Predictive simulation apparatus

Abstract

To propose a mechanism for predicting demographic information in a specific time zone in a predetermined region based on demographic information of the predetermined region under a prediction condition.SOLUTION: An information processing system 1000 performs predictive analysis on demographic information, based on predetermined demographic information including information on the number of people located in each of a plurality of predetermined region meshes included in a region to be analyzed, in a chronological order. A demographic estimation unit 2111 estimates data on people (including the number of people and attributes, such as gender or ages) for each second region mesh unit (estimation region mesh) which is smaller than a first region mesh unit, on the basis of the predetermined demographic information including information on the number of people located in a plurality of first region mesh units included in the region to be analyzed in "predetermined time (length of time or time zone)" or "location information" set as a prediction condition.SELECTED DRAWING: Figure 12

Description

[Technical field]

[0001] The present invention relates to a prediction simulation device that predicts future demographic trends for a specified analysis target area. [Background technology]

[0002] Conventionally, there are services that provide demographic information created using the mechanisms of mobile phone networks (see, for example, Non-Patent Document 1).

[0003] 22 to 24 are diagrams showing an overview of a service that provides such demographic information.

[0004] 22 to 24, "Mobile Spatial Statistics (registered trademark)" provided by NTT DoCoMo, Inc. is described as an example.

[0005] Such "Mobile Spatial Statistics (registered trademark)" is explained in, for example, Non-Patent Document 2.

[0006] "Mobile Spatial Statistics (registered trademark)" is demographic information created using the mechanisms of mobile phone networks. This spatial statistics is demographic information that shows only the number of people in a group, and so has the characteristic that it cannot identify individual mobile phone or smartphone users. In this respect, it differs from the treatment of so-called "personal information," and it is possible to provide third parties with demographic information for a certain area at a certain time without obtaining the consent of the person to whom the information belongs.

[0007] Here, this spatial statistics makes it possible to grasp the population distribution for every hour of Japan, 24 hours a day, 365 days a year. Moreover, this spatial statistics makes it possible to grasp the population structure by gender, age group, and residential area, making it possible to conduct analyses tailored to the purpose. Since the service started in October 2013, it has been widely used by the Japan Tourism Agency, local governments, private companies, etc.

[0008] More specifically, mobile phone networks periodically track the mobile phones located in each base station area so that users can make calls and send emails anytime, anywhere. Using this system, the number of mobile phones can be tallied and the population can be estimated by taking into account the penetration rate of mobile phone carriers. This is Mobile Spatial Statistics (registered trademark).

[0009] Demonstration experiments were carried out in the public sector, such as urban development and disaster prevention planning, and it was confirmed that this spatial statistics could be widely used in society. As mentioned above, the service was launched in October 2013 so that it can be used not only in the public sector but also in academic and industrial fields.

[0010] This spatial statistics makes it possible to analyze data by gender, age group, and residential area (domestic residents), and by country and region (foreign visitors to Japan), and by time period according to purpose.

[0011] As shown in Figure 22(a), the statistical target area is the nationwide DoCoMo service area, and the population distribution throughout the country can be grasped every hour. Therefore, the population distribution can be grasped every hour, 24 hours a day, 365 days a year.

[0012] As shown in Figure 22(b), this demographics makes it possible to grasp the population structure by gender and age group, covering the age range of 15 to 79 years old.

[0013] Furthermore, as shown in Figure 23, this spatial statistics makes it possible to grasp the population by residential area in a specific time period (every hour) in a specific area, which, as mentioned above, makes it possible to grasp the population distribution by residential area based on registered mobile phone information.

[0014] Furthermore, assuming that spatial statistical information is available, technology has been proposed for understanding and predicting "people flow" using demographic information as input data (see, for example, non-patent document 3).

[0015] According to Non-Patent Document 3, "demographic data" can be classified into "trajectory data" and "aggregated data." "Trajectory data" is data related to moving objects such as people, and is often acquired by portable terminals such as smartphones and car navigation systems.

[0016] On the other hand, "aggregated data" is data associated with a location, and is collected by traffic volume sensors installed on roads, wireless base stations, and so on. Because data is acquired by equipment installed in specific locations, once the equipment is installed, data can be collected fairly comprehensively. In addition, because the data is not tied to individual subjects, there are fewer privacy concerns, and it is highly likely that data measured once can be used for a variety of purposes. On the other hand, because information across locations cannot generally be obtained, there are limitations to its use when understanding the flow of people in a city.

[0017] One type of aggregate data is "demographic information," which is data that records the population by location and time in a city. Most demographic information records population information by area and time. When targeting information over a wide area, mobile phone information is often used in consideration of comprehensiveness, and the number of terminals in each area is counted based on terminal connection information at mobile phone base stations and information such as CDR (Call Data Record), and the actual population is estimated while taking into account the utilization rate by attribute, and this is used as demographic information.

[0018] The "spatial statistics" described in the above-mentioned Non-Patent Documents 1 and 2 correspond to "demographic information" in this sense.

[0019] FIG. 24 is a diagram showing a data aggregation procedure in "Mobile Spatial Statistics (registered trademark)" disclosed in Non-Patent Document 2.

[0020] Referring to FIG. 24, in this "spatial statistics", the following processing is carried out to ensure safety by eliminating information that can identify individuals.

[0021] As mentioned above, in order to strictly protect customer privacy, this spatial statistic is demographic information that represents only the number of people in a group, and therefore cannot identify individual users.

[0022] The spatial statistics are created through the following steps: i) De-identification: The process of removing identifying information, such as name, telephone number, date of birth, etc. from operational data. ii) Aggregation processing: Processing to derive statistical "group-related information" by making statistical inferences from de-identified information. iii) Confidentiality processing: Processing to prevent figures from small population areas from being included in the tabulation results

[0023] In addition, Non-Patent Document 3 discloses a technology for performing people flow analysis in order to more effectively utilize demographic information, in which input is demographic information corresponding to multiple time periods, and based on that, it executes a task of outputting the amount of movement between cells between times.

[0024] On the other hand, when using two-dimensional information, there is a technology for a system that distributes demographic data of large meshes to small meshes based on estimated information of building area (see, for example, Patent Document 1).

[0025] Alternatively, Patent Document 2 discloses a sales information analysis device that creates sales activity support information useful to product sellers by analyzing sales performance information based on area clusters where product purchasers reside. Here, the sales information analysis device holds the correspondence between each area and area clusters, which are groups of areas with similar attributes related to residents. The sales information analysis device accesses a product sales information storage unit to calculate the sales volume of the product to be analyzed in each area cluster. The sales information analysis device accesses a store customer information storage unit to calculate the residence ratio of customers of the store to be analyzed in each area cluster. The sales information analysis device calculates an index value indicating the degree to which customers of the store to be analyzed support the product to be analyzed using a predetermined evaluation function that is positively correlated with the sales volume and the residence ratio.

[0026] Patent Document 3 discloses a system for predicting human behavior based on behavioral statistical data showing the occurrence probability for each time period for each of a plurality of human behavior types. Here, the prediction system inputs a plurality of conditions including a prediction period, an annual consumption expenditure, and a usage ratio of each behavior type belonging to a group of a plurality of mutually substitutable behavior types among a plurality of human behavior types, calculates a consumption time ratio for each behavior type in the group based on the usage ratio, satisfies a constraint condition that the ratio of the estimated occurrence probability of each behavior type in the prediction period in the group is equal to the consumption time ratio, and a constraint condition that the sum of the expenditure amount per unit time for each behavior type is equal to the expenditure amount per unit time calculated from the annual consumption expenditure, and calculates the estimated occurrence probability of each of the plurality of behavior types so that the Kullback-Leibler measure between the a priori probability distribution of the plurality of behavior types is minimized, and calculates the behavior time of each behavior type in the prediction period.

[0027] FIG. 25 shows an overview of the "Basic Survey on Social Life" (Non-Patent Document 4) issued by the Statistics Bureau of the Ministry of Internal Affairs and Communications.

