Virtual reservation generation device, virtual reservation generation method, and virtual reservation generation program

Abstract

To predict a movement demand for a non-introduction target area of a MaaS system.SOLUTION: A virtual reservation generation device of the present disclosure includes a calculation unit configured to calculate a feature amount related to a movement demand of a user for each of a plurality of regions obtained by dividing a target area on a map, a prediction unit configured to predict ratio at which each of the plurality of regions is selected as an objective value and ratio at which each of the plurality of regions is selected as a departure value based on the feature amount of each of the plurality of regions, and a generation unit configured to generate a virtual reservation indicating a reservation status of traffic from a departure point to a destination, which occurs in the target area based on a prediction result.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to a virtual reservation generation apparatus, a virtual reservation generation method, and a virtual reservation generation program.

Background Art

[0002] Conventionally, technologies related to on-demand vehicle dispatch services that provide services in accordance with users' service provision requests are known. For example, technologies for predicting service demand for each geographical region in consideration of at least any one of date and time information to be predicted, weather information, and event holding information are known.

Prior Art Literature

Patent Literature

[0003]

Patent Literature 1

Summary of the Invention

Problem to be Solved by the Invention

[0004] In addition, in recent years, MaaS (Mobility as a Service), which provides specific types of services such as a service that dispatches a vehicle equipped with a nurse and medical functions to the vicinity of the home of a patient who has difficulty moving to perform telemedicine with a doctor, is known.

[0005] However, the above-mentioned conventional prediction technologies cannot predict travel demand for areas where the MaaS system has not been introduced yet. Therefore, it is not possible to determine whether it is appropriate to introduce the MaaS system into areas where it has not been introduced, and business expansion cannot be performed quickly.

[0006] The present disclosure has been made in view of the above, and an object of the present disclosure is to provide a virtual reservation generation apparatus, a virtual reservation generation method, and a virtual reservation generation program that enable prediction of travel demand for areas where a MaaS system has not been introduced. [Means for solving the problem]

[0007] The virtual reservation generation device of this disclosure includes: a calculation unit that calculates feature quantities related to the user's travel demand for each of a plurality of regions into which the target area on the map is divided; a prediction unit that predicts the proportion in which each of the plurality of regions is selected as the target value and the proportion in which each of the plurality of regions is selected as the departure value based on the feature quantities of each of the plurality of regions; and a generation unit that generates virtual reservations showing the reservation status of transportation from the departure point to the destination in the target area based on the prediction results. [Effects of the Invention]

[0008] According to one embodiment of the system, it becomes possible to predict travel demand in areas where a MaaS system has not yet been introduced. [Brief explanation of the drawing]

[0009] [Figure 1] This diagram shows the configuration of the provisional reservation generation device according to the embodiment. [Figure 2] This diagram shows the input / output relationship of the mobility demand forecasting unit according to the embodiment. [Figure 3] This diagram illustrates an example of the relationship between the channel input to the mobility demand forecasting unit according to the embodiment and the target area. [Figure 4] This diagram shows the input / output relationship of the destination selection ratio prediction unit according to the embodiment. [Figure 5] This figure shows an example of the area of ​​interest and surrounding areas of the target area according to this embodiment. [Figure 6] This figure shows the channels of the region of interest that are input to the destination selection ratio prediction unit according to the embodiment. [Figure 7] This figure shows the weight of the region in the target area where the generated demand calculated by the destination selection ratio prediction unit according to the embodiment is the region of interest. [Figure 8] This diagram shows the input / output relationship of the departure point selection rate prediction unit according to the embodiment. [Figure 9]This figure shows the weights that indicate where in the surrounding region the demand generated in the region of interest originates, as calculated by the origin selection ratio prediction unit according to the embodiment. [Figure 10] This diagram shows the virtual reservation generation process of the virtual reservation generation device according to the embodiment. [Figure 11] This is a diagram illustrating the effects of the virtual reservation generation device according to the embodiment. [Figure 12] This is a hardware configuration diagram showing an example of a computer that implements the functions of an information processing device. [Modes for carrying out the invention]

[0010] Preferred embodiments of this disclosure will be described in detail below with reference to the attached drawings. In this specification and drawings, components having substantially the same functional configuration are denoted by the same reference numerals, and redundant descriptions will be omitted.

