Simulation system which integrates and uses a plurality of simulators

Abstract

To solve the deviation of output results between a comprehensive simulator and an individual simulator without modifying a model etc., in a simulation system which integrally uses the comprehensive simulator which comprehensively models the entirety of an event, and the individual simulator which individually models each part of the event.SOLUTION: A simulation system which integrally uses a comprehensive simulator which comprehensively models the whole body of an event, and an individual simulator which individually models each part of the event, corrects the simulation result of the comprehensive simulator by using a correction value of the simulation result of the comprehensive simulator determined, based on a statistical value of the difference between a simulation result of the individual simulator and the simulation result of the comprehensive simulator, so as to compensate the deviation of the simulation result of the comprehensive simulator with respect to the simulation result of the individual simulator.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a system for performing a simulation of an event using multiple simulators, and more particularly to a system configured to adjust the simulation results of the entire simulation using the simulation results of the multiple simulators. [Background technology]

[0002] Various systems have been proposed that integrate and execute a simulation of a certain event using simulators that use multiple different models. For example, Patent Document 1 proposes a configuration in which, when a simulation is executed using models with multiple different granularities (levels of granularity in model settings), a simulation of a first granularity and a simulation of a second granularity are executed separately while maintaining consistency of information between the individual models. Furthermore, Patent Document 2 proposes a configuration in an information processing system that executes multiple simulations in cooperation with multiple simulators that are connected to each other in a communicative manner, in which execution order information indicating the execution order of the multiple simulations is stored, and when one simulation reaches a checkpoint during its execution, the execution of the other simulations is controlled, thereby efficiently and quickly executing the multiple simulations as a whole. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2014-238749 [Patent Document 2] Patent Publication No. 2021-77272 [Non-patent literature]

[0004] [Patent Document 1] Y. Fujita, M. Kuwahara and E. Ueda, "MaaS Blender: Integration of Mobility Simulators for Mobility as a Service Evaluation," in IEEE Access, vol. 11, pp. 105105-105114, 2023, doi: 10.1109 / ACCESS.2023.3319079. Summary of the Invention [Problem to be solved by the invention]

[0005] One possible configuration for using multiple simulators in a simulation of a certain event is to simulate each part of the event using multiple simulators (individual simulators), each of which may individually model each part of the event, while simulating the entire event using one simulator (comprehensive simulator) that comprehensively models the entire event, and then integrate the results of these simulators. For example, when simulating the movement of people in a service that uses multiple mobility services (buses, trains, taxis, shared bicycles, etc.) such as MaaS (Mobility as a Service), a possible configuration is to select multiple mobility services to be used in the person's movement from one point to another and set the usage time for each, and then perform a comprehensive simulation using an exhaustive simulator (simulator management module) that is configured to comprehensively simulate the entire movement, and individual simulators (mobility simulators) that are configured to simulate the movement of people in each mobility service used in the movement from one point to another, and then integrate the results of these simulators (Non-Patent Document 1).

[0006] Generally, the comprehensive simulator and the individual simulator described above have different modeling granularity (the latter usually has finer granularity and higher accuracy than the former), which can cause discrepancies in the output results, making it difficult to properly integrate the output results. In such cases, in the past, the details of the model of each simulator were understood and the model was modified, but as the number of simulators increases or the model configuration becomes more complex, modifying the model becomes a cumbersome and difficult task.

[0007] In view of the above circumstances, the main object of the present invention is to provide a simulation system that integrates an exhaustive simulator that comprehensively models the entirety of a certain event with individual simulators that individually model each part of that event, and to eliminate any discrepancy in the output results between the exhaustive simulator and the individual simulators without modifying the models. [Means for solving the problem]

[0008] According to the present invention, the above problem is solved by a simulation system that uses an exhaustive simulator that exhaustively models the entire event and an individual simulator that individually models a part of the event in an integrated manner, a result difference extracting means for extracting a difference between a simulation result of the individual simulator and a simulation result of the comprehensive simulator; a correction value determination means for determining a correction value for the simulation result of the exhaustive simulator to compensate for a deviation of the simulation result of the exhaustive simulator from the simulation result of the individual simulator based on a statistical value of the difference of the simulation results extracted by the result difference extraction means; a result value correcting means for correcting a simulation result in the comprehensive simulator using the correction value determined by the correction value determining means; The system of the present invention can be realized by an operation of any computer device according to a program.

