Area analysis device, human flow analysis device, and human flow analysis system

Abstract

Based on demographic information of the area being analyzed, the system estimates the distribution of people with user-specified attributes at a given location within that area. [Solution] The human flow information estimation device 2000' includes a calculation device 2100 that performs human flow information analysis based on predetermined demographic information, which includes the number of people located in each of a plurality of predetermined area meshes and information on the attributes of those people. The human flow information estimation unit 2103 identifies a set of unit sub-regions that cover the area to be analyzed without overlapping each other, and based on demographic information and predetermined lifestyle information, it distributes the number of people from the demographic information to the unit sub-regions at predetermined time intervals, based on the type of place and the size of the place corresponding to the classification of people's behavior in each area mesh, and estimates aggregated data for each unit sub-region. Based on the aggregated data, the human flow information estimation unit 2103 calculates a human absorption capacity value, which is the ratio of the number of people with the analysis target attributes for each unit sub-region to a predetermined value.

Description

[Technical Field]

[0001] The present invention relates to a region analysis device, a human flow analysis device, and a human flow analysis system that estimate the distribution of objects with predetermined attributes in an analysis target area specified by the user. [Background technology]

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

[0003] Figures 32 to 34 show an overview of services that provide such demographic information.

[0004] Figures 32 to 34 illustrate, as an example, "Mobile Spatial Statistics (registered trademark)" provided by NTT DOCOMO, INC.

[0005] For more information on this type of "mobile spatial statistics (registered trademark)," see, for example, Non-Patent Document 2.

[0006] "Mobile Spatial Statistics (registered trademark)" refers to population statistics created using the mechanisms of mobile phone networks. This spatial statistics is characterized by the fact that it represents only the number of people in a group, and therefore cannot identify individual mobile phone or smartphone users. In this respect, it differs from so-called "personal information" and can be provided to third parties with population statistics for a specific area at a specific time without obtaining the consent of the information recipient.

[0007] This spatial statistics service allows us to understand the population distribution across Japan every hour, 24 hours a day, 365 days a year. Furthermore, it enables us to understand the population structure by gender, age group, and residential area, allowing for analysis tailored to specific purposes. Since its launch in October 2013, it has been widely used by the Japan Tourism Agency, local governments, private companies, and others.

[0008] More specifically, mobile phone networks periodically track the number of mobile phones located within each base station's area to ensure users can make calls, send emails, and access other services anytime, anywhere. By using this system to compile data on the number of mobile phones and taking into account the penetration rate of mobile phone carriers, it is possible to estimate the population. This is known as Mobile Spatial Statistics (registered trademark).

[0009] Demonstration experiments were conducted in public sectors such as urban development and disaster prevention planning, confirming that this spatial statistics can be widely used in society. Furthermore, as mentioned above, it was made available in October 2013 so that it could be used not only in the public sector but also in academic and industrial fields.

[0010] This spatial statistics allows for analysis by gender, age group, residential area (domestic residents), country / region (foreign visitors to Japan), and time of day, depending on the purpose.

[0011] As shown in Figure 32(a), the entire NTT Docomo service area is the statistical target area, making it possible to grasp the nationwide population distribution every hour. Therefore, it is possible to grasp the population distribution every hour, 24 hours a day, 365 days a year.

[0012] Furthermore, as shown in Figure 32(b), this demographic data allows us to understand the population structure by gender and age group. In this case, the target group is those aged 15 to 79.

[0013] Furthermore, as shown in Figure 33, this spatial statistics makes it possible to understand the population by residential area for specific time periods (every hour) within a specific region. As mentioned above, this makes it possible to understand the population distribution by residential area based on mobile phone registration information.

[0014] Furthermore, assuming that spatial statistical information is available, technologies related to understanding and predicting "human flow" using demographic information as input data have also been proposed (see, for example, Non-Patent Document 3).

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

[0016] On the other hand, "aggregated data" is location-based data collected by traffic sensors and wireless base stations installed on roads. Because data is acquired by equipment installed in specific locations, once the equipment is installed, data can be collected fairly comprehensively. Also, since the data is not tied to individual subjects, there are fewer privacy concerns, and the data measured once has a high potential to be used for various purposes. However, since information that spans across locations cannot be obtained, there are limitations to its use when it comes to understanding urban pedestrian flow.

[0017] One type of aggregated data is "demographic information," which records the population of a city by location and by time. Much of this demographic information records population information by area for each time period. When dealing with information covering a wide area, mobile phone data is often used to ensure comprehensiveness of the information. Based on terminal connection information at mobile phone base stations and information such as CDR (Call Data Record), the number of terminals in each area is counted, and the actual population is estimated while considering usage rates for each attribute, and this is used as demographic information.

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

[0019] Figure 34 shows the data aggregation procedure in "Mobile Spatial Statistics (registered trademark)" disclosed in Non-Patent Document 2.

[0020] Referring to Figure 34, this "spatial statistics" employs the following processing to ensure safety by excluding personally identifiable information.

[0021] As described above, in order to strictly protect customer privacy, since this spatial statistic is demographic information representing only the number of people in a group, it is impossible to identify individual users.

[0022] This spatial statistic is created through the following procedure. i) Anonymization process: A process of removing identifying information such as names, phone numbers, dates of birth, etc. from the operation data ii) Aggregation process: A process of deriving statistical "information about the group" by performing statistical inferences from the anonymized information iii) Confidentiality process: A process of ensuring that the aggregated results do not contain numerical values for small population areas

[0023] And in Non-Patent Document 3, in order to more effectively utilize demographic information, as a pedestrian flow analysis, a technology is disclosed that performs a task where the input is demographic information corresponding to multiple times, and based on that, outputs the movement amount between cells over time.

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

[0025] Also, FIG. 35 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.

[0026] As shown in FIG. 35, in the "Basic Survey on Social Life", for each region, living information (behavior classification) in units of 15 minutes is surveyed by age and gender.

Prior Art Documents

Patent Documents

[0027]

Patent Document 1

Non-Patent Documents

[0028] [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: "Understanding and Predicting Human 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 [Overview of the project] [Problems that the invention aims to solve]

[0029] However, the technologies described in Non-Patent Documents 1 and 2 only provide statistics such as the age (age group) and gender of people present in a given area every hour.

[0030] Furthermore, while the technology described in Non-Patent Document 3 estimates "human movement" (the number of people moving) from "demographic data," it cannot determine what kind of people are doing what kind of activities in a particular area at a particular time.

[0031] The information in Non-Patent Document 4 merely shows what activities people in each occupation engage in during a given time period.

[0032] Furthermore, the technology described in Patent Document 1 only discloses a technology for estimating the number of people in a building based on demographic data.

[0033] Therefore, there was a challenge in analyzing information that classifies the behavior of people located in a given area (hereinafter referred to as "lifestyle information") at intervals shorter than the time units for which demographic data is accumulated.

[0034] Furthermore, even if we estimate the distribution of people by attribute within a certain area using the demographic data described above, the following problems arise.

[0035] In other words, as mentioned above, obtaining demographic data involves "anonymization," which inevitably results in some information being omitted from the acquisition. Furthermore, when acquiring information for each attribute, the "estimation method" and "aggregation accuracy" may differ, resulting in a problem where the estimation of the distribution of people for each attribute within a given region is inconsistent in terms of accuracy.

[0036] Furthermore, due to the problems mentioned above, it becomes difficult to accurately estimate the distribution of people within a given region by combining multiple attributes.

[0037] In other words, even if it is possible to estimate the number of people with a specific attribute in a particular area within a region with a certain degree of accuracy, there is a problem in that it is difficult to accurately estimate the number of people corresponding to a combination of multiple attributes (for example, age group, gender, purpose of visit to that area, etc.).

[0038] The present invention was made to solve the above-mentioned problems, and its objective is to provide a region analysis device that can accurately associate distribution information of estimated targets, including moving objects, with the attributes of the targets, within a predetermined time period in a predetermined region, based on dynamic information of the targets in that region.

[0039] Another object of the present invention is to provide a human flow analysis device and a human flow analysis system that can accurately correlate human flow information in a predetermined area during a predetermined time period with human attributes, based on demographic information of that area.