[0011] [0. Configuration of the virtual reservation generation device 100] First, the virtual reservation generation device 100 according to the embodiment will be described. The virtual reservation generation device 100 is an example of a computer that predicts travel demand for areas where user travel systems such as MaaS systems have not yet been introduced.

[0012] Before introducing a MaaS system, estimating the effects of its implementation in advance, including the coverage rate of mobility demand, the number of vehicles, and consideration of area division for the target region, can lead to various benefits such as narrowing down the implementation area, determining priorities, and understanding the expected number of users, thereby promoting business growth. However, conventional mobility demand forecasting models make it difficult to estimate the above-mentioned effects of implementation in advance.

[0013] Therefore, the virtual reservation generation apparatus 100 according to the embodiment predicts the number of demanders in an area occurring across the entire region, the selection ratio of destination values, and the selection ratio of departure points, and generates virtual reservations for traffic occurring in the target area. The virtual reservations generated in this way enable prediction of movement demand for areas where the MaaS system has not yet been introduced. Additionally, it is possible to grasp customer needs and realize efficient provision of services that meet these needs.

[0014] [1. Configuration of Virtual Reservation Generation Apparatus 100] Figure 1 is a diagram showing the configuration of the virtual reservation generation apparatus 100 according to the embodiment. The virtual reservation generation apparatus 100 is an information processing apparatus, and includes a communication unit 101, an input unit 102, an output unit 103, a control unit 104, and a storage unit 105.

[0015] (Communication Unit 101) The communication unit 101 is implemented by, for example, a Network Interface Card (NIC) or the like. The communication unit 101 is connected to a network via a wired or wireless connection, and transmits and receives information to and from an external information processing apparatus.

[0016] The communication unit 101 can communicate the same information as information input via the input unit 102 described later and information output via the output unit 103 to and from an external information processing apparatus. For example, the communication unit 101 executes transmission and reception of information related to an area where the MaaS system has not been introduced, assumption information required when generating virtual reservations, results of the generated virtual reservations, and the like, to and from an external information processing apparatus (not shown) outside the virtual reservation generation apparatus 100.

[0017] (Input Unit 102) The input unit 102 receives various operations input by a user. For example, the input unit 102 may receive various operations from the user via a display surface (e.g., the output unit 103) by means of a touch panel function. Additionally, the input unit 102 may receive various operations from buttons provided on the virtual reservation generation apparatus 100, or from a keyboard or mouse connected to the virtual reservation generation apparatus 100.

[0018] For example, the input unit 102 accepts input of information about the target area for which virtual reservations are to be generated (e.g., latitude and longitude information of the center of the target area), and hypothetical information (number of active users, usage history, number of new users, number of repeat users, location of boarding and alighting points (e.g., in front of a hospital, inside a station, etc.)).

[0019] (Output section 103) The output unit 103 is a display screen, for example, a liquid crystal display or an organic EL (Electro-Luminescence) display, and is a display device for displaying various information. The output unit 103 displays various information, such as virtual reservations, generated by the control unit 104, which will be described later, according to the control of the control unit 104. If a touch panel is used in the virtual reservation generation device 100, the input unit 102 and the output unit 103 are integrated.

[0020] The specific operation of the control unit 104 is related to the information stored in the memory unit 105, so the memory unit 105 will be explained first.

[0021] (Storage unit 105) The memory unit 105 stores the target area information memory unit 121, the hypothetical information memory unit 122, the AI ​​model memory unit 123, and the virtual reservation generation program 124. The AI ​​model memory unit 123 stores the travel demand forecasting model 123-1, the destination selection rate forecasting model 123-2, and the departure point selection rate forecasting model 123-3.

[0022] In this embodiment, the target area for prediction on the map is divided into multiple regions (meshes). Each region is represented by multiple channels. A "channel" is information (feature quantity) that indicates the characteristics of the region, such as a channel for the population, a channel for the number of hospital beds, a channel for the number of train stations, a channel for the number of parks, etc. Channels are generated based on "region information." Region information indicates the characteristics of the region. For example, it may include population information by age group, facility information by category, elevation difference from surrounding regions, weather information including precipitation, and information on the proportion of public holidays.

[0023] The target area information storage unit 121 stores area information of the target area on the map that is the subject of the prediction. This area information is associated with the target area specification information (for example, the central latitude and longitude of the area) entered by the user.

