which the number of trips is associated with each combination of start and end point ODs, such as an OD table. As a result of such a simulation, time-series data of the location information of each agent when the policy response model 16A is not applied is obtained. The data obtained in this way is saved to the storage unit 13 as traffic flow data without policy 13B.

[0053] In another aspect, the simulation unit 16 uses the same OD data 13A as input data and applies the policy response model 16A to simulate the behavior of agents placed on the road network of the area to be simulated. In this case, as models that define the behavior of the agents, the policy response model 16A, as well as route selection models and driving behavior models (not shown), are applied to the simulation. As a result of such simulation, time-series data of the location information of each agent when the policy response model 16A is applied is obtained. The data obtained in this way is stored in the storage unit 13 as policy-implemented traffic flow data 13C.

[0054] The analysis unit 17A is a processing unit that performs analysis on sensing data acquired from sensors 20 after the start of a measure corresponding to the measure implementation zone set during the simulation. As an example, the analysis unit 17A can perform image processing that tracks vehicles chronologically across images while identifying individual vehicles from images sensed by the imaging sensor using license plates, etc. At this time, the vehicle's position information can be determined chronologically from the coordinates on the area where the imaging sensor is placed and the coordinates on the image captured by the imaging sensor. As a result of such analysis, chronological data of the vehicle's position information is obtained for each vehicle. The data obtained in this way is stored in the storage unit 13 as actual traffic flow data 13D during the implementation of the measure.

[0055] The calculation unit 17B is a processing unit that calculates the traffic volume within the divided regions into which the simulation target area is divided, for each of the three types of data: traffic flow data without measures 13B, traffic flow data with measures 13C, and measured traffic flow data 13D. As just one example, the calculation unit 17B can start processing when all three types of data—traffic flow data without measures 13B, traffic flow data with measures 13C, and measured traffic flow data 13D—are acquired.

[0056] More specifically, the calculation unit 17B divides the area to be simulated into multiple sub-regions. Figure 6 shows an example of a sub-region. As shown in Figure 6, the area B to be simulated is divided into n sub-regions A i=1~n It is divided into A1~A 48 It is divided into 48 separate areas, and the policy implementation zone for the charge amount S is divided into area A. 31This is the setting. Figure 6 shows an example where Area B is divided into grid cells, but the shape and size of the divided areas can be arbitrary, and the shape and size may differ between individual divided areas. However, Area B will be divided into shapes and sizes that include one or more divided areas within the measure implementation zone set during the simulation.

[0057] After such area division, the calculation unit 17B calculates the traffic volume within each divided area for each type of data: traffic flow data without measures 13B, traffic flow data with measures 13C, and actual traffic flow data 13D. As an example, if a one-hour simulation is performed in one-minute increments, the calculation unit can determine the total number of vehicles in the divided area per minute over one hour, or the average of the average speed of all vehicles in the divided area per minute over one hour, as an example of the traffic volume within each divided area.

[0058] In the following, for the purpose of distinguishing the labels of traffic volume within the divided areas, traffic volume calculated from the traffic flow data without measures 13B may be referred to as "traffic volume without measures." Furthermore, traffic volume calculated from the traffic flow data with measures 13C may be referred to as "traffic volume with measures." Furthermore, traffic volume calculated from the measured traffic flow data 13D may be referred to as "measured traffic volume."

[0059] After the traffic volume within the divided area is calculated in this way, the calculation unit 17B calculates the difference between the traffic volume without measures and the traffic volume with measures. This difference between the traffic volume without measures and the traffic volume with measures is sometimes referred to as the "difference in traffic volume with and without measures."

[0060] Figure 7 is a schematic diagram showing an example of traffic volume without any measures. Figure 8 is a schematic diagram showing an example of traffic volume with measures. Figure 9 is a schematic diagram showing an example of the difference in traffic volume with and without measures. Figures 8 and 9 show n divided regions A1 to A n Among these, divided area A has a designated zone for implementing measures. 31 The boundaries are indicated by thick solid lines in white or black. Furthermore, Figures 7 and 8 show n divided regions A1~A nAmong them, the traffic volume α without measures i or the traffic volume β with measures i As it increases, the divided area A i is shown by high-density hatching. Further, in FIG. 9, among the n divided areas A1 to A n The traffic volume difference c between the presence and absence of measures i The divided area A where it is equal to or greater than the threshold value δ j Is shown by hatching, and as the traffic volume difference c between the presence and absence of measures i Increases, the divided area A j Is shown by high-density hatching.