[0028] As shown in Figure 25, the Basic Survey on Social Life surveys collects lifestyle information (behavioral classification) in 15-minute increments by region, age, and sex. [Prior art documents] [Patent documents]

[0029] [Patent Document 1] Patent No. 6226493 [Patent Document 2] Patent No. 4897713 [Patent Document 3] Patent No. 4834624 [Non-patent literature]

[0030] [Non-Patent Document 1] https: / / mobaku.jp / [Non-Patent Document 2] https: / / www.bcm.co.jp / bcm / wp-content / uploads / 2018 / 03 / ntt-docomo-2017-10-01.pdf [Non-Patent Document 3] Hiroyuki Toda, author of "Understanding and Predicting People Flow in Cities" https: / / www.jstage.jst.go.jp / article / oubutsu / 90 / 8 / 90_481 / _pdf / -char / ja [Non-Patent Document 4] https: / / www.stat.go.jp / data / shakai / 2021 / gaiyou.html Summary of the Invention [Problem to be solved by the invention]

[0031] However, the techniques described in Non-Patent Document 1 and Non-Patent Document 2 merely provide statistics such as the age (generation), sex, etc. of people present in a certain area on an hourly basis.

[0032] In addition, the technology described in Non-Patent Document 3 estimates "people flow" (the number of people moving) from "demographic data," but does not disclose how to display demographic information for a certain area at a certain time in the future in a way that is easy for users to understand.

[0033] The information in Non-Patent Document 4 merely indicates what activities people in different occupations are engaged in at certain times.

[0034] Furthermore, the technology described in Patent Document 1 merely discloses a technology for estimating the number of people in a building based on demographic data, while the technology described in Patent Document 2 calculates an index value indicated by a customer for a specific product in a given area based on the product sales volume, etc., and does not predict customer behavior.

[0035] For this reason, even if demographic data is accumulated, it is difficult to estimate the movement of people in a specified area based on demographic information at intervals shorter than the time interval at which the demographic data is accumulated. In addition, the technology described in Patent Document 3 predicts the type of human behavior for each time period, but does not predict the movement of people in a specified area at a specified time based on the movement of people in the target area.

[0036] The present invention has been made to solve the problems described above, and its object is to provide a future prediction technology device and a prediction simulation device that estimates the movements of people in a target area based on demographic information of a specified area, and simulates the behavior of people in each area included in the specified area at a specified time. [Means for solving the problem]

[0037] In order to achieve the above object, according to one aspect of the present invention, a predictive simulation device for an analysis target area includes a storage device that stores predetermined demographic information including information on the number of people located in each of a plurality of predetermined area meshes included in the analysis target area in a time series of a series of first time units, and a calculation device that analyzes the demographic information, and the calculation device includes an analysis target condition acquisition means for receiving a user's specification of prediction conditions, a human behavior information estimation means for estimating, in the prediction conditions, the number of people for each classification of human behavior located in the predetermined area mesh included in the analysis target area for each second time unit shorter than the first time unit based on the predetermined demographic information, a demographic estimation means for estimating, in the prediction conditions, aggregated data for each estimated area mesh in the analysis target area by distributing the number of people estimated by the human behavior information estimation means to the estimated area mesh based on the estimation result of the human behavior information estimation means and the type of place and the size of the place corresponding to the classification of human behavior for each estimated area mesh divided into the predetermined area mesh, and an estimation means for estimating the distribution of human attributes in the analysis target area based on the classification of human behavior and the aggregated data estimated in the estimated area mesh in the analysis target area.

[0038] Preferably, the prediction conditions include a first condition in which the user virtually specifies at least one of the type of place and the size of the place for each estimated area mesh, and a second condition in which no such specification is made, and the demographics estimation means further estimates aggregated data for each estimated area mesh within the analysis target area by distributing demographic information to the estimated area mesh based on the estimation results of the human behavior information estimation means and the human behavior for each estimated area mesh corresponding to at least one of the virtually specified type of place and the size of the place, and the estimation means outputs the results of estimating the number of people for each estimated mesh in the analysis target area based on the first condition and the second condition, respectively.

[0039] Preferably, the demographic information is predicted demographic information for a future time point specified in the prediction conditions based on predetermined statistical information, and the type of place and the size of the place are information planned for the place at the future time point specified in the prediction conditions.

[0040] Preferably, the storage device stores human behavior performance information indicating past human behavior performance in the same specified time units as the series of time series, and the human behavior information estimation means estimates predictive data related to human behavior in estimated area mesh units in the specified time units using a model generated from the human behavior performance information and the results of the demographics estimation means.

[0041] Preferably, the storage device stores demographic information, lifestyle information, and area location information as data acquired from an external server via a network, the demographic information being aggregated data of demographic data for a first time interval corresponding to a first time unit, and being information on the number of people by age and sex present in a predetermined area including the analysis target area for each predetermined area mesh at the first time interval, the lifestyle information being information for each predetermined time unit, the predetermined time unit being a second time interval shorter than the first time interval and corresponding to a second time unit, and the lifestyle information being information on the age, sex, and life and behavior of people. The area location information is location information for each predetermined section, and includes information on the total floor area of ​​facilities within the predetermined area and the occupations within the facilities. The human behavior information estimation means includes an area life information estimation means for estimating the number of employed people by occupation located in the specified area for each second time interval based on the life information for each predetermined section, the demographic information and the life information. The demographic estimation means apportions the number of employed people by occupation to each facility based on the total floor area and the area allocated to the occupation of each facility, and estimates aggregate data of the number of people in the estimated area mesh of the demographic information.

[0042] Preferably, the model is a model that estimates the number of people associated with a specified human behavior as prediction data, and further comprises a human behavior information designation receiving means that accepts designation of the number of people for the human behavior to be analyzed in a specified time unit and estimated area mesh unit, and a required number of people information output means that outputs required number of people information indicating the number of people in a specified time unit and specified area unit required for the specified number of people to perform the human behavior in the specified time unit and estimated area mesh unit.

[0043] Preferably, the specified person behavior is purchasing behavior, and the model is a model that estimates the number of people who will perform the specified purchasing behavior and the average customer spending as prediction data.

[0044] Preferably, the demographics prediction information for a future point in time includes first prediction information in the case where a specified event occurs and second prediction information in the case where the specified event does not occur, and the human behavior information estimation means estimates the number of people for each classification of human behavior corresponding to each of the first prediction information and the second prediction information, and the demographics estimation means further estimates aggregated data for each estimated area mesh within the analysis target area by distributing the number of people estimated by the human behavior information estimation means to the estimated area mesh, and the estimation means outputs the results of estimating the number of people for each estimated mesh within the analysis target area based on the first prediction information and the second prediction information, respectively. Effect of the Invention

[0045] According to the present invention, it is possible to predict and simulate the behavior of people in area units contained in a predetermined area at a predetermined time based on demographic information and demographic information of the target area. [Brief description of the drawings]

[0046] [Figure 1] FIG. 1 is a diagram showing an overview of a lifestyle information estimating system according to a first embodiment of the present invention. [Diagram 2] FIG. 2 is a functional block diagram showing an outline of the system of a life information service providing server 2000. [Diagram 3] 2 is a diagram for explaining the hardware configuration of a life information service providing server 2000. FIG. [Figure 4] FIG. 1 is a diagram for explaining the contents of data provided from data providing servers 5000.1 to 5000.M. [Diagram 5] FIG. 11 is a first diagram illustrating a processing flow of the life information service providing server 2000. [Figure 6] FIG. 2 is a second diagram illustrating the processing flow of the life information service providing server 2000. [Figure 7] FIG. 2 is a diagram showing an outline of data output from a calculation device 2100. [Figure 8]FIG. 13 is a conceptual diagram showing data generated when facility occupancy estimation is performed from behavior-specific number of people estimation at a first time point. [Figure 9] FIG. 13 is a conceptual diagram showing data generated when facility occupancy estimation is performed from behavior-specific number of people estimation at a second time point. [Figure 10] FIG. 13 is a diagram showing the concept of the process of improving the accuracy of spatial resolution in the behavior-specific number of people estimation process (S120, S210). [Figure 11] 13 is a diagram showing another example of the map image and the superimposed image of the number of people distribution output by the map output unit 2104. FIG. [Figure 12] FIG. 2 is a diagram showing functional blocks when the life information service providing server 2000 functions as a prediction simulation device. [Figure 13] FIG. 11 is a diagram illustrating a flow of a prediction simulation process. [Figure 14] This is a summary of performance information for each estimated area mesh included in the analysis target area AR. [Figure 15] FIG. 1 is a diagram illustrating an example in which the predictive simulation device is used for urban planning. [Figure 16] FIG. 13 is a conceptual diagram for explaining a “prediction model.” [Figure 17] FIG. 1 is a conceptual diagram showing a hierarchy of data and a hierarchy of data processing used in the lifestyle information estimation system. [Figure 18] FIG. 2 is a conceptual diagram showing an overview of the flow of data processing. [Figure 19] FIG. 1 is a diagram showing an overview of services realized by using a lifestyle information estimating system 1000. [Figure 20] FIG. 20 is a conceptual diagram showing an example of the service of FIG. 19 provided via a smartphone. [Figure 21] FIG. 13 is a diagram showing another example of data visualization by integrated analysis. [Figure 22] FIG. 1 is a diagram showing an overview (1) of mobile spatial statistics of the prior art. [Figure 23] FIG. 1 is a diagram showing an overview (2) of mobile spatial statistics of the prior art. [Figure 24] FIG. 1 is a diagram showing an overview (3) of mobile spatial statistics according to the prior art. [Diagram 25] FIG. 1 is a diagram showing the contents of the Basic Survey on Social Life as a background art. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0047] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0048] The following describes the configuration of a prediction simulation device that simulates the behavior of people in area units contained in a specified area at a specified time in the future based on the number of people in area units contained in a specified area over a series of time units (including prediction data based on behavioral history) and prediction conditions.