[0024] The hypothetical information storage unit 122 stores hypothetical information about the virtual reservation entered by the user making the virtual reservation. The hypothetical information is hypothetical information about the user at the time of prediction. For example, the hypothetical information is the number of active users, which is the number of users in the target area.

[0025] The mobility demand forecasting model 123-1 is a model that uses a convolutional neural network (Conventional Neural Network) that is trained to take feature quantities of the region information of a target area as input and output the number of mobility demands that will occur in the target area within a certain period of time. The mobility demand forecasting model 123-1 is trained using the region information of the learning area and the number of mobility demands that will occur in the learning area within a certain period of time.

[0026] The destination selection rate prediction model 123-2 is a model that uses a trained convolutional neural network that takes feature quantities of region information of a target area as input and outputs the rate at which that region is selected as a destination. The destination selection rate prediction model 123-2 is trained using the region information of the learning area and the rate at which that region is selected as a destination.

[0027] The starting point selection rate prediction model 123-3 is a model that uses a trained convolutional neural network to input feature quantities of region information of a target area and predict the rate at which that region is selected as a starting point. The starting point selection rate prediction model 123-3 is trained using the region information of the learning area and the rate at which that region is selected as a starting point.

[0028] The virtual reservation generation program 124 is a program that performs processing according to the embodiment and is executed by the control unit 104.

[0029] (Control Unit 104) The control unit 104 is a controller, and is implemented, for example, by a CPU (Central Processing Unit) or MPU (Micro Processing Unit) executing various programs (for example, a virtual reservation generation program 124) stored in the memory unit 105 inside the virtual reservation generation device 100, using RAM as the working area. Alternatively, the control unit 104 is a controller and can be implemented, for example, by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array).

[0030] The control unit 104 includes a calculation unit 110, a prediction unit 111, and a virtual reservation generation unit 115. The prediction unit 111 includes a travel demand prediction unit 112, a destination selection rate prediction unit 113, and a departure point selection rate prediction unit 114. Note that the internal configuration of the control unit 104 is not limited to the configuration shown in Figure 1, and other configurations are also acceptable as long as they perform the information processing described later. Furthermore, each functional unit indicates the function of the control unit 104 and does not necessarily have to be physically separated.

[0031] The calculation unit 110 calculates feature quantities related to the user's travel demand for each of the multiple regions into which the target area on the map is divided, and outputs them to the prediction unit 111.

[0032] The prediction unit 111 predicts the number of travel demands, the destination selection rate, and the departure point selection rate for the target area based on the feature quantities of each of the multiple domains. The prediction unit 111 includes a travel demand prediction unit 112, a destination selection rate prediction unit 113, and a departure point selection rate prediction unit 114.

[0033] (Mobility demand forecasting unit 112) The mobility demand forecasting unit 112 inputs area information of the target area and assumption information for forecasting into the mobility demand forecasting model 123-1 and obtains the number of mobility demands that will occur in the target area within a certain period (for example, within 4 weeks).

[0034] Figure 2 shows the input / output relationship of the mobility demand forecasting unit 112 according to the embodiment. As shown in Figure 2, static information, namely the target area 211, and dynamic information, namely the assumption information for the forecast 212, are input to the mobility demand forecasting unit 112. The mobility demand forecasting unit 112 then outputs the mobility demand 213 that will occur within the target area within a certain period (for example, 4 weeks).

[0035] The information 211 for the target area includes, for example, population information by age group for the area, facility information by category (e.g., hospitals, grocery stores, etc.), elevation difference from surrounding areas, weather information including precipitation, and information on the proportion of public holidays. The weather information and the proportion of public holidays may be changed depending on the period being predicted.

[0036] The virtual reservation generation device 100 acquires the target area information 211 based on information specifying the target area entered by the user of the virtual reservation generation device 100 (for example, the latitude and longitude of the center of the target area). If the location of the center of the target area is entered, the target area is defined as a rectangular area within a predetermined distance from the center. For example, the target area is defined as an area of ​​20 km from the center. The method of acquiring the target area information 211 is not specified. The acquired target area information 211 is stored in the target area information storage unit 121.

[0037] The assumption information 212 used during the prediction is, for example, the number of active users in the target area assumed during the prediction. The channels (features) of this target area information 211 and the assumption information 212 used during the prediction are input to the mobility demand prediction unit 112.

[0038] Figure 3 is a diagram illustrating an example of the relationship between the channels input to the mobility demand forecasting unit 112 according to this embodiment and the target area 321. As shown in Figure 3, the target area 321 is divided into multiple regions 322.