[0061] As an example, the traffic volume difference c between the presence and absence of measures in the divided area A shown in FIG. 9 i Can be calculated by subtracting the traffic volume α without measures in the divided area A shown in FIG. 7 from the traffic volume β with measures in the divided area A shown in FIG. 8, that is, according to the following formula (1). i Is the traffic volume β with measures in the divided area A shown in FIG. 8 i From the traffic volume α without measures in the divided area A shown in FIG. 7 i The divided area A shown in FIG. 7 i Of the traffic volume α without measures i That is, it can be calculated according to the following formula (1).

[0062] c i =β i -α i ···(1)

[0063] In parallel with this, the calculation unit 17B calculates the difference between the traffic volume without measures and the measured traffic volume. The difference between the traffic volume without measures and the measured traffic volume may be described as the "traffic volume difference between prediction and measurement".

[0064] FIG. 10 is a schematic diagram showing an example of the measured traffic volume. FIG. 11 is a schematic diagram showing an example of the traffic volume difference between prediction and measurement. In FIGS. 10 and 11, among the n divided areas A1 to A n The boundary of the divided area A where the measure implementation zone is set 31 Is shown by a solid line of a thick white or black line. Further, in FIG. 10, among the n divided areas A1 to A n Of the measured traffic volume β′ i As it increases, the divided area A iThis is shown by high-density hatching. Furthermore, Figure 11 shows n divided regions A1~A n c′ - Difference between planned and actual traffic volume i Divided region A where the threshold γ is greater than or equal to the threshold γ i This is indicated by hatching, along with the difference in traffic volume c' between planned and actual. i Therefore, the divided region A becomes larger. i This is shown with high-density hatching.

[0065] As an example, see the divided region A shown in Figure 11. i The difference between planned and actual traffic volume c' i This is the divided region A shown in Figure 10. i Actual measured traffic volume β′ i From the divided region A shown in Figure 7 i Traffic volume α without measures i This can be calculated by subtracting, that is, according to equation (2) below.

[0066] c′ i =β′ i -α i ...(2)

[0067] Returning to the explanation of Figure 1, the setting unit 17C is a processing unit that sets the policy impact area, which is affected by the policy response model 16A in the area to be simulated, based on the policy implementation zone.

[0068] As just one example, the setting unit 17C divides region A by the calculation unit 17B. i The difference in traffic volume with and without the policy is calculated for each case. i Based on this, the above policy impact area can be set. For example, the setting unit 17C sets the traffic volume difference c between the presence and absence of the policy among the n divided areas. i Divided region A where the threshold δ is greater than or equal to the threshold δ i Element A of the policy impact area M m It is extracted as follows. Then, the setting unit 17C sets the extracted element A m The set of these is the policy impact area M(=U |ci|≧δ A i ) is set as and the boundary ∂M of the policy impact area M is set. In the example shown in Figure 9, the difference in traffic volume c with and without the policy is set.i Divided region A where the threshold δ is greater than or equal to the threshold δ i , that is, hatched divided region A 3~ A8, divided area A 20~ A 24 , divided area A 30~ A 32 , divided area A 37~ A 40 , divided area A 42 This is set as the policy impact area M. Furthermore, the boundary ∂M of the policy impact area M is set.

[0069] The determination unit 17D is a processing unit that determines whether or not the cause of the discrepancy between the budget and actual results is something other than the policy response model. In one embodiment, if the discrepancy between the budget and actual results exists outside the policy influence area set by the setting unit 17C, the determination unit 17D determines that the cause of the discrepancy is something other than the policy response model.

[0070] More specifically, the determination unit 17D determines the divided region A that exists outside the boundary ∂M of the policy influence region M set by the setting unit 17C within the area B to be simulated. j Extract the set BM. The partitioned region A is included in the set BM extracted in this way. j For each division region A, the determination unit 17D determines j The difference between planned and actual traffic volume c' j Determine whether the value is above the threshold γ.