[0040] Furthermore, the present invention provides a human flow analysis device and a human flow analysis system that can estimate human flow information for a given time period in a given area based on demographic information of that area, for combinations of multiple human attributes. [Means for solving the problem]

[0041] To achieve the above objective, according to one aspect of the present invention, a region analysis device comprising a storage device that stores predetermined statistical information including the number of objects located in each of a plurality of predetermined region meshes and information on the attributes of the objects, and a computing device that performs an analysis of the distribution of objects within the region to be analyzed,

[0042] The object includes at least a first object that resides within a predetermined region mesh and is movable between predetermined region meshes, the predetermined statistical information includes dynamic statistical information in a time series of first time units, including the number of first objects located in each predetermined region mesh and information on the first attributes of the first objects, the computing device includes analysis condition setting means that accepts the designation of a predetermined analysis target area and analysis target attributes by user operation and identifies a set of unit sub-regions that cover the analysis target area without overlapping each other, and object distribution estimation means that estimates aggregated data for each unit sub-region within the analysis target area by distributing the number of first objects to the unit sub-regions within the region mesh based on the dynamic statistical information, the object distribution estimation means estimates the distribution of the analysis target attributes of the first objects within the analysis target area based on the aggregated data and calculates a first object absorption value which is the ratio of the analysis target attributes of the first objects for each unit sub-region to a first predetermined value.

[0043] Preferably, the first predetermined value is an average value obtained by dividing the first target of the analyte attribute within the analyte area by the number of unit sub-regions present within the analyte area.

[0044] Preferably, the target further includes a second target which is a facility located within a predetermined regional mesh, the predetermined statistical information includes facility distribution information which includes the number of second targets located in each predetermined regional mesh and information on the second attributes of the second targets, the analysis target attributes include at least one first attribute and at least one second attribute, the target distribution estimation means i) estimates the number of second targets in each unit sub-region within the analysis target region by distributing the number of second targets to unit sub-regions within the regional mesh based on the facility distribution information, ii) estimates a second target absorptive value which is the ratio of the number of second targets of the analysis target attribute in each unit sub-region to a second predetermined value for each second attribute, and iv) calculates a target absorptive vector value which has the first target absorptive value and the second target absorptive value as its components.

[0045] Preferably, the second predetermined value is an average value obtained by dividing the second target of the analyte attribute within the analyte region by the number of unit subregions present within the analyte region.

[0046] Preferably, the first object is a person.

[0047] Preferably, the system further includes a map output means that superimposes and displays the first target absorption value onto a map of the area to be analyzed.

[0048] Preferably, the system further includes a map output means that superimposes the target absorption force vector values ​​onto a map of the area to be analyzed.

[0049] According to another aspect of the present invention, the present invention comprises a storage device that stores predetermined demographic information, including the number of people located in each of a plurality of predetermined area meshes and information on the attributes of those people, in a time series of a first time unit, and a computing device that performs analysis of human flow information within the area to be analyzed, wherein the computing device accepts the designation of a predetermined area to be analyzed and the attributes of the area to be analyzed by user operation, and includes analysis condition setting means that identifies a set of unit sub-regions that cover the area to be analyzed without overlapping each other, and based on the demographic information and predetermined lifestyle information, calculates the behavior of people located in the area meshes included in the area to be analyzed at predetermined time intervals. The system includes a human behavior information estimation means for estimating the number of people with each attribute of the target of analysis, and a human flow information estimation means for estimating aggregated data for each unit sub-region within the target of analysis by distributing the number of people estimated by the human behavior information estimation means to unit sub-regions within the area mesh, based on the estimation results of the human behavior information estimation means and the type and size of places corresponding to the classification of human behavior for each area mesh. The human flow information estimation means estimates the distribution of people with the target of analysis within the target of analysis based on the aggregated data and calculates a human absorption value, which is the ratio of the number of people with the target of analysis in each unit sub-region to a predetermined value.

[0050] Preferably, the predetermined value is an average value obtained by dividing the number of people with a predetermined attribute within the area to be analyzed by the number of unit sub-regions present within the area to be analyzed.

[0051] Preferably, the human behavior information estimation means estimates the number of people for each category of human behavior located in the area mesh included in the area to be analyzed, based on demographic information and predetermined lifestyle information, at a second time unit shorter than the first time unit, and the human flow information estimation means estimates aggregated data for each estimated area mesh obtained by dividing the area mesh.

[0052] Preferably, the attribute to be analyzed includes multiple human attributes, and the human flow information estimation means estimates a human absorption value for each of the multiple human attributes, which is the ratio of the number of people of the attribute to be analyzed in each unit sub-domain to a predetermined value, and calculates a human absorption vector value whose components are the human absorption values ​​for each of the multiple human attributes.

[0053] Preferably, the system further includes a map output means that superimposes and displays the human absorption value on a map of the area to be analyzed.

[0054] Preferably, the system further includes a map output means that superimposes and displays the human absorption force vector values ​​onto a map of the area to be analyzed.

[0055] Preferably, the analysis condition setting means accepts the user's operation to specify the time period for analysis for the area to be analyzed and the location for analysis within the area to be analyzed, and the human flow information estimation means estimates the time-dependent changes in the human absorption capacity value and human absorption capacity vector at the location to be analyzed during the time period.

[0056] Preferably, the analysis condition setting means accepts the user's operation to specify the analysis time period and reference time period for the area to be analyzed, and the specification of the location to be analyzed within the area to be analyzed, and the human flow information estimation means estimates the time-dependent changes in the human absorption capacity value and human absorption capacity vector at the location to be analyzed during the analysis time period and reference time period.

[0057] Preferably, the analysis time period is the time period during which a predetermined event is held at the location to be analyzed, and the reference time period is the time period during which the predetermined event is not held at the location to be analyzed. The human flow information estimation means calculates information indicating the human attraction force due to the predetermined event at the location to be analyzed by comparing the changes in human absorption value and human absorption vector over time during the analysis time period and the reference time period.

[0058] Preferably, the human flow information estimation means calculates information indicating the duration of the effect of human attraction due to a predetermined event at the analysis location by comparing the temporal changes of human absorption value and human absorption vector during the analysis period and the reference period.

[0059] Preferably, the storage device stores demographic information, lifestyle information, area location information, and stay type information, wherein the demographic information is aggregated demographic data for each first time interval, and is information on the number of people by attribute located in each area mesh in a predetermined area during the first time interval, and the attributes of people include stay type information, which is the type of stay in which a person stays in the predetermined area.

[0060] Preferably, the type of stay information includes attributes of residents in a given area, attributes of workers in a given area, and attributes of non-residents and non-workers.

[0061] Preferably, when the human flow information estimation means allocates the number of people based on demographic information to each location within a predetermined area according to the attributes of the area location information corresponding to the type of stay information, if the type of stay is that of a resident, priority is given to the attribute of residential; if the type of stay is that of an employee, priority is given to business facilities; and if the type of stay is neither that of a resident nor an employee, priority is given to short-term stay locations and entertainment facilities.

[0062] According to another aspect of this invention, the invention comprises a first data provision server that provides predetermined demographic information including the number of people and information on the attributes of people located in each of a plurality of predetermined area meshes in a series of first time units, a second data provision server that provides predetermined lifestyle information including information on people's lives and behaviors in a series of first time units, a third data provision server that provides area location information including information on the type and size of places for each predetermined section located in a predetermined area, and a human flow analysis device, the human flow analysis device comprising a storage device that receives data from the first data provision server and stores predetermined demographic information including the number of people and information on the attributes of people located in each of a plurality of predetermined area meshes in a series of first time units, and a computing device that performs analysis of human flow information within the area to be analyzed, the computing device accepts the designation of a predetermined area to be analyzed and the attributes to be analyzed by user operation, and analyzes The system includes: an analysis condition setting means for dividing the target area into predetermined unit sub-areas; a human behavior information estimation means for estimating the number of people with analysis target attributes for each classification of human behavior located in the area mesh included in the target area at predetermined time intervals, based on demographic information and lifestyle information from a second data provision server; and a human flow information estimation means for estimating aggregate data for each unit sub-area within the target area by distributing the number of people estimated by the human behavior information estimation means to the unit sub-areas within the area mesh based on the estimation results of the human behavior information estimation means and area location information received from a third data provision server, based on the type and size of the location corresponding to the classification of human behavior for each area mesh. The human flow information estimation means estimates the distribution of people with analysis target attributes within the target area based on the aggregate data and calculates a human absorption capacity value, which is the ratio of the number of people with analysis target attributes in each unit sub-area to a predetermined value. [Effects of the Invention]

[0063] According to the present invention, it is possible to estimate the distribution information of an estimated target, including a moving object, within a predetermined time period in a predetermined region, based on dynamic information of that region, in an accurate association with the attributes of the target.