[0009] In the system of the present invention, an "event" may be any event to be simulated, such as the movement of a person in a service that uses multiple mobility services, such as MaaS. In this case, more specifically, in a trip from one point (starting point) to another point (destination) using various mobility services, the departure time (or arrival time) may be input, and the arrival and departure times at intermediate points, the arrival time at the destination (or departure time from the starting point), the total travel time, and the usage time of each mobility service may be output as simulation results. A "comprehensive simulator" is a simulator that simulates the entire "event" using a model that generates an "event" configured at any granularity and outputs the results. A "specific simulator" is a simulator that simulates a portion of the "event" using a model that generates a portion of the "event" configured at any granularity, as described above, and outputs the results. For example, if the "event" is the movement of a person from a departure point to a destination using various mobility services, the "comprehensive simulator" may be configured to, in response to the input of a departure time, select the mobility services to be used in the movement and determine the departure and arrival times of each mobility service, and calculate the arrival time at the destination, and the "individual simulator" may be configured to, in response to the input of the departure time or start time of use of a mobility service (instructed by the comprehensive simulator), output the arrival time at the destination or end time of use of that mobility service.

[0010] In a configuration in which the above-described "exhaustive simulator" and "individual simulator" each execute a simulation of the events for which they are responsible and output the results, as already mentioned, differences in the output results that make it difficult to integrate the results may occur due to differences in the granularity of each simulator. In this regard, since an individual simulator, which simulates only a portion of an event, is usually more accurate than an exhaustive simulator, which simulates the entire event, the system of the present invention, as described above, extracts the differences between the simulation results of the individual simulators (obtained by executing multiple simulations) and the simulation results of the exhaustive simulator, and determines a correction value for the simulation results of the exhaustive simulator to compensate for the deviation of the simulation results of the exhaustive simulator from the simulation results of the individual simulators based on a statistical value (e.g., an average value) of the differences. The simulation results of the exhaustive simulator are then corrected using the correction value. For example, if the "event" is the movement of a person using various mobility services from a departure point to a destination, the difference between the usage time of each mobility service in the simulation results of the comprehensive simulator and the usage time of each mobility service in the simulation results of the individual simulators for each mobility service is averaged for each usage time period and for each travel distance, and the simulation results output for each mobility service in the comprehensive simulator are corrected to compensate for this average difference. With this configuration, the simulation results of the comprehensive simulator are aligned with the simulation results of the individual simulators, making it easier to integrate the simulation results of each simulator and also expected to improve the accuracy of the simulation.

[0011] In the system of the present invention, if an abnormal value occurs in the statistical value of the difference between the simulation results of the individual simulator and the simulation results of the comprehensive simulator extracted by the result difference extraction means, the abnormal value may be removed before determining a correction value for the model setting value in the comprehensive simulator. The abnormal value can be detected by any method based on the trend of change in the statistical value of the difference. [Effects of the Invention]

[0012] Thus, the system according to the present invention can eliminate the discrepancy in simulation results due to the difference in granularity between the comprehensive simulator and the individual simulator, without requiring modification of the configuration of the models of the comprehensive simulator or the individual simulator. The configuration according to the present invention is advantageous in that when multiple individual simulators are used, such as in a simulation of human movement in MaaS using multiple mobility services, modification of the models of each individual simulator is not required. The configuration according to the present invention may be used to simulate various events.

[0013] Other objects and advantages of the present invention will become apparent from the following description of preferred embodiments of the invention. [Brief explanation of the drawings]