[0071] Here, divided region A j The difference between planned and actual traffic volume c' j If the value is above the threshold γ, it is unlikely to be near the policy implementation zone or on a detour route, so the possibility of the discrepancy being caused by something other than the policy response model increases. In this case, the determination unit 17D determines the difference in traffic volume c' between the planned and actual traffic. j Divided region A where the threshold γ is greater than or equal to the threshold γ j The deviation element A outside the boundary ∂M k Extract it as follows.

[0072] For example, Figure 11 shows n divided regions A1 to A n c′ - Difference between planned and actual traffic volume iDivided region A where the threshold γ is greater than or equal to the threshold γ i This is indicated by hatching, along with the difference in traffic volume c' between planned and actual. i Therefore, the divided region A becomes larger. i This is indicated by high-density hatching. When the boundary ∂M of the policy impact area M shown in Figure 9 is superimposed on this map of area B in Figure 11, the map of area B shown in Figure 5 is obtained.

[0073] In the example shown in Figure 5, the divided region A j Among the set of BMs, the difference between planned and actual traffic volume c' i Divided region A where the threshold γ is greater than or equal to the threshold γ j As shown in Figure 5, divided region A is indicated by hatching. 26 Such a divided region A exists. 26 The deviation element A outside the boundary ∂M k It is extracted as follows.

[0074] Thus, the partitioned region A outside the boundary ∂M j Difference between planned and actual traffic volume c' j Determination and the deviation element A outside the boundary ∂M according to the determination result k After the extraction is performed, the determination unit 17D determines the deviation element A outside the boundary ∂M. k Determine whether or not it exists.

[0075] At this time, the deviation element A outside the boundary ∂M k If such a factor exists, the determination unit 17D determines that the cause of the discrepancy between the planned and actual results lies outside the policy response model. On the other hand, the discrepancy element A outside the boundary ∂M k If none exists, the determination unit 17D determines that there is no discrepancy between the planned and actual results outside the boundary ∂M.

[0076] Such a determination result can be output to any output destination. As an example only, the determination result can be output to an external device such as the client terminal 30, or software or service that uses a sectional table. When outputting the determination result in this way, it is possible to execute display such as the above-mentioned determination result message or icon, voice output such as a message or alert sound, and other outputs, for example, printing output. For example, examples of messages include "A deviation occurs when the simulation and the actual implementation of measures are carried out" and "Since there is a high possibility that there is a cause other than the policy response model, please investigate models other than the policy response model."

[0077] Furthermore, as a determination result, a map in which the policy influence area M and its boundary ∂M shown in FIG. 9 and the deviation element A shown in FIG. 5 are mapped on the map of area B shown in FIG. 6 k can also be output. For example, according to the display of the policy influence area M and its boundary ∂M shown in FIG. 9, it is possible to grasp the current situation that the influence of the implementation of the policy is not only in the policy implementation zone and its vicinity, but also the highways included in the divided areas A3 to A8 shown in FIG. 6 are affected as detour routes. Furthermore, according to the display of the deviation element A 26 shown in FIG. 5, since it exists not in the policy implementation zone and its vicinity, nor on the highway that can be used as a detour route, it can be grasped that the cause of the deviation in the deviation element A 26 is likely to be other than the policy response model 16A.

[0078] Also, as a determination result, the number of elements of the deviation element A k can also be output. In this case, while increasing the alert warning level as the number of elements increases, the alert warning level can be decreased as the number of elements decreases.

[0079] <Specific example> Here, specific examples of factors causing divergence other than the policy response model are given. FIG. 12 is a diagram showing an example of a factor causing divergence. Area B shown in FIG. 12 is the same area as Area B shown in FIG. 6. On the other hand, Area B shown in FIG. 12 further includes a new road R1 that is not in the road network of Area B shown in FIG. 6.

[0080] In an example such as FIG. 12, similar to the example shown in FIG. 6, the traffic volume without measures shown in FIG. 7 and the traffic volume with measures shown in FIG. 8 are calculated, while the difference is that the measured traffic volume shown in FIG. 13 is calculated. In this case, the calculation result of the difference between the predicted and actual traffic volumes is as shown in FIG. 14.