[0039] Each region 322 has multiple channels 323. In Figure 3, we assume that one region 322 has N channels. In Figure 3, we show the channels of 2D data for population channel 323-1 and hospital bed number channel 323-2. In Figure 3, in population channel 323-1, the upper left region 322 has a population of 0, and the region 322 to its right has a population of 50. Similarly, in hospital bed number channel 323-2, the upper left region 322 has 0 hospital beds, and the region 322 to its right has 10 hospital beds. The 2D data for N channels is input to the mobility demand forecasting unit 112. The mobility demand forecasting unit 112 outputs mobility demand 324 occurring within the area using the mobility demand forecasting model 123-1.

[0040] (Destination selection rate prediction unit 113) The destination selection rate prediction unit 113 inputs area information of the target area 321 and assumption information for the prediction into the destination selection rate prediction model 123-2 and obtains the rate at which the area is selected as a destination.

[0041] Figure 4 is a diagram showing the input / output relationship of the destination selection rate prediction unit 113 according to the embodiment. As shown in Figure 4, static information, namely the target area 431, and dynamic information, namely the hypothetical information at the time of prediction 432, are input to the destination selection rate prediction unit 113. The destination selection rate prediction unit 113 then outputs the percentage of each region 322 within the target area 321 that is selected as a destination.

[0042] The information 431 for the target area includes, for example, population information by age group, facility information by category, elevation difference from surrounding areas, travel time by vehicle, weather information including precipitation, and information on the proportion of public holidays.

[0043] The assumption information 432 used during the prediction process includes, for example, the usage history of the target area in the previous week, the number of new users, the user repeat rate, and the arrangement of boarding and alighting points. The channels (features) of this target area information 211 and the assumption information 212 used during the prediction process are input to the destination selection ratio prediction unit 113.

[0044] Figure 5 shows an example of a region of interest 541 and a surrounding region 542 in the target area 321 according to the embodiment. In Figure 5, the region of interest 541 and the surrounding region 542 around the region of interest 541 are shown in the target area 321 on the map. The surrounding region 542 represents a 3x3 area around the region of interest 541, but is not limited to this. For example, the surrounding region 542 may be an 11x11 area around the region of interest 541.

[0045] Figure 6 shows the channels of the area of ​​interest 541 input to the destination selection ratio prediction unit 113 according to the embodiment. In Figure 6, channel 323 is a channel generated based on information about the target area of ​​the area of ​​interest and assumptions made during prediction. As shown in Figure 6, the upper left region of a channel stores 0, and the region to its right stores 50. Channel 643 is a channel that indicates the time required to move from the surrounding region to the area of ​​interest. In Figure 6, the upper surrounding region of the area of ​​interest stores 5, 7, and 5 minutes, the left region stores 5 minutes, the right region stores 4 minutes, and the lower region stores 3, 4, and 4 minutes. By providing such channels, the destination selection ratio prediction unit 113 can take into account traffic conditions such as the difficulty of moving to the area of ​​interest. The destination selection ratio prediction unit 113 uses channel 643 to calculate the weight of the regions in the target area, converts them into ratios, and probabilistically determines the destination of the generated demand.

[0046] Figure 7 shows the weights of the regions in the target area 321 whose destination is the region of interest, as calculated by the destination selection ratio prediction unit 113 according to this embodiment. A weight is calculated for each region of the target area 321. In Figure 7, only the weights of the region of interest 541 and the surrounding region 542 are shown, and the other regions are omitted. Specifically, the destination selection ratio prediction unit 113 calculates the weight of the entire target area 321 while shifting the region of interest 541. Next, the destination selection ratio prediction unit 113 divides the regions of the target area 321 into proportions so that the sum of the weights of the target area 321 is 1. Then, the destination selection ratio prediction unit 113 probabilistically determines the destination of the generated demand. In Figure 7, the weight of the region of interest 541 is 30, and the weight of the surrounding region 542 is 10.

[0047] (Departure location selection rate prediction unit 114) The departure point selection rate prediction unit 114 inputs the area information of the target area 321 and the assumption information for the prediction into the departure point selection rate prediction model 123-3 and obtains the percentage of the area of ​​the target area that is selected as the departure point.