[0014] [Figure 1] FIG. 1 is a diagram schematically illustrating a computer in which the simulation system according to this embodiment is implemented. [Figure 2] FIG. 2 is a block diagram showing the configuration of the simulation system according to this embodiment. [Figure 3] Figure 3 is a table showing the simulation results of the departure and arrival times of several mobilities for several trips by the simulation management module in the simulation system of this embodiment, and the corresponding simulation results by the mobility simulator. [Figure 4]Fig. 4(A) is a schematic diagram of the average value of the difference for each time period, obtained by subtracting the simulation result of the simulation management module from the simulation result of the mobility simulator for the usage time (arrival time - departure time) of a certain mobility. Fig. 4(B) is a diagram of the state in which abnormal values ​​in the average value of the difference of the simulation results for each time period in Fig. 4(A) have been removed. [Figure 5] FIG. 5 is a schematic diagram showing the average value of the difference, obtained by subtracting the simulation result of the simulation management module from the simulation result of the mobility simulator, for the usage time (arrival time-departure time) of a certain mobility, for each usage distance. [Figure 6] FIG. 6 is a diagram showing, in a table format, an example of correction values ​​for model setting values ​​in the simulation management module. [Explanation of symbols]

[0015] 1...Computer main body, 2...Computer terminal, 3...Monitor, 4...Keyboard, mouse (input device) BEST MODE FOR CARRYING OUT THE INVENTION

[0016] The present invention will now be described in detail with reference to some preferred embodiments thereof with reference to the accompanying drawings, in which like reference numerals indicate like parts.

[0017] Computer equipment configuration The simulation system according to this embodiment, which integrates and utilizes multiple simulators, may be realized by the operation of a computer program on a computer device 1 of a type commonly used in this field, as illustrated in Fig. 1. The computer device 1 is equipped, in a typical manner, with a CPU, memory storing each program that executes calculations, a storage device having work memory and data memory used during calculations, and a computer terminal device 2 having a monitor 3 and input devices 4 such as a keyboard and a mouse that display and output instructions from the operator to the computer device 1, calculation results, and other information, and various results of the simulation may be displayed on the monitor 3, etc. Each simulator and module in the simulation system illustrated below in relation to Fig. 2 is realized by the operation of the computer device 1 in accordance with the program.

[0018] Simulation system configuration Below, we will explain the simulation system of this embodiment in the case of simulating the movement of people in a service that uses multiple mobility services such as MaaS, but it should be understood that this embodiment can also be applied to cases where simulations are carried out that integrate and use multiple other simulators.

[0019] Referring to FIG. 2, in a simulation system for simulating human movement using multiple mobility services according to this embodiment, when a request for movement from a departure point to a destination and the departure time from the departure point (or arrival time at the destination) of a certain person are given, a simulation management module (exhaustive simulator) determines a schedule including the selection of mobility services to be used from the departure point to the destination and the start and end times of use of each mobility service (this schedule is the simulation result of the exhaustive simulator). In the simulation for determining the schedule by the simulation management module, model setting values ​​set to standard values ​​for the movement speed and required time for each mobility service are used. The system also includes mobility simulators 1, 2, ..., N (individual simulators) that simulate human movement in each mobility service corresponding to various mobility services. Each mobility simulator is configured to receive the start point, end point, and start time (or end time) of a certain person's movement as input, and output the end time (or start time) of use as a simulation result using a model configured at the respective granularity. Input to each mobility simulator is provided by a simulation broker (coordinator) based on a schedule set by the simulation management module. In an actual simulation, the simulation management module receives input of various situations (time period, departure point, destination, number of people) and simulates the overall movement of people, while each mobility simulator simulates the movement of people when they use each mobility service. These results are integrated by the simulation broker to generate evaluations of the travel time required for each time period and location, the usage status of each mobility service, and people flow.

[0020] Correction of schedule (simulation results) by simulation management module As mentioned above, in the simulation of human mobility described above, differences in the granularity between the simulation management module and each mobility simulator can lead to discrepancies in the usage time and distance of each mobility service in the simulation results of both models. This can also lead to discrepancies in the start and end times of each mobility service. For example, as illustrated in FIG. 3, the start and end times (departure and arrival times) of each mobility service in each movement (event) can differ between the simulation results of the simulation management module and those of each mobility simulator. If such discrepancies remain, inconsistencies can occur in the results when integrating the simulation results, resulting in a deterioration in accuracy. In the past, attempts to resolve discrepancies between the results of the simulation management module and each mobility simulator have been made by internally modifying the models of each simulator. However, the cumbersomeness and difficulty of such work increases with the complexity of the models. Therefore, in this embodiment, the schedule, which is the simulation result of the simulation management module, is corrected so that the discrepancy (input / output discrepancy) between the start time and end time of use of each mobility between the simulation management module and each mobility simulator is easily eliminated without any internal modifications to the model.