[0081] FIG. 13 is a schematic diagram showing an example of the measured traffic volume. FIG. 14 is a schematic diagram showing an example of the difference between the predicted and actual traffic volumes. That is, the difference c' in the predicted and actual traffic volumes i of the divided area A shown in FIG. 14 i is the measured traffic volume β' of the divided area A shown in FIG. 13 i minus the traffic volume α without measures of the divided area A shown in FIG. 7 i , that is, it can be calculated according to the above formula (2). i i j i

[0082] In the example shown in FIG. 14, among the sets B - M of the divided area A j , the divided areas A where the difference c' in the predicted and actual traffic volumes i is greater than or equal to the threshold value γ j , such as the divided area A shown hatched in FIG. 14 17 , the divided area A 25 and the divided area A 26 exist. These divided areas A 17 , the divided area A 25 and the divided area A 26 are extracted as the divergence elements A k outside the boundary ∂M. Such divergence elements A 17 , divergence elements A 25 and divergence elements A 26According to this, it is clear that when a discrepancy between planned and actual results occurs due to factors other than the policy response model 16A, namely the opening of the new road R1 shown in Figure 12, the factors contributing to that discrepancy can be identified.

[0083] Let's look at a more specific example. Figure 15 shows another example of a factor causing the discrepancy. Area B shown in Figure 15 is the same area as Area B shown in Figure 6. On the other hand, Area B shown in Figure 15 experiences an irregularity I1, such as an accident, which does not occur in Area B shown in Figure 6.

[0084] In this example shown in Figure 15, similar to the example in Figure 6, the traffic volume without measures (shown in Figure 7) and the traffic volume with measures (shown in Figure 8) are calculated, but the difference is that the actual measured traffic volume (shown in Figure 16) is also calculated. In this case, the calculation result of the difference between planned and actual traffic volume is as shown in Figure 17.

[0085] Figure 16 is a schematic diagram showing an example of actual traffic volume. Figure 17 is a schematic diagram showing an example of the difference between planned and actual traffic volume. That is, divided area A shown in Figure 17. i The difference between planned and actual traffic volume c' i This is the divided region A shown in Figure 16. i Actual measured traffic volume β′ i From the divided region A shown in Figure 7 i Traffic volume α without measures i This can be calculated by subtracting, that is, according to equation (2) above.

[0086] In the example shown in Figure 17, division region A j Among the set of BMs, the difference between planned and actual traffic volume c' i Divided region A where the threshold γ is greater than or equal to the threshold γ j As shown in Figure 17, divided region A is indicated by hatching. 25 ~A 27 , divided area A 33 ~A 35 and divided region A 41 ~A 42 These divided regions A exist. 25 ~A 27 , divided area A 33 ~A 35and divided region A 41 ~A 42 The deviation element A outside the boundary ∂M k This is extracted as such. Such a divided region A 25 ~A 27 , divided area A 33 ~A 35 and divided region A 41 ~A 42 According to this, it is clear that when a discrepancy between the forecast and actual results occurs due to factors other than the policy response model 16A, namely, the irregular I1 shown in Figure 15, the discrepancy element can be identified.

[0087] <Processing flow> Figures 18 and 19 are flowcharts (1) and (2) showing the procedure for the determination process. This process can be activated, as an example, when three data sets are acquired: traffic flow data without measures 13B, traffic flow data with measures 13C, and actual measured traffic flow data 13D. As shown in Figure 18, first, the calculation unit 17B divides the area B to be simulated into multiple n divided regions A i=1~n Divide into (step S101).

[0088] Next, divided region A i Loop process 1 is executed, repeating the process from step S102 to step S106 below a number of times corresponding to the number n. Note that although Figure 18 shows an example in which the process from step S102 to step S106 is executed repeatedly, the process from step S102 to step S106 may also be executed in parallel for each of the n divided regions.

[0089] Furthermore, the calculation unit 17B executes a loop process 2 that repeats the process of step S102 below a number of times corresponding to the number of data types K for the traffic flow data without measures 13B, traffic flow data with measures 13C, and measured traffic flow data 13D. Although Figure 18 shows an example in which the process of step S102 is executed repeatedly, the process of step S102 may also be executed in parallel for each of the K types of traffic flow data.