[0048] Figure 8 shows the input / output relationship of the departure point selection rate prediction unit 114 according to the embodiment. As shown in Figure 8, static information, namely the target area 861, and dynamic information, namely the assumption information at the time of prediction 862, are input to the departure point selection rate prediction unit 114. The departure point selection rate prediction unit 114 then outputs the percentage of each region 322 within the target area 321 that is selected as the departure point. Specifically, the departure point selection rate prediction unit 114 predicts the percentage of surrounding regions 542 that become the departure point when a certain area of ​​interest 541 is designated as the destination region. The surrounding region 542 is, for example, 5 regions (±5 km) around the area of ​​interest 541. In Figure 8, it is shown that the surrounding region 542 above the area of ​​interest 541 accounts for 20%, the surrounding region 542 to the right accounts for 20%, the surrounding region 542 below accounts for 30%, and the surrounding region 542 to the lower right accounts for 30%.

[0049] The information 861 for the target area includes, for example, population information by age group, facility information by category, elevation difference with surrounding areas, travel time by vehicle, and the percentage of public holidays.

[0050] The hypothetical information 432 used during the prediction process is, for example, information about the arrangement of boarding and alighting points. The channels (features) of this target area information 8611 and the hypothetical information 862 used during the prediction process are input to the destination selection ratio prediction unit 113.

[0051] Figure 9 is a diagram showing the weights that indicate where in the surrounding areas 542 the demand generated in the area of ​​interest 541 originates, as calculated by the departure point selection ratio prediction unit 114 according to the embodiment. In Figure 9, the surrounding areas 542 represent a 3x3 area, but are not limited to this. For example, the surrounding areas 542 may be an 11x11 area. In Figure 9, the weights of 30, 10, and 0 are stored for the three upper surrounding areas 542 relative to the area of ​​interest 441, the weight of 10 is stored for the left surrounding area 542, the weight of 30 is stored for the right surrounding area 542, and the weights of 10, 0, and 30 are stored for the three lower surrounding areas 542.

[0052] The departure point selection ratio prediction unit 114 divides the weights of the surrounding region 542 so that the sum of the weights equals 1. Next, the departure point selection ratio prediction unit 114 probabilistically determines the departure point of the generated demand.

[0053] (Virtual reservation generation unit 115) Let's return to Figure 1 for explanation. The virtual reservation generation unit 115 generates virtual reservations that show the reservation status of transportation from the departure point to the destination in the target area, based on the prediction results from the prediction unit 111. For example, the number of reservations from home to the hospital, the number of reservations from home to the restaurant, etc. Specifically, the virtual reservation generation unit 115 generates virtual reservations for the target area based on the predicted number of travel demands in the target area, the selection rate of destinations, and the selection rate of departure points.

[0054] [2. Operation] Figure 10 shows the virtual reservation generation process of the virtual reservation generation device 100 according to the embodiment.

[0055] The travel demand forecasting unit 112 of the virtual reservation generation device 100 acquires the number of travel demands that will occur in the target area within a certain period of time (step S1). Figure 10 shows the case where 100 demands occur in the target area.

[0056] Next, the destination selection rate prediction unit 113 of the virtual reservation generation device 100 obtains the rate at which an area is selected as a destination when movement occurs within the target area (step S2).

[0057] Next, the departure point selection rate prediction unit 114 of the virtual reservation generation device 100 outputs the percentage of each region within the target area that is selected as the departure point when a trip to a certain point as the destination occurs (step S3).

[0058] Subsequently, the virtual reservation generation unit 115 of the virtual reservation generation device 100 generates virtual reservations for the target area (step S4) based on the number of travel demands obtained in step S1, the percentage selected as a destination obtained in step S2, and the percentage selected as a departure point obtained in step S3. Note that the order of processing in steps S1 to S3 may be different.

[0059] [effect] Figure 11 is a diagram illustrating the effects of the virtual reservation generation device 100 according to the embodiment. Figure 11(a) shows the target area demand forecast of the virtual reservation generation device 100 using a CNN model according to the embodiment. Figure 11(b) shows the target area demand forecast of the virtual reservation device using a multiple regression model.

[0060] Figures 11(a) and 11(b) are histograms showing the error of area demand forecasting for a dataset of area demand for a given date and region. For example, 0.0 indicates a range of 0.0 ≤ relative error < 0.1, and 0.1 indicates a range of 0.1 ≤ relative error < 0.2. Here, the relative error is calculated as relative error = (actual demand - demand given by the forecaster) / absolute value of the actual demand.