[0049] As described below, the predictive simulation device in this embodiment is a device that estimates human movement and the distribution of the number of people within a specified area during a specified time period under set prediction conditions, based on demographic information collected as a series of time series for the specified area.

[0050] In another aspect, the predictive simulation device of this embodiment also functions as a simulation device that simulates the behavior of people in a specified area at a specified time, based on the number of people in a specified area and their behavioral history in that area over a series of time units.

[0051] FIG. 1 is a diagram showing an overview of a demographic analysis system according to a first embodiment of the present invention.

[0052] Referring to FIG. 1, demographic analysis system 1000 of embodiment 1 of the present invention includes a life information service providing server 2000 operating as a demographic analysis device, a life information user terminal 3000 (for example, there may be a plurality of terminals 3000.1 to 3000.N, where N is a natural number of 2 or more. Hereinafter, collectively referred to as "life information user terminal 3000") of a user receiving services from life information service providing server 2000, data providing servers 5000.1 to 5000.M (M: a natural number) that provide life information service providing server 2000 with data necessary for estimating life information as described below, and a mobile phone or smartphone 200 (for example, there may be a plurality of smartphones 200.1 to 200.M, where M is a natural number of 2 or more. Hereinafter, collectively referred to as "smartphone 200") used by user 100 to receive services as described below.

[0053] FIG. 2 is a functional block diagram showing an outline of a system when the life information service providing server 2000 functions as a life information estimating device.

[0054] 4 is a diagram for explaining the contents of data provided from the data providing servers 5000.1 to 5000.M. Note that hereinafter, when the data providing servers 5000.1 to 5000.M are collectively referred to, they will simply be referred to as "data providing server 5000."

[0055] With reference to FIG. 2, the life information service providing server 2000 includes a computing device 2100 and a storage device 2300 .

[0056] The storage device 2300 first stores predetermined "demographic information 2302." Here, the "demographic information 2302" includes at least information on the number of people present in a predetermined area at a predetermined time.

[0057] Here, the "predetermined time" refers to, for example, a first predetermined time interval, for example, every hour, and the "predetermined area" refers to, for example, an area in a specific region, for example, a first area mesh unit, for example, a 500m mesh (a square of 500m on each side). The "predetermined demographic information" is "demographic data" and means "aggregated data", and corresponds to, for example, the above-mentioned "Mobile Spatial Statistics Information (registered trademark)" as shown in FIG. 4.

[0058] Furthermore, the storage device 2300 stores a predetermined "lifestyle information 2306". The "lifestyle information 2306" includes at least information related to a person's life. Here, the predetermined "lifestyle information 2306" is, more specifically, information on a classification of a predetermined behavior by region, age, and sex, in a second time interval that is shorter than a first predetermined time interval and is preferably a divisor of the first time interval. Here, the "lifestyle information" corresponds to information in a database based on the "Basic Survey on Social Life" by the Ministry of Internal Affairs and Communications, which classifies behaviors into 20 predetermined types by region, age, and sex, and in 15-minute increments, as shown in FIG. 4. Here, the "classification of behavior" corresponds to classifications such as "sleep", "commuting", "school", "work", and "academic work".

[0059] The storage device 2300 also stores "occupation-specific employment information 2308." Here, the "occupation-specific employment information 2308" is data showing the employment ratio of each occupation by age and sex, and corresponds to information in the Ministry of Internal Affairs and Communications' Labor Force Survey database, for example, as shown in FIG.

[0060] In addition, the "predetermined living information" may include such "occupation-specific employment information."

[0061] Therefore, the life information estimation unit 2110 can estimate the number of people by behavior (hereinafter, referred to as "number of people by behavior") who are present in the "predetermined area" set by the area setting unit 2106 in response to an instruction from the user, during the "second time interval", based on the "demographic information 2302" and the "predetermined life information 2306". Furthermore, the life information estimation unit 2110 can calculate the classification data of the "occupation" of people who are present in the "predetermined area" during the "second time interval", by applying the "work information by occupation" to people with the behavior classification of working or studying, based on the "number of people by behavior" estimated in this way. Note that the "occupation" does not only refer to the so-called "type of occupation", but may also refer to "tasks" such as "study" that are not business, but in which a person is obviously engaged in the same behavior for a certain period of time or more in that place (facility / building).

[0062] The storage device 2300 further stores a predetermined "area location information 2310". Here, the "predetermined area location information" refers to "information on a predetermined place existing in the above-mentioned predetermined area (including information on the type of job at that place)". For example, the "area location information" preferably includes information on buildings and facilities in the area of ​​the second area mesh unit, which is smaller than the above-mentioned first area mesh unit and preferably has a divisor of each side when the first area mesh unit is a rectangle, and information on the types of jobs performed in the buildings and facilities. For example, as shown in FIG. 4, the "area location information 2310" corresponds to map data including information on buildings and facilities in a town, information on the types of jobs performed in the buildings and facilities, and information on roads and water areas in 62.5m mesh units. Some of this "map data" is provided by private companies.

[0063] Based on such area location information 2310, the area information estimation unit 2108 estimates the area corresponding to each job type in the total floor area of ​​facilities and buildings located in the "specified location" of the "specified area" set by the area setting unit 2106.

[0064] Therefore, the lifestyle information estimation unit 2110 of the calculation device 2100 of the lifestyle information service providing server 2000 calculates and outputs information on the work of people located at a specified location in a specified area at the second time interval (hereinafter referred to as "people's area lifestyle information") by expanding (for example, proportionally allocating) the "number of people per occupation" into the "area corresponding to each occupation" of the "specified location."

[0065] For example, when calculating such "human area life information," the classification of "tasks" that people are engaged in at a certain location (facility / building) within the target area is calculated according to the method described above.

[0066] As a relationship between a certain life information and a certain area location information, for example, during a certain time period, it is possible to assume that doctors and nurses work in hospital facilities, farmers work on farmland, fishermen work in fishing grounds, housewives raise children, and pensioners live in houses. Also, during commuting hours, a certain percentage of people will have the attribute "movement." The specific processing will be described later.

[0067] Furthermore, the map output unit 2104 generates and outputs a screen that displays the thus estimated "personal area life information" in association with each "predetermined place" in the "predetermined area" (for example, superimposed on the map information). By generating such an output, the user can visually grasp the life information of people in the "predetermined area" that he or she has specified.

[0068] As described above, since the "number of people by behavior" or the "number of people corresponding to people's area life information" is calculated for each area in the second area mesh unit and at the second time interval, the people flow path information estimation unit 2102 can be configured to estimate and output people flow path information (the route along which people moved) by tracking changes in the number of people distribution for the area in the second area mesh unit and its surrounding areas along the time axis and integrating this with "information on routes along which people can move" (for example, roads, railways, etc.) on the map data. The output result of the people flow path information output unit 2102 is stored in the people flow path information 2312 of the storage device 2300.

[0069] The storage device 2300 further stores "traffic route information 2304." Here, the "traffic route information 2304" includes the following data located in a predetermined area: Route information on road maps, information on transportation route nodes such as the location of highway interchanges (ICs) - Average travel speed on a road at a given time on a given day based on historical road congestion information Location of public transport stations and stops (railroads, buses, etc.) Public transport (train, bus, etc.) routes on maps Public transport timetables (average travel speed can also be calculated based on timetable and route information)

[0070] However, as long as the information is related to transportation, the data included in the transportation route information 2304 is not limited to these.