[0090] In other words, the calculation unit 17B divides the region A based on the traffic flow data of type k. i The traffic volume within is calculated (step S102). This loop process 2 is repeated, and the divided region A i Traffic volume α without measures i , with measures in place, traffic volume β i and actual traffic volume β′ i This is calculated.

[0091] Next, the calculation unit 17B calculates the divided region A i Traffic volume α without measures i and traffic volume β with measures in place i By substituting this into equation (1) above, the divided region A i Difference in traffic volume due to the presence or absence of the policy c i Calculate (Step S103).

[0092] Then, the setting unit 17C sets the traffic volume difference c with and without the measure calculated in step S103. i It is determined whether the value is greater than or equal to the threshold δ (step S104). At this time, the difference in traffic volume c with and without the measure is determined. i If the value is greater than or equal to the threshold δ (step S104 Yes), the setting unit 17C sets the divided region A i Element A of the policy impact area M m Extract as follows (Step S105). Note that the difference in traffic volume with and without the policy is c. i If the value is not greater than or equal to the threshold δ (step S104No), the process in step S105 above is skipped.

[0093] Before, after, or in parallel with the processing in steps S103 to S105 above, the calculation unit 17B can perform the following processing. That is, the calculation unit 17B processes the divided region A i Traffic volume α without measures i and actual traffic volume β′ i By substituting this into equation (2) above, the divided region A i The difference between planned and actual traffic volume c' i Calculate (step S106).

[0094] As this loop process 1 is repeated, the divided region A i Element A of the policy impact area M for each m Attempting to extract the difference between planned and actual traffic volume c′ i This is calculated.

[0095] Furthermore, the setting unit 17C processes element A extracted in step S105. m The set of these is the policy impact area M(=U |ci|≧δ A i ) is set as such, and the boundary ∂M of the policy impact area M is set (step S107).

[0096] Subsequently, as shown in Figure 19, the determination unit 17D determines the divided region A that exists outside the boundary ∂M of the policy influence area M set in step S107 within the area B to be simulated. j Extract the set BM (step S108).

[0097] Then, the determination unit 17D determines the division region A outside the boundary ∂M. j Loop process 3 is executed, repeating the processes of steps S109 and S110 below a number of times corresponding to the number L of partitioned regions. Although Figure 19 shows an example in which the processes of steps S109 and S110 are executed repeatedly, the processes of steps S109 and S110 may also be executed in parallel for each of the L partitioned regions.

[0098] In other words, the determination unit 17D determines the division region A j The difference between planned and actual traffic volume c' j It is determined whether or not it is greater than or equal to the threshold γ (step S109). At this time, divided region A j The difference between planned and actual traffic volume c' j If the value is greater than or equal to the threshold γ (step S109 Yes), the determination unit 17D determines the difference between the planned and actual traffic volume c′. j Divided region A where the threshold γ is greater than or equal to the threshold γ j The deviation element A outside the boundary ∂M k Extract as (step S110). Note that divided region A j The difference between planned and actual traffic volume c' jIf the value is not the threshold γ (step S109No), the process in step S110 above is skipped.

[0099] As this loop process 3 is repeated, the partitioned region A outside the boundary ∂M is formed. j For each boundary ∂M, the deviation element A outside k An attempt is made to extract it.

[0100] Furthermore, the determination unit 17D determines the deviation element A outside the boundary ∂M k It is determined whether or not the divergence element A outside the boundary ∂M exists (step S111). k If such a condition exists (step S111 Yes), the determination unit 17D determines that the cause of the discrepancy between the planned and actual results lies elsewhere than the policy response model (step S112), and terminates the process. On the other hand, the discrepancy element A outside the boundary ∂M k If none exists (step S111No), the determination unit 17D determines that there is no discrepancy between the planned and actual figures outside the boundary ∂M (step S113), and terminates the process.

[0101] <One aspect of the effect> As described above, the server device 10 in this embodiment sets the policy impact area, which is affected by the policy response model in the area to be simulated, based on the policy implementation zone. Then, if a discrepancy between the budget and actual results exists outside the policy impact area, the server device 10 in this embodiment determines that the cause of the discrepancy is something other than the policy response model. Therefore, the server device 10 in this embodiment can determine whether or not the cause of the discrepancy between the budget and actual results is something other than the policy response model. [Examples]

[0102] Now, while embodiments of the disclosed apparatus have been described, the present invention may be implemented in various other forms besides those described above. Therefore, other embodiments included in the present invention will be described below.