[0061] As shown in Figure 11(a), the virtual reservation generation device 100 of the embodiment achieves a relative error of 20% or less in demand forecasts 75% of the time, and a relative error of 30% or less in 83% of the time. On the other hand, the virtual reservation generation device using a multiple regression model achieves a relative error of 20% or less in demand forecasts 35% of the time, and a relative error of 30% or less in 70% of the time. Therefore, it can be seen that the virtual reservation generation device 100 of the embodiment has higher accuracy in forecasting travel demand than the virtual reservation device using a multiple regression model.

[0062] According to the virtual reservation generation device 100 of this embodiment, the virtual reservation generation unit 115 generates virtual reservations for the target area.

[0063] Therefore, for example, it is possible to predict travel demand in areas where the MaaS system has not yet been implemented. Furthermore, since users can generate virtual reservations simply by specifying the target area, it is possible to rapidly expand business into areas where the MaaS system has not yet been implemented.

[0064] [3. Other Embodiments] The processes described in each embodiment above may be carried out in various other forms besides those described above.

[0065] Furthermore, among the processes described in each of the above embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically by known methods. In addition, the processing procedures, specific names, and information including various data and parameters shown in the above document and drawings can be changed at will unless otherwise specified. For example, the various information shown in each figure is not limited to the information shown.

[0066] Furthermore, the components of each illustrated device are functionally conceptual and do not necessarily need 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.

[0067] Furthermore, the embodiments and modifications described above can be combined as appropriate, provided that the processing content is not inconsistent.

[0068] Furthermore, the effects described herein are merely illustrative and not limiting; other effects may also occur.

[0069] [4. Hardware Configuration] Figure 12 is a hardware configuration diagram showing an example of a computer 1000 that implements the arithmetic unit of the virtual reservation generation device 100, which is an information processing device according to the embodiment.

[0070] Computer 1000 has a CPU 1100, RAM 1200, ROM (Read Only Memory) 1300, HDD (Hard Disk Drive) 1400, a communication interface 1500, and an input / output interface 1600. The various parts of computer 1000 are connected by a bus 1050.

[0071] The CPU 1100 operates based on programs stored in the ROM 1300 or HDD 1400, and controls various parts. For example, the CPU 1100 loads the programs stored in the ROM 1300 or HDD 1400 into the RAM 1200 and executes processing corresponding to various programs.

[0072] ROM1300 stores boot programs such as the BIOS (Basic Input Output System) executed by CPU1100 when computer 1000 starts up, as well as programs that depend on the computer 1000's hardware.

[0073] HDD1400 is a computer-readable recording medium that non-temporarily records programs executed by CPU1100 and data used by such programs. Specifically, HDD1400 is a recording medium that records an application program related to this disclosure, which is an example of program data 1450.

[0074] The communication interface 1500 is an interface for the computer 1000 to connect to an external network 1550 (e.g., the Internet). For example, the CPU 1100 can receive data from other devices or transmit data it generates to other devices via the communication interface 1500.

[0075] The input / output interface 1600 is an interface for connecting the input / output device 1650 and the computer 1000. For example, the CPU 1100 receives data from input devices such as a keyboard or mouse via the input / output interface 1600. The CPU 1100 also transmits data to output devices such as a display, speaker, or printer via the input / output interface 1600. The input / output interface 1600 may also function as a media interface for reading programs recorded on a predetermined recording medium (media). Examples of media include optical recording media such as DVDs (Digital Versatile Discs) and PDs (Phase Change Rewritable Disks), magneto-optical recording media such as MOs (Magneto-Optical Disks), tape media, magnetic recording media, or semiconductor memory.

[0076] The CPU 1100 reads and executes the program data 1450 from the HDD 1400, but as an alternative, these programs may be obtained from other devices via the external network 1550.

[0077] While preferred embodiments of the present disclosure have been described in detail above with reference to the attached drawings, the technical scope of the present disclosure is not limited to such examples. It is clear to any person with ordinary skill in the art of the present disclosure that various modifications or alterations may be conceived within the scope of the technical ideas described in the claims, and these will naturally also fall within the technical scope of the present disclosure.