[0071] The traffic route information 2304 stores information as described below. By relating pedestrian flow information to traffic route information, the pedestrian flow path information estimation unit 2102 can estimate information on the amount of people moving along each traffic route, and stores the estimation results in the storage device 2300 as pedestrian flow path information 2312.

[0072] FIG. 3 is a diagram for explaining the hardware configuration of the life information service providing server 2000. As shown in FIG.

[0073] 3, the life information service providing server 2000 may be configured such that a calculation device (CPU: Central Processing Unit) in its own housing executes calculation processing, or a part of the program processing may be further executed on another server. In the following, the description will be given assuming that the calculation device in its own housing executes the calculation processing.

[0074] Referring to FIG. 3, server 2000 includes a computer device 2010, a network communication unit 2012 for communicating with a network, a recording medium (e.g., a memory card) 2210 for recording data from the outside and providing the data to computer device 2010, a keyboard 2400 as an input device, and a display 2420 as a display device.

[0075] For example, a USB memory, a memory card, or an external storage device can be used as the recording medium 2210. Also, for example, a communication function of a wired LAN or a wireless LAN can be used as the network communication unit 2012. The network communication unit 2012 and the input / output interface 2090 configure a communication interface.

[0076] As shown in FIG. 3, the computer main body constituting the computer device 2010 includes, in addition to the disk drive 2030 and the memory drive 2020, a CPU (Central Processing Unit) 2100 connected to a bus 2050, a memory 2200 including a ROM (Read Only Memory) 2200.1 and a RAM (Random Access Memory) 2200.2, a non-volatile rewritable non-volatile storage device 2300, and an input / output interface 2090 for communicating via a network and sending and receiving data with the outside. For example, a HDD (Hard Disc Drive) or an SSD (Solid State Drive) is used as the non-volatile storage device 2300. In the following description, it is assumed to be an SSD. An optical disk can be attached to the disk drive 2030. A memory card 2210 can be attached to the memory drive 2020.

[0077] When the programs of the computer device 2010 operate, the data and programs that store the information that is the basis for the operation of the computer are stored in the SSD 2300.

[0078] 3, the medium capable of recording information such as a program to be installed in the computer main body may be, for example, a DVD-ROM (Digital Versatile Disc), a memory card, a USB memory, etc. In order to accommodate such cases, the computer main body is provided with a drive device (memory drive 2020, disk drive 2030) capable of reading such media.

[0079] The main components of the computer device 2010 are computer hardware and software executed by the CPU 2100. Generally, such software is stored in a storage medium and distributed or distributed via a network, and is acquired via the disk drive 2030 or the network communication unit 2012 and temporarily stored in the SSD 2300. The software is then read from the SSD 2300 to the RAM 2200.2 in the memory and executed by the CPU 2100. Note that, when connected to a network, the software may be directly loaded into the RAM and executed without being stored in the SSD 2300.

[0080] When a program for functioning as computer device 2010 is distributed, the program does not necessarily need to include an operating system (OS) that causes computer main body 2010 to execute functions of an information processing device, etc. The program only needs to include an instruction portion that calls appropriate functions (modules) in a controlled manner and produces desired results. How computer system 2010 operates is well known, and a detailed description will be omitted.

[0081] Furthermore, the CPU 2100 may be a one-core processor or a multiple-core processor, i.e., a single-core processor or a multi-core processor. The server 2000 may also be configured with multiple servers to execute distributed processing.

[0082] The hardware configurations of the lifestyle information user terminal 3000 and the data providing servers 5000.i (1≦i≦M) are basically the same, and therefore the description thereof will not be repeated. (Estimation process of life information within a specified area)

[0083] In the following, as a premise for the life information service providing server 2000 to predict future demographics, the process of estimating the number of people by behavior and the number of people in a facility based on the life information of people present in a specific area during a specific time period will be described.

[0084] Fig. 5 is a first diagram illustrating the processing flow of the number of people estimation by behavior and the number of people in a facility estimation by the life information service providing server 2000. Fig. 6 is a second diagram illustrating the processing flow of the number of people estimation by behavior and the number of people in a facility estimation by the life information service providing server 2000.

[0085] FIG. 7 is a diagram showing an outline of data output from the arithmetic device 2100. As shown in FIG.

[0086] FIG. 8 is a conceptual diagram showing data generated when facility population estimation is performed from behavior-specific population estimation at a first time point.

[0087] FIG. 9 is a conceptual diagram showing data generated when facility population estimation is performed from behavior-specific number of people estimation at a second time point.

[0088] 5 and 6, when the processing of the behavior-specific number of people estimation and facility number of people estimation starts, the calculation device 2100 acquires spatial statistical data via the network (S110, S200).

[0089] As mentioned above, to obtain spatial statistical data, it is possible to use "Mobile Spatial Statistics (registered trademark)" provided by NTT DoCoMo. Spatial statistical data is, for example, data on the number of people by age and sex in a given time unit (for example, 1 hour) within a mesh of a given width (for example, 500m square).

[0090] Next, for a "predetermined area" specified by the area setting unit 2106 based on the user's designation, the lifestyle information estimation unit 2110 classifies people (whose age and gender are obtained from the spatial statistical data) who are staying in a specific area (for example, the Umeda area of ​​Osaka) during a specific time period (for example, a certain hour) into states (behavior classification information) such as work and movement (for example, commuting) in more detailed time units (for example, 15 minutes) within this specific time period, referring to the Basic Survey of Social Life database, and estimates the number of people in each category (attributes of age and gender) (S120, S210).

[0091] The lower part of Figure 7(a) shows an overview of such estimation of the number of people by behavior.

[0092] The lower diagram in Figure 8 shows the estimated number of people by activity in a specific area at 2 a.m. Given the time of day, most people's activities are classified as "sleeping."

[0093] The lower graph in Figure 9 shows the estimated number of people by activity in a specific area at 3:00 p.m. (15:00). Judging from the time of day, most people are classified as "working" or "studying."

[0094] Returning to FIG. 5 and FIG. 6 , further, the lifestyle information estimation unit 2110 estimates the number of employees (age and gender attributes) by occupation type for every 15 minutes for the above-mentioned specific area based on the estimated number of people (age and gender attributes) for each behavior category, for example, based on the Ministry of Internal Affairs and Communications Labor Force Survey database (S130, S220).

[0095] Next, the lifestyle information estimation unit 2110 expands (for example, proportionally apportions) the estimated number of employees (attributes of age and gender) by occupation for every 15 minutes to each building (facility) based on the total floor area of ​​the facilities and buildings on the map estimated by the area location information estimation unit 2108 and the area allocated to the occupation of each facility and building, and estimates aggregated data of the number of people in a mesh (for example, a 62.5 m mesh) that is smaller than the unit mesh of the spatial statistical data (S140, S230).

[0096] The upper part of Fig. 7(a) shows an overview of such facility occupancy estimation.

[0097] The upper diagram in Figure 8 shows the results of estimating the number of people in facilities in a specific area at 2 a.m. Judging from the time of day, most people are located in "residences" or "hotels and inns."

[0098] The upper diagram in Figure 9 shows the results of facility population estimation in a specific area at 3:00 p.m. (15:00). Based on the time of day, it can be seen that most people are assigned to the "occupation" or "task" corresponding to the facility or building in each facility.

[0099] 5 and 6, the map output unit 2104 further superimposes the number of people deployed for each facility / building on the map and outputs it by color-coding, for example, using a heat map method (S140, S240). Here, when displaying the heat map, a method of displaying a chart showing the ratio of the number of people doing the work at that location by color coding may be used.

[0100] FIG. 7(b) is a diagram showing an example of an output that visualizes such a distribution of the number of people.

[0101] FIG. 10 is a diagram showing the concept of the process of improving the accuracy of spatial resolution in the behavior-specific number of people estimation process (S120, S210).

[0102] As shown in Figure 10, when conducting analysis using various information such as spatial statistical data (or people flow data) and map information, some data may have a spatial resolution of 500m, other data may have a spatial resolution of 100m, and still other data may have a spatial resolution of land lot number (address unit).

[0103] Because the units of the data do not match, analysis usually ends up being performed based on data with larger units.

[0104] However, as described above, in the life information service providing server 2000, people flow data (including number distribution data at a specific point in time), building data, facility data, land data (including information on roads, railways, roads and water bodies), etc. are all decomposed and converted into the same unit (area mesh) and stored.