[0064] Furthermore, according to the present invention, it is possible to estimate the flow of people located in a predetermined area at a predetermined time, so that the estimated flow of people can be utilized in various ways by businesses, researchers, government officials, etc.

[0065] Furthermore, according to the present invention, it is possible to estimate human flow information for a predetermined time period in a predetermined area, based on demographic information of that area, by accurately associating it with the attributes of the people.

[0066] Furthermore, the present invention makes it possible to estimate human flow information for a predetermined time period in a predetermined area based on demographic information of that area, for combinations of multiple human attributes. [Brief explanation of the drawing]

[0067] [Figure 1] This figure shows an overview of the human flow analysis system according to Embodiment 1 of the present invention. [Figure 2] This is a functional block diagram showing the system overview of the Information Service Provisioning Server 2000. [Figure 3] This is a diagram illustrating the hardware configuration of the information service server 2000. [Figure 4] This diagram illustrates the content of the data provided by data provision servers 5000.1 to 5000.M. [Figure 5] This is the first diagram illustrating the processing flow of the information service provider server 2000. [Figure 6] This is the second diagram illustrating the processing flow of the information service provider server 2000. [Figure 7] This figure shows an overview of the data output from the arithmetic unit 2100. [Figure 8] This is a conceptual diagram showing the data generated when estimating the number of people in a facility based on the number of people per activity at the first time point. [Figure 9] This is a conceptual diagram showing the data generated when the number of people in a facility is estimated from the number of people in each activity at the second time point. [Figure 10]This figure shows the concept of improving the accuracy of spatial decomposition in the action-based number estimation process (S120, S210). [Figure 11] This figure shows another example of a superimposed image of a map image and population distribution output by the map output unit 2104. [Figure 12] This is a functional block diagram illustrating the configuration of the pedestrian flow information service provider server 2000' in Embodiment 2. [Figure 13] This is a conceptual diagram showing how a predetermined area RM is specified on a digital map. [Figure 14] This is a conceptual diagram to explain the data structure of digital maps. [Figure 15] This is a conceptual diagram illustrating an example of building information. [Figure 16] This flowchart shows the process of estimating the distribution of the number of people in each sub-region within the area being analyzed. [Figure 17] This is a conceptual diagram using a map of Osaka City as an example, where the analysis area is specified as a circle (radius 10km) with Osaka Station as the center. [Figure 18] This figure shows the human absorption capacity value as a heatmap, with a unit sub-region defined to cover the area to be analyzed, and age specified as the attribute to be analyzed. [Figure 19] This diagram shows the human absorption capacity value as a heatmap, with a unit sub-region defined to cover the area to be analyzed, and the visitor's place of residence (origin area) specified as the attribute to be analyzed. [Figure 20] This diagram shows the human absorption capacity value as a heatmap, with unit sub-regions defined to cover the area to be analyzed, and the type of stay specified as the attribute to be analyzed. [Figure 21] This diagram illustrates the concept of setting up unit sub-regions to cover the area to be analyzed and synthesizing them as human absorption force vectors. [Figure 22] This figure shows a heatmap of human absorption capacity values ​​for every 500m mesh, for men in their 30s living outside the Kansai region within the analysis area who are "not working away from home." [Figure 23]This figure shows an example of analyzing the temporal changes in pedestrian flow by specifying the location within the analysis area, the attributes of the people, and the time period to be analyzed. [Figure 24] This figure shows an example of analyzing the acceleration of changes in pedestrian flow over time, by specifying the location within the analysis area, the attributes of the people, and the time period to be analyzed. [Figure 25] This figure shows an example of analyzing the acceleration of changes in pedestrian flow over time, by specifying the location within the analysis area, the attributes of the people, and the time period to be analyzed. [Figure 26] This figure shows an example of setting the analysis area to include the region where a specific event is held. [Figure 27] This figure shows the change in human attractiveness over time when all human attributes are specified for a designated area, and the time period of the day including the event duration is specified as the analysis time period. [Figure 28] This is a conceptual diagram showing the data hierarchy and data processing hierarchy used in a domain analysis system. [Figure 29] This is a functional block diagram illustrating the configuration of the region analysis server 2000'' of Embodiment 3. [Figure 30] This is a conceptual diagram showing the first example of the target absorption capacity value. [Figure 31] This is a conceptual diagram showing a second example of the target absorption capacity value. [Figure 32] This figure shows an overview of conventional mobile spatial statistics (1). [Figure 33] This figure shows an overview of conventional mobile spatial statistics (2). [Figure 34] This figure shows an overview of conventional mobile spatial statistics (3). [Figure 35] This diagram shows the contents of the Basic Survey on Social Life, which is a background technology. [Modes for carrying out the invention]

[0068] Embodiments of the present invention will be described below with reference to the drawings.

[0069] The following describes the configuration of a human flow analysis device for estimating the distribution of people (distribution of human attractiveness) within an analysis area from lifestyle information and location information, as well as an information processing system for estimating the distribution of people within an analysis area (hereinafter referred to as the "human flow analysis system").

[0070] As described below, the human flow analysis device in each of the following embodiments estimates lifestyle information for a predetermined time period in a predetermined area based on demographic information of that area.

[0071] Furthermore, the pedestrian flow analysis device in each of the following embodiments is a device that estimates and analyzes pedestrian flow information, such as information on the distribution of people, during a predetermined time period related to an analysis target area specified by the user, based on demographic information of a predetermined area.

[0072] Figure 1 shows an overview of the human flow analysis system according to Embodiment 1 of the present invention.

[0073] Referring to Figure 1, the pedestrian flow analysis system 1000 of Embodiment 1 of the present invention includes an information service provision server 2000 operating as a pedestrian flow analysis device, a user terminal 3000 for users receiving services from the information service provision server 2000, a data provision server 5000.1~5000.M (M: natural number) that provides the information service provision server 2000 with data necessary for estimating pedestrian flow information as described later, a building information registration terminal 4000 for registering building information as described later, and a mobile phone or smartphone 200 used by a user 100 to receive services as described later.

[0074] Here, the building information registration terminal 4000 is a terminal used by building owners or managers to register building information (such as floor plans and 2D or 3D polygon data). Visitors to the building may also register information such as photographs of the interior. In this case, the building information registration terminal 4000 may be a mobile terminal, such as a smartphone with a photo-taking function. For example, in the case of a smartphone, the information service server 2000 may receive notifications from users using a terminal with application software installed that has an input format that allows them to submit photographs along with information such as the building name, floor number, and area information (such as store name and type of store).

[0075] Figure 2 is a functional block diagram showing the system overview of the information service provision server 2000.

[0076] Figure 4 illustrates the content of the data provided by data provision servers 5000.1 to 5000.M. Hereafter, when referring to data provision servers 5000.1 to 5000.M collectively, they will simply be called "data provision server 5000".

[0077] Referring to Figure 2, the information service provider server 2000 comprises a computing unit 2100 and a storage device 2300.

[0078] The storage device 2300 first stores predetermined "population dynamics information 2302". Here, "population dynamics information 2302" includes information on the number of people located in a predetermined area at least at a predetermined time.

[0079] Here, "predetermined time" refers to, for example, a first predetermined time interval, such as every hour; and "predetermined area" refers to a specific region, such as a first area mesh unit, such as a 500m mesh (a 500m square). "Predetermined demographic information" refers to "demographic data," meaning "aggregated data," such as the aforementioned "Mobile Spatial Statistics Information (registered trademark)" as shown in Figure 4.