[0103] <Examples of applications of threshold γ> In the above example 1, all divided regions A jAn example was given where a common threshold γ is applied to the divided region A. j The threshold γ can be changed according to the distance from the policy implementation zone. For example, the threshold γ can be set to be larger as the distance decreases, and smaller as the distance increases. This is because, when the cause of the discrepancy between budget and actuals lies in the policy response model, there is a high probability that the error in budget and actuals will be distributed near the policy implementation zone. This improves the accuracy of determining whether the cause of the discrepancy between budget and actuals lies elsewhere than the policy response model. Furthermore, the threshold γ can also be changed according to the amount charged S. For example, the threshold γ can be set to be smaller as the amount charged S decreases, and larger as the amount charged S increases.

[0104] <Distributed and Integrated> Furthermore, the components of each illustrated device do not necessarily have to be physically configured as shown. In other words, the specific forms of distribution and integration of each device are not limited to those shown, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads and usage conditions. For example, the simulation unit 16, analysis unit 17A, calculation unit 17B, setting unit 17C, or determination unit 17D may be connected to the server device 10 as external devices via a network. Alternatively, the simulation unit 16, analysis unit 17A, calculation unit 17B, setting unit 17C, or determination unit 17D may each be located in a separate device, connected via a network, and working together to realize the functions of the server device 10. For example, the simulation unit 16 and the analysis unit 17A, calculation unit 17B, setting unit 17C, and determination unit 17D may be located on different computers, or their functions may be provided by different service providers or vendors.

[0105] <Hardware Configuration> Furthermore, the various processes described in the above embodiments can be implemented by executing a pre-prepared program on a computer such as a personal computer or workstation. Therefore, below, an example of a computer that executes a determination program having the same functions as in Embodiments 1 and 2 will be described using Figure 20.

[0106] Figure 20 shows an example of a hardware configuration. As shown in Figure 20, the computer 100 has an operating unit 110a, a speaker 110b, a camera 110c, a display 120, and a communication unit 130. Furthermore, the computer 100 has a CPU 150, a ROM 160, an HDD 170, and RAM 180. These parts 110 to 180 are connected via a bus.

[0107] As shown in Figure 20, HDD 170 stores a determination program 170a that performs the same functions as the analysis unit 17A, calculation unit 17B, setting unit 17C, and determination unit 17D shown in Embodiment 1 above. This determination program 170a may be integrated or separated, similar to the components of the analysis unit 17A, calculation unit 17B, setting unit 17C, and determination unit 17D shown in Figure 1. In other words, HDD 170 does not necessarily have to store all the data shown in Embodiment 1 above; it is sufficient that the data used for processing is stored in HDD 170.

[0108] Under these conditions, the CPU 150 reads the judgment program 170a from the HDD 170 and then loads it into the RAM 180. As a result, the judgment program 170a functions as a judgment process 180a, as shown in Figure 20. This judgment process 180a loads the various data read from the HDD 170 into the memory area of ​​the RAM 180 allocated to the judgment process 180a, and then executes various processes using the loaded data. For example, the processes shown in Figures 18 to 19 may be examples of processes executed by the judgment process 180a. Note that the CPU 150 does not necessarily need to operate all of the processing units shown in the above embodiment 1; it is sufficient if the processing units corresponding to the processes to be executed are virtually implemented.

[0109] The judgment program 170a described above does not necessarily have to be stored in the HDD 170 or ROM 160 from the beginning. For example, the judgment program 170a could be stored on a "portable physical medium" such as a flexible disk, floppy disk, CD-ROM, DVD disk, magneto-optical disk, or IC card inserted into the computer 100. The computer 100 could then retrieve and execute the judgment program 170a from these portable physical media. Alternatively, the judgment program 170a could be stored on another computer or server device connected to the computer 100 via a public network, the internet, LAN, WAN, etc. The computer 100 could then download and execute the judgment program 170a stored in this manner.