[0105] Hereafter, such a database stored in the same unit will be referred to as a "data lake." Therefore, for example, it is possible to perform real-time analysis in the same unit (area mesh) at any point in Japan.

[0106] As described above, the same area mesh has a side length of 62.5 m, as shown in FIG.

[0107] If the resolution is 62.5 m, since Mobile Spatial Statistics (registered trademark) is a mesh with sides of 500 m, it is possible to create an area mesh with sides of 1 / 8 (1 / 8) of this 500 m, which is approximately the distance a person walks in one minute. When using other spatial statistics, it is possible to unify the area mesh with sides equivalent to the distance a person walks in one minute in the same way. By using the distance a person walks in one minute, it can be said that the resolution is suitable for analyzing people flow, including people who are moving by foot, for example.

[0108] However, it is possible to divide it into units smaller than 62.5 m (for example, the smallest unit of 1 m). However, this would increase the amount of data to be analyzed and the amount of calculation, so from the viewpoint of cost-effectiveness, it is preferable to use, for example, 62.5 m.

[0109] FIG. 11 is a diagram showing another example of the map image and the superimposed image of the number of people distribution output by the map output unit 2104. In FIG.

[0110] As shown in FIG. 11, it is also possible to designate an area of ​​a certain size as the specified region, and furthermore, to allow the user to specify a portion of that area as the "specified region" in which the estimation process will be performed. (Prediction processing at a future point in time)

[0111] The following describes the estimation process of demographic information of people in a specific area at a specific time period in the future and the display process thereof by the life information service providing server 2000. By estimating demographic information in this manner, the life information service providing server 2000 functions as a prediction simulation device.

[0112] FIG. 12 is a diagram showing functional blocks when the life information service providing server 2000 functions as a prediction simulation device.

[0113] The configuration in FIG. 12 illustrates the software functions for executing the "prediction process of future demographic trends" as described below in the configuration of the life information service providing server 2000 shown in FIG. 2, and the data configuration required for executing such functions.

[0114] There are many commonalities between the processing described in the configuration of Fig. 2 and the processing executed in the configuration of Fig. 12. Therefore, the functional blocks described in Fig. 2 may be omitted from illustration as appropriate, but the same functions are executed by the same functional blocks in Fig. 2 and Fig. 12. Furthermore, the same reference numerals are given to the same parts in Fig. 2 and Fig. 12, and descriptions thereof will not be repeated in principle.

[0115] 12, when performing a prediction process of demographics at a future time point, the arithmetic device 2100 of the life information service providing server 2000 executes the following functions in addition to the functions of FIG. 2: an analysis target time / area information acquisition unit 2101 that acquires prediction conditions for information on the area to be analyzed (hereinafter, "analysis target area information") and the time or time period to be analyzed, which are specified by the user via the user terminal 3000; a human behavior estimation unit 2103 that estimates a classification of human behavior at the above-mentioned specified second time interval for each area mesh based on specified demographic information; a human behavior information designation acceptance unit 2105 that acquires information on the behavior of people to be analyzed in the analysis target area (hereinafter, "specified human behavior information") by user designation on the user terminal 3000; and a number of people required information estimation unit 2107 that estimates the number of people required in the analysis target area to satisfy the specified number of people performing a specific behavior (including attribute information such as the number of people, gender, and age) specified by the user for the specific behavior (hereinafter, "number of people required information").

[0116] Furthermore, the location information provisional setting unit 2115 acquires provisional setting information of the location designated by the user as a prediction condition.

[0117] 12, the functions of the life information estimation unit 2110 in FIG. 2 will be explained by dividing them into three functions: a human behavior information estimation unit 2103, an area life information estimation unit 2109, and a vital movement information estimation unit 2111.

[0118] As in Fig. 2, the "demographic information 2302" includes information on the number of people present in a predetermined area at least at a predetermined time. Here, the "predetermined time" refers to, for example, a first predetermined time interval, for example, every hour, and the "predetermined area" refers to, for example, an area in a specific region, for example, a first area mesh unit, for example, a 500m mesh (a square of 500m on each side). The "predetermined demographic information" is "demographic data" and means "aggregated data", and corresponds to, for example, the above-mentioned "Mobile Spatial Statistics Information (registered trademark)" as shown in Fig. 4.

[0119] Furthermore, the "lifestyle information 2306" stored in the storage device 2300 includes at least information related to a person's life. Here, the predetermined "lifestyle information 2306" is, more specifically, information on classification of predetermined behaviors by region, age, and sex, in a second time interval that is shorter than a first predetermined time interval and is preferably a divisor of the first time interval. Here, the "lifestyle information" is, for example, information classified into 20 predetermined types of behaviors by region, age, and sex, and in 15-minute intervals, as shown in FIG. 4, and corresponds to information in a database based on the "Basic Survey on Social Life" by the Ministry of Internal Affairs and Communications.

[0120] The "employment information by occupation 2308" stored in the storage device 2300 is data showing the employment ratio of each occupation by age and sex, and corresponds to the information in the Ministry of Internal Affairs and Communications' Labor Force Survey Database, for example, as shown in Fig. 4. Note that, here too, such "employment information by occupation" may be included in the "predetermined life information".

[0121] Therefore, the human behavior estimation unit 2103 estimates the number of people by behavior (hereinafter referred to as the "number of people by behavior") who are present in the "predetermined area" set by the analysis target time / area information acquisition unit 2101 in response to an instruction from a user, during the "second time interval", based on the "demographic information 2302" and the "predetermined life information 2306". Furthermore, by applying the "occupation-specific employment information" to people with the behavior classification of working or studying, based on the "number of people by behavior" estimated in this way, the area life information estimation unit 2109 can calculate data on the number of people by occupation for the "occupation" classification of people who are present in the "predetermined area" during the "second time interval". Note that the "occupation" may not only be the so-called "type of occupation", but may also be a "job" such as "study" that is not a business, but in which a person is obviously engaged in the same behavior for a certain period of time or more in that place (facility / building).

[0122] The "area location information 2310" stored in the storage device 2300 is "information on a specific place existing in the above-mentioned specific area (including information on the type of job at that place)." Here, the "area location information" includes information on buildings and facilities within the area of ​​the second area mesh unit, which is smaller than the above-mentioned first area mesh unit and preferably has sides that are divisors of the first area mesh unit when the first area mesh unit is a rectangle, and information on the types of jobs performed within the buildings and facilities. For example, as shown in FIG. 4, the "area location information 2310" corresponds to map data in 62.5m mesh units, including information on buildings and facilities in a town, information on the types of jobs performed within the buildings and facilities, and information on roads and water areas.

[0123] Based on such area location information 2310, the area location information estimation unit 2108 estimates the area corresponding to each job type in the total floor area of ​​the facility or building located in the "predetermined place" of the "predetermined area".

[0124] Finally, the demographic information estimation unit 2111 of the calculation device 2100 of the life information service providing server 2000 calculates and outputs information on the number of people (demographics) present in each estimated area mesh during the second time interval by expanding (for example, proportionally allocating) the "number of people per occupation" into the "area corresponding to each occupation" of the "specific location."

[0125] In addition to the information stored in FIG. 2, the memory device 2300 stores analysis time / area information 2301 specified by the user, demographic information 2302 including information estimated by the demographic information estimation unit 2111 in addition to specified demographic information, human behavior performance information 2303 as described below, human behavior information 2305 estimated by the human behavior information estimation unit 2103, area life information 2311 estimated by the area life information estimation unit 2109, and statistical information 2313 as described below.

[0126] Here, the area life information 2311 is life information for each estimated area mesh about people who are present in the analysis target area at a given time, based on the "relationship between predetermined life information and predetermined area location information".

[0127] FIG. 13 is a diagram illustrating the flow of the prediction simulation process.

[0128] When the processing starts, the analysis target time / area information acquisition unit 2101 of the calculation device 2100 acquires the analysis target area information and the prediction conditions for the time or time period information to be analyzed in response to instructions from the user terminal 3000, and the location information provisional setting unit 2115 acquires the prediction conditions for the provisional setting of the location specified by the user (S102).

[0129] Here, the "prediction conditions" include the analysis target area and analysis target time (time period) to be predicted, as well as provisional location settings such as the following:

[0130] i) Buildings that actually exist in a specific area (whose total floor area and workplaces / occupations within the building are known) are removed from the calculation (for example, the building height is set to 0 floors).

[0131] By setting prediction conditions in this way and simulating people flow, it is possible to evaluate a building's ability to attract people by comparing conditions with and without the building.