[0080] Furthermore, the storage device 2300 stores predetermined "lifestyle information 2306". "Lifestyle information 2306" includes at least information relating to a person's life. Here, predetermined "lifestyle information 2306" more specifically refers to information classifying predetermined behaviors by region, age, and gender in a second time interval that is shorter than a first predetermined time interval, and preferably a divisor of the first time interval. Here, "lifestyle information" corresponds to, for example, the information from the Ministry of Internal Affairs and Communications' "Basic Survey on Social Life" database, which classifies predetermined behaviors into 20 types by region, age, gender, and in 15-minute increments, as shown in Figure 4. Here, "classification of behaviors" corresponds to classifications such as "sleep," "commuting," "going to school," "work," and "studying."

[0081] Furthermore, the storage device 2300 stores "occupational employment information 2308". Here, "occupational employment information 2308" is data showing the employment ratio for each occupation on an age and gender basis, and corresponds to information from the Ministry of Internal Affairs and Communications' Labor Force Survey Database, as shown in Figure 4.

[0082] Furthermore, "specified lifestyle information" may include such "occupation-specific employment information."

[0083] Therefore, using "population dynamics information 2302" and "predetermined lifestyle information 2306," the lifestyle information estimation unit 2110 can estimate the number of people engaged in different activities (hereinafter referred to as "number of people by activity") in the "predetermined area" set by the area setting unit 2106 at the "second time interval" based on instructions from the user. Furthermore, by applying "occupational employment information" to the "number of people by activity" estimated in this way, for people whose activities fall under categories such as work or study, the lifestyle information estimation unit 2110 can calculate data (hereinafter referred to as "person activity information") on the number of people corresponding to the classification of "occupation" of people located in the "predetermined area" at the "second time interval." Note that "occupation" does not only refer to the so-called "type of occupation," but also to "work" such as "studying," which is not business but in which a person is explicitly engaged in the same activity for a certain period of time or longer at that place (facility / building).

[0084] Furthermore, while the lifestyle information 2306 is defined as a "classification of a predetermined behavior" in a second time interval that is shorter than the first time interval and preferably a divisor of the first time interval, the second time interval may, for example, be the same as the first time interval. For example, the system may estimate a "classification of a predetermined behavior" coarsely in the first time interval and present lifestyle information, pedestrian flow, and human distribution to the user, and then, in response to the user's instructions, estimate a "classification of a predetermined behavior" in a more detailed "second time interval" and present lifestyle information, pedestrian flow, and human distribution to the user.

[0085] The storage device 2300 further stores predetermined "area location information 2310". Here, "predetermined area location information" refers to "information about predetermined locations in the predetermined area described above (including information about the occupations of those locations)". For example, preferably, the "area location information" is smaller than the first area mesh unit described above, and preferably, for a second area mesh unit whose sides are divisors of the first area mesh unit when the first area mesh unit is made into a rectangle, it includes information about buildings and facilities within this second area mesh unit and information about the occupations performed within those buildings and facilities. For example, as shown in Figure 4, the "area location information 2310" corresponds to map data in 62.5m mesh units that includes information about buildings and facilities in a town, information about occupations performed within those buildings and facilities, and information about roads and water areas. Some of this "map data" is provided by private companies.

[0086] Based on the area location information 2310, the area location information estimating unit 2108 estimates the area corresponding to each occupation in the total floor area of ​​facilities and buildings located in "predetermined locations" within the "predetermined areas" set by the area setting unit 2106. It is preferable to perform such estimations in units of the second area mesh (estimated area mesh). However, depending on the user's settings, it is also possible to perform such estimations in units of the first area mesh, where each side is a multiple of the second area mesh.

[0087] Therefore, the lifestyle information estimation unit 2110 of the computing unit 2100 of the information service provision server 2000 calculates and outputs information about the work of people located in a predetermined location within a predetermined area (hereinafter referred to as "people's domain lifestyle information") at a second time interval by expanding (for example, proportionally apportioning) the "number of people for each occupation" into the "area corresponding to each occupation" of the "predetermined location".

[0088] For example, in calculating this kind of "human domain lifestyle information," the classification of "tasks" that people are engaged in at a certain location (facility / building) within the target domain is calculated according to the method described above.

[0089] As a relationship between predetermined lifestyle information and predetermined location information, for example, we can assume that during a given time period, doctors and nurses are working in hospital facilities, farmers are working on farmland, fishermen are working in fishing grounds, and housewives are raising children or pensioners are living in residences. The specific processing of this will be described later.

[0090] Furthermore, the map output unit 2104 generates and outputs a screen that displays the "human living information" estimated in this way, corresponding to "predetermined locations" within "predetermined areas" (for example, superimposed on top of the map information). By generating such an output, users can visually grasp the human living information in the "predetermined area" they have specified.

[0091] As explained above, since the "number of people by activity" or "number of people corresponding to people's area lifestyle information" is calculated for each area of ​​the second area mesh unit and at the second time interval, the pedestrian flow path information estimation unit 2102 can be configured to estimate and output pedestrian flow path information (what paths people moved along) by tracking the change in the distribution of people in the area of ​​the second area mesh unit and its surrounding areas along the time axis and integrating it with "information on possible routes for people to move" (for example, roads, railways, etc.) on the map data. The output result of the pedestrian flow path information output unit 2102 is stored in the pedestrian flow path information 2312 of the storage device 2300.

[0092] Furthermore, by associating pedestrian flow information with traffic route information, the pedestrian flow path information estimation unit 2102 can also estimate the amount of people moving along each traffic route, and the estimation results are stored in the storage device 2300 as traffic route information 2304.

[0093] Figure 3 is a diagram illustrating the hardware configuration of the information service server 2000.

[0094] Referring to Figure 3, the information service server 2000 may be configured so that its own CPU (Central Processing Unit) performs the calculations, or it may be configured so that part of the program's processing is executed on another server. In the following explanation, we will assume that the CPU within the server itself performs the calculations.

[0095] Referring to Figure 3, the server 2000 comprises a computer device 2010, a network communication unit 2012 for communicating with the network, a recording medium (e.g., a memory card) 2210 for recording external data and providing it to the computer device 2010, a keyboard 2400 as an input device, and a display 2420 as a display device.

[0096] For example, the recording medium 2210 can be a USB memory stick, memory card, or external storage device. The network communication unit 2012 can utilize, for example, wired LAN or wireless LAN communication functions. The network communication unit 2012 and the input / output interface 2090 constitute the communication interface.

[0097] As shown in Figure 3, the computer body constituting this computer device 2010 includes, in addition to the disk drive 2030 and memory drive 2020, a CPU (Central Processing Unit) 2100 connected to the bus 2050, memory 2200 including ROM (Read Only Memory) 2200.1 and RAM (Random Access Memory) 2200.2, a non-volatile rewritable non-volatile storage device 2300, and an input / output interface 2090 for communication over a network and data exchange with the outside. The non-volatile storage device 2300 can be, for example, an HDD (Hard Disk Drive) or an SSD (Solid State Drive). In the following description, it will be described as an SSD. An optical disc can be installed in the disk drive 2030. A memory card 2210 can be installed in the memory drive 2020.

[0098] For the operation of the Computer Device 2010 program, the data and programs that store the information fundamental to the computer's operation will be described assuming they are stored on the SSD2300.

[0099] In Figure 3, the medium capable of recording information such as programs installed on the computer itself may be, for example, a DVD-ROM (Digital Versatile Disc), a memory card, or a USB memory stick. To accommodate such cases, the computer is equipped with drive devices (memory drive 2020, disk drive 2030) capable of reading these media.

[0100] The main components of the computer device 2010 consist of computer hardware and software executed by the CPU 2100. Generally, this software is stored and distributed on a storage medium or via a network, retrieved via the disk drive 2030 or network communication unit 2012, and temporarily stored in the SSD 2300. It is then read from the SSD 2300 into the RAM 2200.2 in memory and executed by the CPU 2100. However, if a network connection is established, the software may be loaded directly into RAM and executed without being stored in the SSD 2300.

[0101] The program for functioning as computer device 2010 does not necessarily need to include an operating system (OS) that causes the computer 2010 to execute functions such as information processing devices. The program only needs to contain the instruction portion that calls appropriate functions (modules) in a controlled manner to obtain the desired result. How computer system 2010 operates is well known, and a detailed explanation is omitted.