[0110] With regard to embodiments including the above examples, the following additional information is disclosed.

[0111] (Note 1) Among the virtualized areas used as the simulation target, the policy impact area is defined based on the policy implementation zone, where the policy response model, which specifies the agent's response to the policy implementation zone where the policy is implemented, is affected. In the divided region of the area into which the area is divided, if the difference between the simulation results when the policy response model is applied and the actual measurement results when the policy is implemented corresponding to the policy implementation zone satisfies certain conditions, it is determined that the cause of the discrepancy between the simulation results and the actual measurement results is something other than the policy response model. A method for determining whether a computer should execute a process.

[0112] (Note 2) The process for setting the above includes a process for setting the policy impact area based on the difference between the simulation results when the policy response model is not applied and the simulation results when the policy response model is applied. The determination method described in Appendix 1, characterized by the features described herein.

[0113] (Note 3) The process of setting the above includes the process of setting the area within a predetermined distance from the measure implementation zone as the measure's influence area. The determination method described in Appendix 1, characterized by the features described herein.

[0114] (Note 4) The process for making the determination includes a process for determining that the cause of the deviation in the divided region where the difference is greater than or equal to a threshold is something other than the policy response model. The determination method described in Appendix 1, characterized by the features described herein.

[0115] (Note 5) The threshold is set according to the distance between the divided area and the measure implementation zone. The determination method described in Appendix 4, characterized by the features described herein.

[0116] (Note 6) The computer further executes a process to output that, as a result of the determination made in the above determination process, the cause of the discrepancy between the simulation result and the actual measurement result is something other than the policy response model. The determination method described in Appendix 1, characterized by the features described herein.

[0117] (Note 7) The output process includes the process of outputting a map on which the boundaries of the policy impact area and the divided areas where the difference is greater than or equal to a threshold are mapped onto the area. The determination method described in Appendix 6, characterized by the features described herein.

[0118] (Note 8) Among the virtualized areas used as simulation targets, the policy impact area is defined based on the policy implementation zone, where the policy response model, which specifies the agent's response to the policy implementation zone where the policy is implemented, is affected. In the divided region of the area into which the area is divided, if the difference between the simulation results when the policy response model is applied and the actual measurement results when the policy is implemented corresponding to the policy implementation zone satisfies certain conditions, it is determined that the cause of the discrepancy between the simulation results and the actual measurement results is something other than the policy response model. A decision-making program that causes a computer to execute a process.

[0119] (Note 9) The process for setting the above includes a process for setting the policy impact area based on the difference between the simulation results when the policy response model is not applied and the simulation results when the policy response model is applied. The determination program described in Appendix 8, characterized by the above.

[0120] (Note 10) The process of setting the above includes the process of setting the area within a predetermined distance from the measure implementation zone as the measure's influence area. The determination program described in Appendix 8, characterized by the above.

[0121] (Note 11) The process for making the determination includes a process for determining that the cause of the deviation in the divided region where the difference is greater than or equal to a threshold is something other than the policy response model. The determination program described in Appendix 8, characterized by the above.

[0122] (Note 12) The threshold is set according to the distance between the divided area and the measure implementation zone. The determination program described in Appendix 11, characterized by the above.

[0123] (Note 13) The computer further executes a process to output, as a result of the determination made in the above determination process, that the cause of the discrepancy between the simulation result and the actual measurement result is something other than the policy response model. The determination program described in Appendix 8, characterized by the above.

[0124] (Note 14) The output process includes the process of outputting a map on which the boundaries of the policy impact area and the divided areas where the difference is greater than or equal to a threshold are mapped onto the area. The determination program described in Appendix 13, characterized by the above.

[0125] (Note 15) Among the virtualized areas used as the simulation target, the policy impact area is defined based on the policy implementation zone, in which the policy response model, which specifies the agent's response to the policy implementation zone where the policy is implemented, is affected. In the divided region of the area into which the area is divided, if the difference between the simulation results when the policy response model is applied and the actual measurement results when the policy is implemented corresponding to the policy implementation zone satisfies certain conditions, it is determined that the cause of the discrepancy between the simulation results and the actual measurement results is something other than the policy response model. A determination device including a control unit that performs processing.