[0132] ii) Conversely, a new building (setting the total floor area and the workplaces and occupations within the building) is calculated and registered in a specific area (for example, the building height is set to 5 stories).

[0133] By setting prediction conditions in this way and simulating people flow, it is possible to evaluate the building's potential for attracting people by comparing the state when the building does not exist (the current state) with the state when the building is constructed (a hypothetical future state).

[0134] Furthermore, in FIG. 13, the process from steps S110 to S140 is similar to the process of estimating life information described in FIG. 5, and therefore description thereof will not be repeated.

[0135] In practice, for example, the data calculated in the processing from steps S110 to S140 in the data lake may be temporarily stored as common data, and may be configured to be shared in the processing for estimating the number of people by behavior and estimating the number of people in facilities described in FIG. 5 and the processing for analyzing demographic information described in FIG. 13.

[0136] Therefore, by processing up to step S140, the demographics estimation unit 2111 estimates demographic information (people data (including attributes such as number of people, gender, age, etc.)) for each estimated area mesh in a state in which no particular prediction conditions have been set at this time.

[0137] Furthermore, following the processing of steps S110 to S140, the demographics estimation unit 2111 of the calculation device 2100 also refers to the human behavior record information and statistical information to set prediction conditions (prediction time, prediction area, and attributes of the prediction location) as requested by the user (S170).

[0138] Here, the human behavior record information includes, for example, the following information:

[0139] i) Information on human behavior over a series of time and area units (for example, statistics such as the Basic Survey of Social Life, and information on the purchase of goods and services based on POS (Point of Sales) data).

[0140] As will be described later, the information may include estimated behavioral performance (e.g., rough performance information or predicted / complemented performance information (which may be a prediction that takes into account the influence of events)) based on actually observed / detected behavioral performance using a model such as a "state space model."

[0141] The statistical information includes, for example, the following information:

[0142] ii) For example, information on "future estimated regional population" based on "demographic survey" + "future population growth rate calculation by age and sex". Furthermore, it may include correction by "census data" in a specified area mesh unit.

[0143] In setting the prediction conditions as described above, for example, if the prediction time is set to a future time, a prediction model can be used to predict data for the estimated area mesh at a future time.

[0144] FIG. 16 is a conceptual diagram for explaining such a "prediction model."

[0145] Although not particularly limited, a model such as a "state space model" can be used as the prediction model.

[0146] Here, the "state space model" is a method of time series analysis that mathematically represents a system in which the variables that represent the actual state differ from the variables that can be actually observed, and is applied in various fields such as weather forecasting and control theory.

[0147] For example, the following document discloses a configuration that uses a state space model to predict purchasing patterns with a small amount of calculation while taking into account time information of purchases and responding to changes in purchasing data over time. Publicly known document 1: JP 2010-61456 A

[0148] In addition, the following literature discloses a technology that can prevent an increase in the load when providing highly accurate sales forecasts for each product by using models such as a state space model. Publicly known document 2: JP 2016-115157 A

[0149] Furthermore, the following document discloses a prediction device that predicts congestion levels through simple processing using time-series data of past congestion levels and a state space model. Publicly known document 3: Patent No. 7099770

[0150] Therefore, as shown in Figure 16, a "prediction model" can be generated that inputs multiple past time-series actual data of "number of people, attributes, and behavior classification" for each estimated area mesh along with date and time information, and outputs predicted data for a specified time in the future for each estimated area mesh.

[0151] For example, the predicted data can be the "purchase volume" and "sales amount" of a certain type of product (or a specific product) in the estimated mesh area. In this case, the past performance data can include POS data within the estimated mesh area, as described above.

[0152] Alternatively, for example, the predicted data may be the number of people and the classification of people's behavior in the estimated area mesh at a future time.

[0153] If the prediction condition is, for example, a specified time in the future, the demographics estimation unit 2111 performs the same processing as steps S120 to S140 under the prediction condition as described above, thereby generating data that predicts and displays the distribution of people in the area to be analyzed, and returns this data to the user terminal (S180).

[0154] Therefore, when the lifestyle information service providing server 2000 executes a prediction process of demographic information of people located within a specified area at a specified time period, the analysis target time / area information acquisition unit 2101 acquires analysis target area information indicating the analysis target area specified by the user, and the demographic estimation unit 2111 estimates people's data (including attributes such as number of people, gender, age, etc.) for each second area mesh unit (estimated area mesh) smaller than the first area mesh unit for the time set as the prediction condition, based on the specified demographic information including information on the number of people located in multiple first area mesh units included in the analysis target area at a "specified time (hour or time period)".

[0155] Here, as in the case of Fig. 2, "a specified time" refers to, for example, a "first specified time interval", for example, every hour, and "a specified area" refers to, for example, a first area mesh unit, for example, a 500m mesh (a square of 500m on each side) unit, of a specific region. "Specified demographic information" refers to "demographic data" and means "aggregated data", such as the above-mentioned "Mobile Spatial Statistics Information (registered trademark)".

[0156] Furthermore, the predetermined "lifestyle information 2306" is the same as that described in FIG. 2 to FIG. 4, and more specifically, it is information on classification of predetermined behaviors by region, age, and sex in a "second time interval" that is shorter than the first predetermined time interval and is preferably a divisor of the first time interval. Here, the "lifestyle information" corresponds to information in a database according to the "Basic Survey on Social Life" by the Ministry of Internal Affairs and Communications, which classifies predetermined 20 types of behaviors by region, age, and sex, and in 15-minute increments. Furthermore, the "predetermined area location information" refers to "information on a predetermined place that exists in the above-mentioned predetermined area (including information on the type of job at that place)", and preferably, the "area location information" includes information on buildings and facilities in the area of ​​the second area mesh unit (estimated area mesh) that is smaller than the above-mentioned first area mesh unit and preferably includes information on the types of jobs performed in the buildings and facilities for an area of ​​the second area mesh unit (estimated area mesh) that is smaller than the above-mentioned first area mesh unit and has sides that are divisors of the first area mesh unit when the first area mesh unit is a rectangle.

[0157] FIG. 14 shows a summary of performance information for each estimated area mesh included in the analysis target area AR.

[0158] The estimated area mesh MC contains area demographic information such as number of people, gender, and age, and area behavior information such as number of people for each behavior category, building information (number of floors, floor area), and facility attributes since the industry corresponds to the floor area.

[0159] As described above, the performance information may also include information on the purchase of products and services based on POS data.

[0160] Therefore, this performance information can be used as an input for learning the prediction model (state space model) shown in FIG.

[0161] The performance information includes the corresponding date and time information, which may include not only date and time information but also designation of days / weekends / holidays, morning / afternoon, daytime / night, etc.

[0162] In addition, when the demographics estimation unit 2111 adds up the number of people corresponding to a specific behavior category predicted for each estimated area mesh present in the analysis area, it is also possible to estimate the total number of people performing this specific behavior in the analysis area at the time specified as the predicted date and time.

[0163] Conversely, if the user specifies as a prediction condition the number of people (including attribute information such as the number of people, gender, age, etc.) in the analysis target area who will perform this specific behavior at the prediction date and time, the number of people required information estimation unit 2107 estimates the number of people required in the analysis target area to satisfy the number of people performing the specific behavior specified by the user for this specific behavior (hereinafter, "number of people required information"). To satisfy this number of people required information, the number of people required for each estimation area mesh may be estimated.

[0164] FIG. 15 is a diagram showing an example in which the predictive simulation device is used for urban planning.

[0165] For example, as shown in FIG. 15(a), assume that the distribution of the number of people in each estimated area mesh in the analysis target area AR is estimated when a specific building exists in the area B01.

[0166] Here, in FIG. 15(a), the human density in each estimated area mesh is shown in a heat map, assuming that the human density is distributed between a minimum value p1 and a maximum value p2.

[0167] Figure 15(b) shows a heat map of the density of people in each estimated area mesh when the buildings in area B01 are mathematically removed (for example, the building height is set to 0 floors) as a prediction condition.

[0168] Here, it can be seen that if the building that existed in area B01 disappears, the distribution of people will be dispersed to areas B02 and B03.

[0169] Therefore, in urban planning, it is possible to simulate how the distribution of people within a given analysis area will change if a building is virtually constructed at a specified location within the analysis area, or if an existing building is removed.