[0102] Furthermore, the CPU 2100 may be a single-core processor or a multi-core processor. In other words, it may be a single-core processor or a multi-core processor. Also, the server 2000 may be configured with multiple servers to perform distributed processing.

[0103] Furthermore, the hardware configuration of user terminal 3000 and data provision server 5000.i (1≦i≦M) is basically the same, so we will not repeat the explanation.

[0104] Figure 5 is the first diagram illustrating the processing flow of the information service provider server 2000. Figure 6 is the second diagram illustrating the processing flow of the information service provider server 2000.

[0105] Figure 7 shows an overview of the data output from the arithmetic unit 2100.

[0106] Figure 8 is a conceptual diagram showing the data generated when the number of people in a facility is estimated from the number of people in each activity at the first time point.

[0107] Figure 9 is a conceptual diagram showing the data generated when the number of people in a facility is estimated from the number of people in each activity at the second time point.

[0108] Referring to Figures 5 and 6, once processing starts, the computing unit 2100 acquires spatial statistical data via the network (S110, S200).

[0109] As mentioned above, it is possible to use "Mobile Spatial Statistics (registered trademark)" provided by NTT Docomo to obtain spatial statistical data. Spatial statistical data is, for example, data on the number of people by age and gender within a predetermined time unit (for example, 1 hour) within a predetermined width mesh (for example, 500m square).

[0110] Next, with respect to the "predetermined area" specified by the area setting unit 2106 based on the user's specifications, the lifestyle information estimation unit 2110, using spatial statistical data, classifies the people (age and gender obtained from spatial statistical data) who are staying in a specific area (for example, the Osaka Umeda area) during a specific time period (for example, a certain hour) into more detailed time units (for example, 15 minutes) within that specific time period (behavioral classification information), such as work and movement (e.g., commuting), by referring to the Social Life Basic Survey Database, and estimates the number of people (attributes of age and gender) in each classification (S120, S210).

[0111] The lower part of Figure 7(a) shows an overview of this type of activity-based population estimation.

[0112] The lower part of Figure 8 shows the estimated number of people engaged in specific activities at 2:00 AM. Given the time of day, most people fall into the "sleeping" activity category.

[0113] The lower part of Figure 9 shows the estimated number of people engaged in specific activities at 3 PM (15:00). Given the time of day, most people are categorized as either "working" or "studying."

[0114] Returning to Figures 5 and 6, the lifestyle information estimation unit 2110 further estimates the number of employed persons (with age and gender attributes) by occupation every 15 minutes for the above-mentioned specific domain, based on the estimated number of persons (with age and gender attributes) for each activity category, for example, based on the Ministry of Internal Affairs and Communications Labor Force Survey Database (S130, S220).

[0115] Next, the area location information estimation unit 2108 expands the estimated number of employed persons (attributes of age and gender) by occupation every 15 minutes to each building (facility) based on the total floor area of ​​facilities and buildings on the map and the area allocated to each occupation in each facility and building (for example, proportionally apportions), and estimates aggregated data of the number of people in meshes smaller than or equal to the unit mesh of spatial statistical data (for example, a 62.5m mesh) (S140, S230).

[0116] Hereinafter, when referring to the estimation processing functions of the "domain location information estimation unit 2108" and the "lifestyle information estimation unit 2110" collectively, we will use the term "human behavior information estimation unit."

[0117] The upper part of Figure 7(a) shows an overview of this facility population estimation.

[0118] The upper part of Figure 8 shows the estimated number of people in a specific area at 2:00 AM. Given the time of day, most people are located in either their "residence" or "hotel / inn."

[0119] The upper part of Figure 9 shows the estimated number of people in a specific area at 3 PM (15:00). Based on the time of day, it can be seen that most people are assigned to the "job type" or "task" corresponding to the facility / building at each facility.

[0120] Returning to Figures 5 and 6, the map output unit 2104 further overlays the number of people for each facility / building onto the map and outputs it color-coded, for example, using a heatmap method. Here, the heatmap may be displayed using a color-coded chart showing the proportion of people performing tasks at that location.

[0121] Figure 7(b) shows an example of output that visualizes this distribution of people.

[0122] Figure 10 shows the concept of the process for improving the accuracy of spatial decomposition in the action-based number estimation process (S120, S210).

[0123] As shown in Figure 10, when performing analysis using spatial statistical data (or pedestrian flow data) and various other information such as map information, some data may have a spatial resolution of 500m units, others 100m units, and still others at the land parcel level (address level).

[0124] Because the data units do not match, the analysis is usually performed using the larger unit size.

[0125] However, as mentioned above, the information service server 2000 stores all data, including pedestrian flow data (including population distribution data at a given time), building data, facility data, and land data (including information on roads, railways, and road / water areas), by breaking them down and converting them into the same unit (area mesh).

[0126] A database where data is stored in such a uniform unit will be referred to as a "data lake" below. Therefore, for example, it is possible to perform real-time analysis at any location throughout Japan using the same unit (regional mesh).

[0127] Furthermore, as mentioned above, the same region mesh is defined as a mesh with sides of 62.5m, as shown in Figure 10.

[0128] If we assume a distance of 62.5m, and since Mobile Spatial Statistics (registered trademark) uses a mesh with sides of 500m, we can create a region mesh with sides that are approximately equal to the distance a person walks in one minute (1 / 8 of 500m divided by 2). Similarly, when using other spatial statistics, it is possible to unify the sides to a region mesh corresponding to the distance a person walks in one minute (hereinafter referred to as the "estimated region mesh"). By using the distance a person walks in one minute, it can be said that the resolution is suitable for analyzing pedestrian flow, including people moving on foot.

[0129] However, it is possible to divide the data into units smaller than 62.5m (for example, a minimum of 1m). However, this would increase the amount of data to be analyzed and the computational load, so from a cost-effectiveness standpoint, it is preferable to use 62.5m, for example.

[0130] Figure 11 shows another example of a superimposed image of a map image and population distribution output by the map output unit 2104.

[0131] As shown in Figure 11, a predetermined area can be set up in which a certain size of region is designated, and furthermore, the user can specify a portion of the "predetermined area" where the estimation process is performed. [Embodiment 2]

[0132] Figure 12 is a functional block diagram illustrating the configuration of the pedestrian flow information service provider server 2000' in Embodiment 2.

[0133] The hardware configuration of the pedestrian flow information service provider server 2000' is the same as that described in Figure 3, so the explanation will not be repeated. In this specification, the "information service provider server 2000" and the "pedestrian flow information service provider server 2000'" are collectively referred to as the "pedestrian flow analysis device."

[0134] In this specification, "people flow information" means information that represents the "distribution and movement patterns of people" over a predetermined period. For example, it includes information on the estimated number of people staying in a user-specified area (hereinafter referred to as the "specific analysis target area") at a particular time or time period, based on the set attributes of the people. By tracking the changes in such people distribution information over time, it is also possible to estimate information that represents the flow of people. Here, "people attributes" include, but are not limited to, age, age group, gender, occupation classification, and "stay attributes" of people as described later.

[0135] Components identical to those in the information service server 2000 of Embodiment 1 are denoted by the same reference numerals, and, in principle, their descriptions will not be repeated.

[0136] Referring to Figure 12, the computing unit 2100 of the pedestrian flow information service provider server 2000' in Embodiment 2 includes a pedestrian flow information estimation unit 2103 instead of the pedestrian flow path information estimation unit 2102, and an analysis condition setting unit 2107 instead of the area setting unit 2106, as functions to be executed according to the program.

[0137] The pedestrian flow information estimation unit 2103 will be described later.

[0138] The area setting unit 2106 has the function of specifying a "predetermined area (a "specific area" designated by the user as the target of analysis)" by specifying, for example, a building or administrative district, based on instructions from the user. In contrast, the analysis condition setting unit 2107 has the function of specifying a "specific area" as the area to be analyzed by the user specifying buildings or administrative districts on the displayed map, as well as by specifying the boundaries of the area using operations such as a mouse or touch panel. The analysis condition setting unit 2107 also accepts the user's specification of the attributes of the person performing the analysis. Furthermore, the analysis condition setting unit 2107 identifies a set of predetermined unit sub-regions (for example, unit regions of the same shape and area) that cover the area to be analyzed without overlapping each other.