[0078] Furthermore, the effects described herein are merely descriptive or illustrative and not limiting. In other words, the technology relating to this disclosure may produce other effects that are obvious to those skilled in the art from the description herein, in addition to or instead of the effects described herein. [Explanation of Symbols]

[0079] 100 Virtual Reservation Generator 101 Communications Department 102 Input section 103 Output section 104 Control Unit 105 Storage section 110 Calculation Unit 111 Prediction Section 112 Transportation Demand Forecasting Department 113 Destination Selection Proportion Prediction Unit 114. Prediction section for the proportion of departure location selections 115 Virtual Reservation Generation Unit 121 Target Area Information Storage Unit 122 Hypothetical Information Storage Unit 123 AI Model Memory Unit 123-1 Transportation Demand Forecasting Model 123-2 Destination Selection Proportion Prediction Model 123-3 Prediction Model for Departure Destination Selection Proportion 124 Virtual Reservation Generation Program Information on the target areas: 211, 431, 861 212, 432, 862 Assumptions made during the prediction 321 Target Area 322 areas Channels 323, 643 324 Travel demand 541 Areas of Interest 542 Peripheral area

Claims

1. A calculation unit calculates feature quantities related to the user's travel demand, based on regional information concerning the regions belonging to each of the multiple regions into which the target area on the map is divided, and hypothetical information assumed during the prediction process. A prediction unit inputs the feature quantities corresponding to the aforementioned regional information and the feature quantities corresponding to the aforementioned hypothetical information as channels of the input data into a prediction model using a convolutional neural network, and predicts the proportion in which each of the multiple regions is selected as a destination, the proportion in which it is selected as a departure point, and the number of travel demands that will occur in the target area within a certain period of time, based on the output results of the prediction model. A generation unit generates virtual reservations that show the reservation status of transportation from a departure point to a destination in the target area, based on the proportion in which each of the aforementioned multiple areas is selected as a destination, the proportion in which each of the aforementioned multiple areas is selected as a departure point, and the number of travel demands that occur in the target area within a certain period of time. A virtual reservation generation device having the following features.

2. The aforementioned feature quantities are generated based on the correlation between the target region of interest and multiple surrounding regions around the region of interest. The virtual reservation generation device according to claim 1.

3. The aforementioned correlation is, This is the time it takes to move from the surrounding region to the region of interest. The virtual reservation generation device according to claim 2.

4. The aforementioned regional information is, The information obtained in accordance with the specified target area entered by the user, The virtual reservation generation device according to claim 1.

5. The aforementioned regional information includes at least one of the following: population information by age group, facility information by category, elevation difference from the surrounding area, weather information including precipitation, and the percentage of public holidays. The virtual reservation generation device according to claim 1.

6. An information processing method implemented by a program executed by an information processing device, A calculation process for calculating features related to user travel demand, using regional information about the regions belonging to each of the multiple regions into which the target area on the map is divided, and hypothetical information assumed during the prediction process. A prediction step in which features corresponding to the aforementioned regional information and features corresponding to the aforementioned hypothetical information are input as channels of the input data into a prediction model using a convolutional neural network, and based on the output results of the prediction model, the proportion in which each of the multiple regions is selected as a destination and the proportion in which it is selected as a departure point, and the number of travel demands that will occur in the target area within a certain period are predicted. A generation process that generates virtual reservations showing the reservation status of transportation from a departure point to a destination in the target area, based on the proportion in which each of the aforementioned multiple areas is selected as a destination, the proportion in which each of the aforementioned multiple areas is selected as a departure point, and the number of travel demands that occur in the target area within a certain period of time. A virtual reservation generation method including the following.

7. A calculation procedure for calculating features related to user travel demand, using regional information about the areas belonging to each of the multiple regions into which the target area on the map is divided, and hypothetical information assumed during the prediction process. A prediction procedure that involves inputting the feature quantities corresponding to the aforementioned regional information and the feature quantities corresponding to the aforementioned hypothetical information as channels of the input data into a prediction model using a convolutional neural network, and predicting the proportion in which each of the multiple regions is selected as a destination, the proportion in which it is selected as a departure point, and the number of travel demands that will occur in the target area within a certain period of time, based on the output results of the prediction model, A generation procedure for generating virtual reservations that show the reservation status of transportation from a departure point to a destination in the target area, based on the proportion in which each of the aforementioned multiple areas is selected as a destination, the proportion in which each of the aforementioned multiple areas is selected as a departure point, and the number of travel demands that occur in the target area within a certain period of time. A virtual reservation generation program that causes a computer to execute a command.

Images