[0170] Therefore, by changing area location information (for example, the construction of new facilities (including "locations" such as stations and shops) or the construction of new roads and railways) as a setting of predetermined assumption conditions (prediction conditions) based on predetermined statistical information (including demographic information), it is possible to calculate the "impact area" in which the effects of such changes will appear.

[0171] Examples of assumed conditions include redevelopment plans for the area in front of the station, construction of public housing and housing for the elderly, and changes to bus routes.

[0172] The area life information estimation unit outputs area life information based on the area place information output by the area place information estimation unit.

[0173] With the above configuration, the impact of future increases / decreases in local population, redevelopment plans, etc. can be reflected in the simulation.

[0174] And, "human behavior information" here includes "behavior in the narrow sense" such as "purchasing a product" or "using the toilet," as well as "behavior in the broad sense" that occurs as an outcome of human behavior, such as "people pooling in front of train stations," "traffic congestion," "public transportation congestion," and "shortage of housing."

[0175] Furthermore, if the assumed conditions here are not those of the type of location, but rather those of the presence or absence of events such as the Olympics or the World Expo, and actual information on past events held is used in a prediction model such as that described in Figure 16, it will be possible to predict the demographic trends for each estimated area mesh based on actual information on two (or more) different types of demographic trends for cases where an event is held and cases where it is not held.

[0176] The so-called economic effects can be analyzed based on the difference in human behavior information depending on whether or not there are events such as the Olympics or the World Expo, redevelopment of areas in front of stations, expansion of bus routes, urban planning such as the construction of public housing, etc.

[0177] By collecting performance information, it is possible to concretely grasp the characteristics of human behavior and the effects of disturbances such as specific events (including synergies resulting from combinations of different events) and urban planning as "differences" in human behavior, providing unique input information for calculating future "economic effects" as well as quantitatively assessing past "opportunity losses" (including simulations for relative points in the past).

[0178] FIG. 17 is a conceptual diagram showing the hierarchy of data and the hierarchy of data processing used in the prediction simulation device.

[0179] FIG. 18 is a conceptual diagram showing an outline of the flow of data processing.

[0180] As shown in Figure 17, the "public domain," which is a public and open domain, includes, for example, the "urban OS" as "software that supports urban infrastructure" and the data integration platform.

[0181] In Japan, for example, the following Cabinet Office documents are publicly available: Publicly known documents: https: / / www8.cao.go.jp / cstp / stmain / a-whitepaper3_200331.pdf

[0182] Here, there are three challenges to realizing smart cities in Japan: i) reuse and horizontal deployment of services, ii) utilization of data across fields, and iii) low scalability.

[0183] i) In terms of reusing and expanding services, traditionally systems were specialized for each field or organization, which made it difficult to reuse or expand services to other regions.

[0184] ii) When it comes to utilizing data across sectors, traditional services have separate data for each sector and organization, making it difficult to build new services that span sectors.

[0185] iii) Low scalability: Conventional specialized systems have the problem that expanding functions requires large costs and effort, making it difficult to continuously and easily evolve services.

[0186] In order to address the challenges in realizing these smart cities in Japan, the City OS will be designed to have the following features: i) interoperability (connection), ii) data distribution (flow), and iii) ease of expansion (sustainability).

[0187] The "demographic information prediction system area" of this embodiment is conceptually configured to exist on such a foundation.

[0188] As shown in Figure 17, in addition to "Docomo Spatial Statistics (registered trademark)" as shown in Figure 4, system 100 is also expected to utilize weather data for a specific time period in a specific area obtained from the Japan Meteorological Agency server, "disaster risk information" obtained from the National Research Institute for Earth Science and Disaster Prevention, and the like.

[0189] When such data is stored in the above-mentioned "data lake," it is first stored as general-purpose raw data, and then finalized for a certain time period and stored as a "unified format for the demographic information prediction system (for example, unifying the unit area mesh)" as mentioned above.

[0190] As shown in Figures 17 and 18, data in a general-purpose data format provided to the "Data Lake" may be posted from a smartphone via a smartphone application, for example. The posted data may be subjective evaluation data on the congestion situation at a specific facility in a specific location, or may be photographs or videos of the congestion situation.

[0191] Furthermore, as shown in Figures 17 and 18, above the "data lake" there is a data analysis layer, which corresponds to the data integration and analysis processing performed by the above-mentioned life information service providing server 2000.

[0192] The system is structured so that a section that adds some sort of evaluation to the analysis results, or collaborates with other companies on the results, and finally visualizes and presents them, exists at a higher data display level.

[0193] In particular, while spatial statistics are based on an hourly basis, the above-mentioned processing makes it possible to perform data analysis and visualization in 15-minute increments, which is almost real-time.

[0194] As shown in Figure 18, after the above-mentioned integrated analysis process is performed, facilities in the town are evaluated based on the information, and people are allocated to each facility. Therefore, since the people flow distribution is created from independent information (people flow → behavior → town information), the people flow distribution changes when the "town information (general map information)" is updated. Therefore, it is possible to visualize not only current real-time information, but also people flow distribution in towns that differ from the present in the past and future.

[0195] For example, by using city information from the Edo period, we can estimate the distribution of people at that time and visualize the life of the city at that time, and by using city information from virtual worlds such as movies, anime, and games, we can estimate the distribution of people in the virtual world. Of course, it is also possible to visualize the life of a specific city in the future (e.g. Osaka City).

[0196] These analytical results can be used for urban planning, environmental optimization, energy optimization, real estate value assessment, logistics optimization, etc. For example, when assessing real estate value, it is possible to analyze and evaluate what kind of changes in population distribution may occur if a certain facility is hypothetically constructed in a certain location in the future, and this information can be added to the real estate value assessment.

[0197] One possible use case for data integration and analysis would be to distribute admission tickets to product promotional events (whether at a physical or virtual venue) via smartphone based on users' behavioral history, etc.

[0198] FIG. 19 is a diagram showing an overview of services realized by using the demographic information prediction system 1000.

[0199] FIG. 20 is a conceptual diagram showing an example of the service shown in FIG. 19, which is provided via a smartphone.

[0200] As shown in Figure 19, maps, people flows, various sensor data, weather data, satellite data, real estate data, and data posted via smartphones stored in the data lake can be analyzed on an integrated analysis platform (artificial intelligence prediction) to analyze the condition of each region, calculate a score for each region for specified evaluation items, and map and visualize each region.

[0201] In this case, the output of the integrated analysis platform will be provided with interfaces suitable for residents, local businesses, and local governments.

[0202] For example, residents could be encouraged to make behavioral changes through a smartphone application to improve the score for their neighborhood or area of ​​residence.

[0203] For example, as shown in Figure 20, in the real world, it is possible to think of a moving person (or the smartphone they carry and move around with) as a kind of sensor that accumulates data in a data lake via a smartphone application. By using city information posted by smartphone users (which can include not only crowding but also data on weather, accidents, crime, etc.), it is possible to provide a location analysis platform as a service in the digital space, which can provide the distribution of people within walking distance, or even the distribution of people in the surrounding area, and even larger statistical data on the state of the city in the living area.

[0204] With this configuration, it is possible to collect information about the town's attractions directly from local residents, and it is also possible to encourage behavioral changes among people who can share information via smartphone (such as visiting the town).

[0205] Furthermore, by using detailed analysis reports and town simulation tools to base policy decisions and implementation effectively and efficiently on objective evidence, the government will be able to keep the cycle of town development for everyone going.

[0206] Regional businesses will be able to carry out area marketing such as store development and sales promotion based on the scores obtained from the integrated analysis platform.

[0207] FIG. 21 is a diagram showing another example of data visualization by integrated analysis.

[0208] We have explained above how the current state of a city can be displayed in real time by using an integrated analysis platform.

[0209] In this case, the types of data that can be superimposed on the map information include "distribution of the number of people (including people flow. People flow can also be shown, for example, using vectors)" and "elevation," as well as the type of use of each location within the area, such as "commercial use," "residential use," or "office use," the spatial utilization rate, and even the distribution of road maintenance conditions.

[0210] As described above, by utilizing the integrated analysis by the life information service providing server 2000, it is possible to predict the distribution of people (including attributes such as age and gender) within the area to be analyzed under specified prediction conditions.

[0211] Furthermore, by utilizing the integrated analysis by the life information service providing server 2000, the analyzed results of the predictive simulation can be provided to the user in a form that is easily understandable to the user through visualized information such as charts.