[0139] Here, although not particularly limited, in this embodiment, the area to be analyzed is defined as a circle with a predetermined radius centered at a specified point, and the unit sub-region is defined as a rectangular area of ​​the same shape. Therefore, the set of these unit sub-regions is set to cover the area to be analyzed by being adjacent to each other without overlapping, so that in the periphery of the area to be analyzed, the unit sub-regions only need to include, for example, a part of the area to be analyzed.

[0140] Furthermore, while it is desirable for the boundaries of the unit sub-region to coincide with the boundaries of the estimated region mesh, this is not necessarily the only possible situation. For example, although not particularly limited, a unit sub-region may be set to correspond to multiple estimated region meshes. In this case, the aggregated data of the estimated region meshes included in the unit sub-region becomes the aggregated data of the unit sub-region. Alternatively, if the unit sub-region is not aligned to have a multiple relationship with the size of the estimated region mesh, the aggregated data of the unit sub-region may be estimated by distributing the number of people with activity information within the estimated region mesh to the area of ​​each occupation of facility / building within the unit sub-region for each overlap between the unit sub-region and the estimated region mesh.

[0141] Furthermore, the storage device 2300 of the human flow information service provider server 2000' stores map information 2314 as a digital map, as will be described later.

[0142] Therefore, the storage device 2300 of the human flow information service provider server 2000' stores demographic information (e.g., "Mobile Spatial Information (registered trademark)"), lifestyle information, and area location information as data acquired from an external server via the network. The demographic information is aggregated demographic data for each first time interval (e.g., 1 hour), and contains information on the number of people by age, gender, and place of residence located in each area mesh (e.g., 500m square) within a predetermined area at the first time interval. On the other hand, the predetermined time unit for lifestyle information is a second time interval (e.g., 15-minute intervals) that is shorter than the first time interval, and the lifestyle information is information that can associate a person's age, gender, and predetermined classification of a person's life and behavior with a person's occupation. The location information for each predetermined section of the area location information includes the total floor area of ​​facilities within the predetermined area and information on occupations within those facilities.

[0143] The computing unit 2100 of the human flow information service provider server 2000' performs the function of a human behavior information estimation unit that, based on a program, estimates the number of people with each category of human behavior and attribute located in a region mesh (or, preferably, an estimated region mesh) included in the area to be analyzed, at predetermined time intervals (first time interval, or preferably a second time interval), based on demographic information and lifestyle information.

[0144] More specifically, the lifestyle information estimation unit 2110, which functions as a human behavior information estimation unit, i) estimates the number of employed persons by occupation located in a predetermined area for each area mesh (or, preferably, an estimated area mesh) at second time intervals based on demographic information and lifestyle information, and ii) as described later, the human flow information estimation unit 2103 apportions the number of employed persons by occupation to each facility based on the total floor area and the area allocated to each occupation in each facility, and estimates aggregated data of the number of people in the area mesh (or a smaller mesh (estimated area mesh, or unit sub-area)) of the demographic information.

[0145] Furthermore, the storage device 2300 stores information on the type of stay in relation to the demographic data. The demographic information is aggregated demographic data for each first time interval, and contains information on the number of people (for example, age and gender) located in each area mesh within a predetermined area during the first time interval. Here, the attributes of the people include information on the type of stay in which the person is staying within the predetermined area.

[0146] "Information on type of stay" includes, as described later, the attributes of residents in a given area, the attributes of workers in a given area, and the attributes of non-residents and non-workers.

[0147] The function of the pedestrian flow information analysis within the target area, which is performed by the pedestrian flow information estimation unit 2103 based on the program of the computing unit 2100, will be explained in more detail below.

[0148] As a prerequisite, the lifestyle information estimation unit 2110 of the computing device 2100 estimates the number of people for each classification of human behavior located in the area mesh included in the analysis area at each first time interval (or preferably a second time interval), based on demographic information and predetermined lifestyle information (behavioral information and occupational information), and performs human behavior information estimation. The human flow information estimation unit 2103 estimates aggregated data for each estimated area mesh within the analysis area by distributing the number of people estimated by the lifestyle information estimation unit 2110 to the estimated area mesh, based on the estimation results of the lifestyle information estimation unit 2110 and the type and size of the location (buildings, facilities, roads, transportation, etc.) corresponding to the classification of human behavior for each "area mesh (for example, a 500m mesh)" or preferably an "estimated area mesh (for example, a 62.5m square)".

[0149] As described above, the pedestrian flow information estimation unit 2103 directly or indirectly estimates aggregated data for each unit sub-area. At this time, the pedestrian flow information estimation unit 2103 estimates the distribution of human attributes (including a person's gender, age or age group, occupation, type of stay, etc.) specified by the user for each unit sub-area within the analysis target area, based on the classification of human behavior estimated in the estimated area meshes within the analysis target area and the aggregated data.

[0150] While not particularly limited, it is desirable that, similar to Embodiment 1, the lifestyle information estimation unit 2110 not only performs estimations in a first time unit based on demographic information and predetermined lifestyle information, but also estimates the number of people for each category of human behavior located in the area mesh included in the area to be analyzed for each second time unit which is shorter than the first time unit.

[0151] Figure 13 is a conceptual diagram showing an embodiment in which a predetermined area RM is specified on a digital map.

[0152] The Region RM may be specified as a circle with a specified radius centered at a user-specified location, as shown by the dotted line area in the map in Figure 13. Alternatively, the user may specify a rectangular area, as shown in the map in Figure 13. The shape of the Region RM is not limited to these.

[0153] However, in the following explanation, the region RM will be described as being designated by a circle, as shown by the dotted line in Figure 13.

[0154] As shown in Figure 13, a predetermined area RM is displayed as a digital map. At this time, roads Rd and rivers RV are displayed on the digital map.

[0155] Figure 14 is a conceptual diagram illustrating the data structure of a digital map.

[0156] As shown in Figure 14, digital map data consists of multiple layers.

[0157] For example, the map information includes "management data" that is not illustrated, such as the secondary mesh code, the base map used, geomagnetic declination, the actual distance between the corners of the district, the update date of each data, the number of records per data, and the number of items per data.

[0158] For example, Layer 1 is road information, which is stored as a road information database within map information 2314.

[0159] Furthermore, road information could be divided into, for example, "basic road data" and "minor road data."

[0160] Here, "basic road data" is not particularly limited, but includes, for example, "basic road node data," "basic road link data," and "basic road link attribute data." Similarly, "minor road data" is not particularly limited, but includes, for example, "minor road node data," "minor road link data," and "minor road link attribute data."

[0161] Here, "node data" (collectively referring to "basic road node data" and "minor road node data") includes node number, location coordinates, elevation, node type, number of connecting links, and connecting node number. "Basic road node data" also includes intersection names in addition to these.

[0162] "Link data" (collectively referred to as "basic road link data" and "minor road link data") includes the link number (number of the start and end nodes), road administrator, road type, administrative area code, link length, width classification, number of lanes, and the position coordinates and elevation of the interpolation point. "Basic road link data" further includes the route number, roadway width, road census data (median width, 12-hour traffic volume, travel speed (peak hours), traffic regulations such as speed limits), position coordinates and elevation of the interpolation point, emergency transport road classification, and highway numbering. "Minor road link data" further includes the corresponding basic road link number.

[0163] "Link attribute data" (collectively referred to as "basic road link attribute data" and "minor road link attribute data") includes the location and name of link attributes (bridges / elevated structures, tunnels, tunnels, level crossings, pedestrian bridges, toll booths, underpasses, areas prone to road flooding, etc.).

[0164] Therefore, road information generally consists of node data to identify the location of road intersections, link data indicating the configuration of roads connecting the nodes, and in-link attribute data indicating the type of link.

[0165] Layer 2 represents, for example, building information as described later; Layer 3 represents, for example, contour lines; Layer 4 represents, for example, bodies of water (rivers, lakes, ponds, etc.); and Layer 5 represents, for example, administrative boundaries (prefectural borders, city / ward / town / village boundaries, district boundaries, etc.).

[0166] As a digital map, there may also be other layers containing information on the map.