[0126] (Note 16) The process for setting the above includes a process for setting the policy impact area based on the difference between the simulation results when the policy response model is not applied and the simulation results when the policy response model is applied. The determination device according to Appendix 15, characterized by the features described herein.

[0127] (Note 17) The process of setting the above includes the process of setting the area within a predetermined distance from the measure implementation zone as the measure's influence area. The determination device according to Appendix 15, characterized by the features described herein.

[0128] (Note 18) The process for making the determination includes a process for determining that the cause of the deviation in the divided region where the difference is greater than or equal to a threshold is something other than the policy response model. The determination device according to Appendix 15, characterized by the features described herein.

[0129] (Note 19) The threshold is set according to the distance between the divided area and the measure implementation zone. The determination device described in Appendix 18, characterized by the features described herein.

[0130] (Note 20) The control unit further executes a process to output, as a result of the determination made in the above determination process, that the cause of the discrepancy between the simulation result and the actual measurement result is something other than the policy response model. The determination device according to Appendix 15, characterized by the features described herein.

[0131] (Note 21) The output process includes the process of outputting a map on which the boundaries of the policy impact area and the divided areas where the difference is greater than or equal to a threshold are mapped onto the area. The determination device described in Appendix 20, characterized by the features described herein. [Explanation of Symbols]

[0132] 10 Server devices 11. Communication Control Unit 13 Storage section 13A OD data 13B Traffic flow data without any policies 13C Measures in place: Traffic flow data 13D Actual Measured Traffic Flow Data 15 Control Unit 16. Simulation Department 16A Policy Response Model 17A Analysis section 17B Calculation part 17C Setting section 17D Judgment section 20 Sensors 30 client terminals

Claims

1. The policy impact area, which is influenced by a policy response model that defines the agent's response to a policy implementation zone where a policy has been set to be implemented, is set based on the policy implementation zone, among the virtualized areas used as the simulation target. In the divided region of the area into which the area is divided, in the divided region outside the area affected by the measure, it is determined whether the difference between the simulation result when the measure response model is not applied and the actual measurement result when the measure is implemented corresponding to the measure implementation zone is greater than or equal to a threshold. A method for determining whether a computer should execute a process.

2. The process for setting the policy includes a process for setting the policy impact area based on the difference between the simulation results when the policy response model is not applied and the simulation results when the policy response model is applied. The determination method according to feature 1.

3. The process described above includes a process of setting the area within a predetermined distance from the measure implementation zone as the measure's influence area. The determination method according to feature 1.

4. The process for making the determination includes a process for determining that the cause of the discrepancy in the divided region where the difference is greater than or equal to the threshold is a factor other than the measure. The determination method according to feature 1.

5. The threshold is set according to the distance between the divided area and the measure implementation zone. The determination method according to feature 4.

6. The computer further executes a process to output, based on the determination result determined in the aforementioned determination process, that the cause of the deviation in the divided region where the difference is greater than or equal to the threshold is a factor other than the aforementioned measure. The determination method according to feature 1.

7. The output process includes a process that outputs a map on which the boundaries of the policy impact area and the divided areas where the difference is greater than or equal to the threshold are mapped onto the area. The determination method according to feature 6.

8. The policy impact area, which is influenced by a policy response model that defines the agent's response to a policy implementation zone where a policy has been set to be implemented, is set based on the policy implementation zone, among the virtualized areas used as the simulation target. In the divided region of the area into which the area is divided, in the divided region outside the area affected by the measure, it is determined whether the difference between the simulation result when the measure response model is not applied and the actual measurement result when the measure is implemented corresponding to the measure implementation zone is greater than or equal to a threshold. A decision-making program that causes a computer to execute a process.

9. The policy impact area, which is influenced by a policy response model that defines the agent's response to a policy implementation zone where a policy has been set to be implemented, is set based on the policy implementation zone, among the virtualized areas used as the simulation target. In the divided region of the area into which the area is divided, in the divided region outside the area affected by the measure, it is determined whether the difference between the simulation result when the measure response model is not applied and the actual measurement result when the measure is implemented corresponding to the measure implementation zone is greater than or equal to a threshold. A determination device having a control unit that performs processing.

Images