[0212] Furthermore, by utilizing the integrated analysis by the life information service providing server 2000, the user can easily understand the results of the analyzed predictive simulation by relating them to locations on a map.

[0213] Furthermore, by utilizing the integrated analysis by the life information service providing server 2000, it is possible to perform a time series analysis for a predetermined period of time.

[0214] Furthermore, by utilizing the integrated analysis by the life information service providing server 2000, it is possible to simulate life information in the future.

[0215] Furthermore, by utilizing the integrated analysis by the life information service providing server 2000, when a predetermined area location information cannot be obtained as statistical information or the like, it can be supplemented by estimation.

[0216] Furthermore, by utilizing the integrated analysis by the life information service providing server 2000, when certain demographic information is not available as statistical information or the like, it is possible to make up for it by estimation.

[0217] Although one embodiment of the present invention has been described above, the present invention is not limited to the above-described embodiment, and modifications, improvements, etc. within the scope that can achieve the object of the present invention are included in the present invention.

[0218] Furthermore, for example, the series of processes described above can be executed by hardware or software.

[0219] In other words, the functional configurations in FIGS. 1 to 21 are merely examples and are not particularly limited.

[0220] That is, it is sufficient that the information processing system is provided with a function capable of executing the above-mentioned series of processes as a whole, and the type of functional block used to realize this function is not limited to the examples in Figures 1 to 21. Furthermore, the locations of the functional blocks and databases are not particularly limited to those in Figures 1 to 21 and may be arbitrary. For example, at least a portion of the functional blocks and databases required to execute various processes may be transferred to a user terminal or the like. Conversely, the functional blocks and databases of the user terminal may be transferred to a server or the like.

[0221] Furthermore, one functional block may be configured as a single piece of hardware, a single piece of software, or a combination of both.

[0222] When the series of processes is executed by software, the program constituting the software is installed into a computer or the like from a network or a recording medium.

[0223] The computer may be a computer implemented with dedicated hardware.

[0224] Furthermore, the computer may be a computer capable of executing various functions by installing various programs thereon, such as a server, a general-purpose smartphone, or a personal computer.

[0225] A recording medium containing such a program may be composed not only of a removable medium (not shown) that is distributed separately from the device main body in order to provide the program to a user, etc., but also of a recording medium that is provided to a user, etc. in a state where it is pre-installed in the device main body.

[0226] In this specification, the steps of describing a program to be recorded on a recording medium include not only processes that are performed chronologically according to the order, but also processes that are not necessarily performed chronologically but are executed in parallel or individually.

[0227] In addition, in this specification, the term "system" refers to an overall device that is composed of a plurality of devices, a plurality of means, etc.

[0228] The embodiments disclosed herein are merely examples of configurations for specifically implementing the present invention, and do not limit the technical scope of the present invention. The technical scope of the present invention is indicated by the claims, not by the description of the embodiments, and is intended to include modifications within the literal scope of the claims and within the scope of equivalent meanings. [Explanation of symbols]

[0229] 2 network, 100 user, 200 smartphone, 1000 lifestyle information estimation system, 2000 lifestyle information service server, 2100 computing device, 2101 analysis target time and area information acquisition unit, 2102 people flow path information estimation unit, 2103 human behavior information estimation unit, 2104 map output unit, 2105 human behavior information designation reception unit, 2106 area setting unit, 2107 required number of people off estimation unit, 2108 area location information estimation unit, 2109 area lifestyle information estimation unit, 2111 demographic information estimation unit, 2115 location information provisional setting unit, 2110 lifestyle information estimation unit, 2300 storage device, 2301 analysis target time and area information, 2302 demographic information, 2303 human behavior actual information, 2304 traffic route information, 2305 human behavior information, 2306 lifestyle information, 2308 Employment information by occupation, 2310 Area location information, 2311 Area life information, 2312 People flow information, 2313 Statistical information, 3000 Life information user terminal, 5000.1~5000.M Data provision server.

Claims

1. A prediction simulation device for an analysis target area, A storage device that stores predetermined demographic information including information on the number of people present in each of a plurality of predetermined area meshes included in the analysis target area in a time series of a series of first time units; and a computing device for analyzing the demographic information, the computing device comprising: an analysis target condition acquisition means for receiving a prediction condition from a user; a human behavior information estimation means for estimating, under the prediction conditions, the number of people for each classification of human behavior present in the predetermined area mesh included in the analysis target area for each second time unit shorter than the first time unit based on the predetermined demographic information; a demographics estimation means for estimating aggregate data for each estimated area mesh within the analysis target area by distributing the number of people estimated by the human behavior information estimation means to the estimated area mesh based on the estimation result of the human behavior information estimation means and the type of place corresponding to the classification of the human behavior for each estimated area mesh obtained by dividing the specified area mesh and the size of the place, under the prediction conditions; A predictive simulation device comprising: an estimation means for estimating a distribution of human attributes within the analysis target area based on a classification of human behavior estimated in the estimation area mesh within the analysis target area and the aggregated data.

2. the prediction conditions include a first condition in which the user virtually designates at least one of the type of the location and the size of the location for each of the estimated area meshes, and a second condition in which the user does not designate the first condition; The demographics estimation means further estimates aggregated data for each of the estimated area meshes in the analysis target area by distributing the demographic information to the estimated area meshes based on the estimation result of the human behavior information estimation means and the human behavior for each of the estimated area meshes corresponding to at least one of the virtually specified type of place and the size of the place, 2. The predictive simulation device according to claim 1, wherein the estimation means outputs results of estimating the number of people for each estimation mesh within the analysis target area under the first condition and the second condition.

3. The demographic information is prediction information of demographics at a future time point specified by the prediction condition based on predetermined statistical information, 3. The prediction simulation device according to claim 2, wherein the type of place and the size of the place are information planned for the place at the future time point specified by the prediction conditions.

4. the storage device stores human behavior performance information indicating past human behavior performance in a predetermined time unit identical to the series of time series; 2. The predictive simulation device according to claim 1, wherein the human behavior information estimation means estimates prediction data related to the human behavior in the estimation area mesh unit in the specified time unit using a model generated from the human behavior record information and the results of the demographics estimation means.

5. the storage device stores the demographic information, the lifestyle information, and the area location information as data acquired from an external server via a network; The demographic information is aggregated data of demographic data for a first time interval corresponding to the first time unit, and is information on the number of people by age and sex present in each predetermined area mesh in a predetermined area including the analysis target area during the first time interval; the life information is information for each predetermined time unit, the predetermined time unit being a second time interval shorter than the first time interval and corresponding to the second time unit, the life information being information capable of associating the age, sex, and predetermined classification of the person's life and behavior with the occupation of the person, The area location information is location information for each predetermined section, and includes information on the total floor area of ​​a facility within the predetermined area and the type of job within the facility, the human behavior information estimation means includes an area life information estimation means for estimating the number of employed people by occupation type in the specified area for each of the second time intervals based on the life information for each of the specified blocks and the demographic information and the life information; The predictive simulation device according to any one of claims 1 to 3, wherein the demographics estimation means allocates the number of employees by occupation type to each of the facilities based on the total floor area and the area allocated to the occupation type of each of the facilities, and estimates the aggregated data of the number of employees in the estimation area mesh of the demographics information.

6. The model is a model for estimating a number of people associated with a behavior of a specified person as the prediction data, a human behavior information designation receiving means for receiving a designation of the number of people regarding the human behavior to be analyzed in the predetermined time unit and the estimation area mesh unit; The predictive simulation device according to claim 1, further comprising: a workforce requirement information output means for outputting workforce requirement information indicating the number of people in the specified time unit and the specified area unit required for the specified number of people to execute the action of the people in the specified time unit and the estimated area mesh unit.

7. 7. The prediction simulation device according to claim 6, wherein the specified human behavior is purchasing behavior, and the model is a model for estimating the number of people who will perform the specified purchasing behavior and the average customer price as the prediction data.

8. The demographic prediction information at a future time point includes first prediction information in a case where a predetermined event occurs, and second prediction information in a case where the predetermined event does not occur, The human behavior information estimation means estimates the number of people for each classification of the human behavior in response to each of the first prediction information and the second prediction information, The demographics estimation means further estimates aggregated data for each of the estimated area meshes in the analysis target area by distributing the number of people estimated by the human behavior information estimation means to the estimated area meshes; 4. The predictive simulation device according to claim 3, wherein the estimation means outputs results of estimating the number of people for each estimation mesh within the analysis target area based on the first prediction information and based on the second prediction information.

Images