[0167] For road information, you may use the contents of the following database. https: / / www.drm.jp / database /

[0168] Furthermore, the administrative boundary information may be structured in a manner similar to that of the Geospatial Information Authority of Japan's database.

[0169] Figure 15 is a conceptual diagram showing an example of building information.

[0170] Building information is a collection of information about each building or facility, for example, having the following structure:

[0171] Individual building information consists of details such as "Building ID," "Building Name," "Latitude and Longitude of the Building's Representative Point," "Building Use," "Building Height (Height from Reference Point)," "Number of Floors (Basement, Above Ground)," "Total Floor Area," and "Floor Plan Information for Each Floor."

[0172] A "representative point of a building" is, for example, the location of the building's entrance.

[0173] A "reference point for height" is, for example, the ground level.

[0174] "Floor plan information for each floor" is not particularly limited, but includes the FPID (Financial Plan ID) for each floor, the intended use of the area on that floor (parking lot, retail (type of retail), office, etc.), the area of ​​each area, and the attributes of each area for each time period (residential facility, commercial facility, tourist facility, accommodation facility, etc.).

[0175] In other words, in building information (including facility information and location information), if the same building (facility, location) has different attributes (uses) depending on the date and time, the configuration may define the respective attributes for each date and time.

[0176] Furthermore, the building information includes 2D or 3D polygon information (not shown) of the building, associated with the building ID.

[0177] In 2D polygon information, the shapes of buildings and houses are represented by polygons in a GIS (Geographic Information System). 3D polygon information represents the shape of each floor of a building, including its overall shape, in addition to its overall form, in the height direction.

[0178] Furthermore, while there are no particular limitations on the specifications for indoor information, they may, for example, conform to the following "Draft Specifications for 3D Indoor Geospatial Information Data." https: / / www.gsi.go.jp / common / 000212582.pdf

[0179] As described in Embodiment 1, the building owner or manager can register building information (such as floor plans or 2D or 3D polygon data) via the building information registration terminal 4000. Alternatively, the building information registration terminal 4000 may be a mobile terminal such as a smartphone with a photo-taking function, in which visitors to the building can register information such as photos of the building's interior. For example, in the case of a smartphone, the pedestrian flow information service provider server 2000' may receive notifications from users using a terminal with application software installed that has an input format that allows them to notify the building name, floor number, and area information (such as store name and type of store) along with the photos they have taken.

[0180] Furthermore, the pedestrian flow information service provider server 2000' may be configured to store building information as map information 2314 in the storage device 2300.

[0181] As described above, the map information 2314 is a database that stores polygon information of buildings and facilities, roads, and administrative boundaries used by the analysis condition setting unit 2107 when specifying the boundaries of a predetermined area, and the building information registration terminal 4000 accepts the registration of building information (including 2D or 3D polygon information) to the database. Since buildings and the like are constantly updated, it is desirable to have a configuration that can accept the provision of polygon information from external sources.

[0182] Furthermore, by accepting design data for unstarted or unfinished buildings and facilities within a predetermined area ("analysis target area") from the building information registration terminal 4000, it is also possible to predict and estimate the number of future visitors to these buildings and facilities based on this design data. In this case, for example, if the design data specifies the attributes of each floor area for unstarted or unfinished buildings and facilities, such prediction and estimation becomes possible by multiplying the number of people staying within the analysis target area by a predetermined coefficient (for example, 1.3) and distributing this number as the predicted aggregate number within the analysis target area. Here, the predetermined coefficient can be configured to determine its value according to, for example, the ratio of the total floor area of ​​unstarted or unfinished buildings and facilities to the total floor area of ​​currently completed buildings and facilities within the analysis target area.

[0183] The above explanation assumes that the distribution of people within a specified analysis area will be estimated following this process. In other words, the process will generally proceed as follows.

[0184] i) For a given area, "spatial statistical data" is obtained, aggregated based on information such as mobile phone usage, for each given area mesh and for each first time interval.

[0185] ii) For people staying within a designated area (area of ​​analysis) at a first time interval, refer to the Social Life Basic Survey DB and classify them into behavioral classification information for a second time interval shorter than the first time interval, and estimate the number of people in each classification.

[0186] iii) For the area to be analyzed, the number of people by occupation for each of the second time intervals is estimated by referring to the Ministry of Internal Affairs and Communications Labor Force Survey DB, etc.

[0187] iv) For the area under analysis, the number of people by occupation for each second time interval is allocated based on the total floor area of ​​each area of ​​the building on the map and the occupation information of each building. This allows for the estimation of the distribution of people (aggregated data) for a finer mesh (estimated area mesh) than the predetermined area mesh of the spatial statistical data.

[0188] As also mentioned in Embodiment 1, the first time interval and the second time interval may be the same.

[0189] However, this method of apportionment of the number of people does not necessarily take the following points into consideration.

[0190] Let's say we specify the number of people of each age and gender in a certain area as "Attribute 1," the number of people from each region of origin who visited the area as "Attribute 2," and the number of people from each purpose of visit as "Attribute 3."

[0191] Here, the "place of origin" refers to the original location of the person who visited the aforementioned "certain area." In Japan, this could be categorized by prefecture or municipality, while for foreigners, it would typically be categorized as outside of Japan. However, if, for example, a smartphone app allows a user to register their place of origin before starting to use the smartphone in Japan and grants permission for its use, then that place of origin can be used.

[0192] Furthermore, as mentioned above, the purpose of visit can be classified into three categories based on "type of stay information": attributes of residents of the area, attributes of workers in the area, and attributes of people who are neither residents nor workers in the area.

[0193] However, due to discrepancies between aggregated data caused by differences in estimation methods, data anonymization, and aggregation accuracy, when combining all human flow data or different data sets, problems may arise where the numbers don't match even though data for each individual should have been broken down and aggregated by attribute.

[0194] Therefore, a problem may arise in that it is difficult to accurately analyze the flow of people when multiple conditions are set, such as a male tourist in his 30s in Ota Ward, Tokyo.

[0195] However, if the objective is to analyze the tendency for people with certain attributes to gather in a particular area, then it is more important that the trends when multiple attributes are considered are accurately reflected in the pedestrian flow analysis, rather than the consistency between individual aggregated data points.

[0196] Of course, one could consider configuring the system based on the premise of registering the detailed attributes of individual smartphone users. However, such a configuration would create another problem: the aggregated data would not necessarily align with the objective of making data collection easier while ensuring confidentiality.

[0197] Therefore, in this embodiment of pedestrian flow analysis, we will describe a method that makes it possible to quantitatively evaluate the "tendency for people to gather in specific areas" for each attribute of a person by using attribute data of people associated with demographic data, and also makes it possible to consistently estimate the "tendency for people to gather in the above-mentioned specific areas" for "people" with multiple attributes combined.

[0198] Furthermore, when distributing aggregated data to a certain facility or building using the aforementioned "type of stay information," a configuration may be adopted that more accurately estimates the distribution of people by considering the following priorities.

[0199] i) For example, assuming that the behavior is classified as sleep during the night, residents within the analysis area should be preferentially allocated to housing. On the other hand, those who are neither residents nor employees within the analysis area (temporary visitors) should be preferentially allocated to accommodation facilities within the analysis area (e.g., hotels).

[0200] ii) Similarly, during weekday daytime hours, workers within the area under analysis should be preferentially allocated to buildings with attributes such as commercial facilities and business offices.

[0201] Therefore, in Embodiment 2, an attribute called "type of stay" is assigned to each person staying within the analysis target area during each time period, and the priority of the allocation of people is set in advance on the people flow information service provider server 2000' so as to change the priority of the allocation of people according to the attributes of time period and location.

[0202] It should be noted that, although not particularly limited, attributes such as "type of stay" can be obtained from demographic data.

[0203] The human flow information estimation unit 2103 then prioritizes allocating residents to residential facilities, employees to business facilities (such as office buildings and shops within commercial facilities), and others to short-term stay locations and entertainment facilities (short-term stay locations are assumed to include accommodations such as hotels and inns, and entertainment facilities are assumed to include theme parks, parks, museums, various experiential facilities, and aquariums).

[0204] The human flow information estimation unit 2103 also considers the attributes of the location for each time period when assigning people to locations. For example, if the attributes are ("resident" + "sleep" behavior information), it will be preferentially assigned to residences, and if the attributes are ("non-resident / non-worker" + sleep behavior information), it will be preferentially assigned to hotels.