[0021] Specifically, to correct the simulation results, the difference between the usage time in the simulation management module and the usage time in the mobility simulator for each mobility event involving multiple people (which may be the value obtained by subtracting the usage time in the simulation management module from the usage time in the mobility simulator) is first calculated in the difference calculation module in Fig. 2, and stored in a database of difference data. Then, the correction value calculation module in Fig. 2 calculates, from the difference data stored in the database, the average value of the differences in usage time for each time period (for example, every 10 minutes) and the average value of the differences in usage time for each usage distance.

[0022] FIG. 4A is a diagram showing a schematic representation of the change in the average value of the difference in usage time for a certain mobility by time period. As shown in the figure, when the average value of the difference in usage time for each time period is calculated, unnatural changes often occur in the trend of the change in the average value of the difference in usage time. Therefore, preferably, once the average value of the difference for each time period is obtained, the presence or absence of unnatural changes is detected using any outlier detection technology. If an unnatural change is detected, the unnatural change may be treated as an abnormal value and removed using any method, such as smoothing (abnormality correction). Thus, as shown in FIG. 4B, once the average value of the difference in usage time for each mobility by time period, from which the outliers have been removed, is obtained, this value is used as a correction value (a value that increases or decreases the required time) for the time required for each mobility by each time period (required time) in the simulation management module (time period correction).

[0023] Furthermore, as with the case of time periods, the average value of the difference between the usage time in the simulation management module and the usage time in the mobility simulator for each usage distance of each mobility changes depending on the usage distance, as shown schematically in FIG. 5 (when calculating the average value of the difference in usage time for each usage distance, data after removing unnatural variations in the average value of the difference for each time period may be used). Therefore, in this embodiment, as with the case of time periods, the average value of the difference in usage time for each usage distance is used as a correction value (a value that increases or decreases the required time) for the time (required time) required for each distance when using each mobility (time period correction). In this case, for example, a regression equation for the average value of the difference in usage time with distance as a variable may be found as shown in the figure, and the value calculated by the regression equation for each distance may be used as a correction value for the time required for each distance when using each mobility in the simulation management module.

[0024] 6 shows an example of how the usage time of each mobility service is corrected in the schedule (simulation result) output by the simulation management module. For example, for mobility 2, the required time is extended by 1 minute per 1 km of travel distance, and when the usage time is 9:10, the required time is extended by 5 minutes, and when the usage time is 9:20, the required time is extended by 10 minutes. The start and end times of mobility usage in the simulation result of the simulation management module are corrected. This correction is implemented when a simulation is executed after the simulation for which the correction value was obtained.

[0025] Thus, in the system according to the present embodiment, the differences between the simulation results of the simulation management module, which is a comprehensive simulator, and the simulation results of each mobility simulator, which is an individual simulator, are extracted, the statistical changes or distributions of the differences for each time period and for each distance are calculated, and the simulation results of the simulation management module for each time period and for each distance are corrected based on the statistical changes or distributions of the differences. According to the system configuration of the present embodiment, even if a discrepancy occurs in the output simulation results due to the use of models with different granularities between the comprehensive simulator and the individual simulator, the discrepancy between the simulation results of both can be eliminated without modifying the internal models, making it possible to appropriately integrate the simulation results, and improving the accuracy of the simulation results of the entire event is expected.

[0026] The above description has been made in relation to the embodiments of the present invention, but it will be apparent that many modifications and changes will be readily apparent to those skilled in the art, and the present invention is not limited to the above-described exemplary embodiments, but can be applied to various devices without departing from the concept of the present invention.

Claims

[Claim 1] A simulation system that uses an exhaustive simulator that comprehensively models an entire event and an individual simulator that individually models a part of the event in an integrated manner, a result difference extracting means for extracting a difference between a simulation result of the individual simulator and a simulation result of the comprehensive simulator; a correction value determination means for determining a correction value for the simulation result of the exhaustive simulator to compensate for a deviation of the simulation result of the exhaustive simulator from the simulation result of the individual simulator based on a statistical value of the difference of the simulation results extracted by the result difference extraction means; a result value correcting means for correcting a simulation result in the comprehensive simulator using the correction value determined by the correction value determining means; A system including:

Images