[0205] In other words, as explained in Figures 34 and 35, first, in "Mobile Spatial Statistics (registered trademark)," information on the residential area is included as information on the type of stay.

[0206] Here, during the de-identification process shown in Figure 35, it is possible to extract and count the number of "residents" of the area under analysis and those who live outside the area but regularly move to the area under analysis on weekdays (workers). Then, it is possible to classify and count "people who are neither residents nor workers" who are staying in the area under analysis during a certain time period as "temporary visitors".

[0207] In this case, "temporary visitors" are, for example, tourists or shoppers, so if there are entertainment or tourist facilities such as theme parks within the area being analyzed, it is possible to set it up so that priority is given to allocating visitors to those facilities during their operating hours.

[0208] As described above, the map information 2314 of the information service provider server 2000 contains registered attributes of buildings or places, and the area location information 2310 is pre-configured with priority settings for the "type of stay" when distributing building or place attributes (residential, commercial, office, accommodation, entertainment / tourism facilities, etc.). [In this embodiment, "human absorption capacity value" and "human absorption capacity vector value"]

[0209] In the following, we define a "people absorption value" for specified human attributes as an indicator that objectively represents the "tendency for people to gather" in each area, as described above, and we also define a "people absorption vector value" as an indicator that represents the "tendency for people with a combination of attributes to gather" for a combination of multiple human attributes. (Human absorption capacity)

[0210] For example, if the area to be analyzed is 10,000 m 3 Let's assume we have a flat area of ​​land where 1,000 people live. Now, let's say we divide this area into 1,000 unit sub-regions.

[0211] Conceptually, the land within the area is empty, and if it were divided into 1,000 sections, which unit sub-area would be 10m²? 3 ) The same conditions apply, 10m 3 The number of people per unit is 1. The value obtained by simply dividing the total number of people with the specified attribute within the analysis area by the number of unit sub-areas is called the "mean population distribution value." When people are distributed evenly across the analysis area and there is no bias in the distribution within the area, the number of people located in a unit sub-area is equal to the "mean population distribution value."

[0212] However, the placement of a specific store in a particular area can lead to a bias in the distribution of people, as those who wish to purchase from that store will gather there.

[0213] This can be understood as a situation where the presence of a "store," which has the power to attract people, has caused a bias in the distribution of people.

[0214] Therefore, a "human absorption value" is defined for each unit sub-domain as a value that represents the extent to which the "force that attracts people" is acting on a specified location or point within the area under analysis. (Human absorption capacity value) (Aggregated data of the specified attribute estimated for a unit subregion) / (Mean population distribution value)

[0215] Therefore, the sum of the "human absorption capacity values" within the area under analysis will always be equal to the number of area divisions.

[0216] In other words, when the numerator, "aggregate data of specified attributes estimated for a unit sub-region," is summed up for the region being analyzed, it becomes "aggregate data of specified attributes estimated to be located in the region being analyzed," while the denominator is equal to (aggregate data of specified attributes estimated to be located in the region being analyzed) / (number of region divisions).

[0217] Therefore, the following relationship holds true.

[0218] a) If the human absorption capacity value is equal to the mean human distribution value, then there is a force at work that absorbs (attracts) an average person.

[0219] b) If the human absorption capacity value is greater than the mean human distribution value, it indicates a strong ability to absorb (attract) people.

[0220] c) If the human absorption capacity value is smaller than the mean human distribution value, it indicates that a person has a weak ability to absorb (attract) things. (Human absorption force vector value)

[0221] Since the "human absorption power value" for each attribute is defined, for example, for the "human absorption power vector value" of two attributes, it is defined as the "human absorption power vector" with the human absorption power values of each attribute as components, as follows, as the "human absorption power vector value". (Definition of the human absorption power vector value ( = human absorption power value) in the case of degree 1) When the human absorption power value of attribute 1 is x, (human absorption power vector value of degree 1) = |x| (Definition of the human absorption power vector value in the case of degree 2)

[0222] Let the human absorption power value of attribute 1 be x and the human absorption power value of attribute 2 be y. (Human absorption power vector value of degree 2) = |x·y| / (x 2 +y 2 ) 1 / 2 (Definition of the human absorption power vector value in the case of degree 3)

[0223] Let the human absorption power value of attribute 1 be x, the human absorption power value of attribute 2 be y, and the human absorption power value of attribute 3 be z. (Human absorption power vector value of degree 3) = |x·y·z| / ((x·y) 2 +(y·z) 2 +(z·x) 2 ) 1 / 2

[0224] For higher degrees, it can be defined similarly.

[0225] Note that as the calculation method of the high-order human absorption power vector value, it is not necessarily limited to the above calculation formula, and other calculation formulas may be used as long as they satisfy the following condition 1 and condition 2.

[0226] Condition 1: When calculating the "power to attract humans" for a "human" corresponding to a combination of two or more attribute values, if the human absorption power value corresponding to any one of the attributes is 0, then the (human absorption power vector value) is 0. This is because the fact that the human absorption power value corresponding to that attribute is 0 means that it does not have that attribute.

[0227] Condition 2: In combinations of two or more attribute values, the human absorption vector value increases monotonically as the human absorption value corresponding to each attribute increases.

[0228] Condition 3: The human absorption force vector value takes a non-negative value. The value may be normalized to fall within a predetermined range.

[0229] Figure 16 is a flowchart showing the process for estimating the distribution of the number of people in each unit sub-region within the area being analyzed.

[0230] Referring to Figure 16, the human flow information service provider server 2000' first performs the analysis condition setting process (S102) in response to an operation from the user, where the analysis condition setting unit 2107 sets the analysis target area, analysis target attributes, and analysis time period. Here, the analysis time period may be a specified date and time for analysis, or a set time period. Also, the analysis target attributes refer to the attributes of at least one person to be analyzed. As for the analysis target attributes, all attributes may be specified (equivalent to not specifically specifying the attributes of the person).

[0231] Next, the computing unit 2100 acquires spatial statistical data via the network (S110).

[0232] As mentioned above, it is possible to use "Mobile Spatial Statistics (registered trademark)" provided by NTT Docomo to obtain spatial statistical data. Spatial statistical data is, for example, data on the number of people by age and gender within a predetermined time unit (for example, 1 hour) within a predetermined width mesh (for example, 500m square).

[0233] The lifestyle information estimation unit 2110 further acquires information on stay attributes from spatial statistical data (S112).

[0234] Next, with respect to the "analysis target area" specified by the area setting unit 2106 based on the user's specifications, the lifestyle information estimation unit 2110, based on spatial statistical data, classifies the people (age and gender obtained from spatial statistical data) who are staying in the specific area during a specific time period (for example, one hour) into more detailed time units (for example, 15 minutes) within that specific time period, such as work and movement (e.g., commuting) (behavioral classification information), by referring to the Social Life Basic Survey Database, and estimates the number of people in each classification (for age and gender attributes, and for each stay attribute) (S120). Here, stay attributes include not only attributes directly identified by stay type information, but also attributes estimated from stay type information. "Attributes directly identified by stay type information" are, for example, the region from which the visitor came to a certain area, and "attributes estimated from stay type information" are, for example, the purpose of visiting a certain area (work, residence, or non-work activities, for example, estimated as "tourism").

[0235] Furthermore, the lifestyle information estimation unit 2110 estimates the number of people (attributes of age and gender, and stay) for each activity category every 15 minutes for the specific area mentioned above, based on, for example, the Ministry of Internal Affairs and Communications' Labor Force Survey database (S130).

[0236] Next, the area location information estimation unit 2108 estimates the area corresponding to each occupation in the total floor area of ​​facilities and buildings located in the "predetermined location" of the "predetermined area" set by the area setting unit 2106, based on the area location information 2310. Similar to Embodiment 1, it is desirable to perform such estimation in units of the second area mesh (estimated area mesh or unit sub-area). However, depending on the user's settings, it is also possible to perform such estimation in units of the first area mesh, where each side is a multiple of the second area mesh.

[0237] The human flow information estimation unit 2103 then distributes the estimated number of people employed by occupation every 15 minutes (by age, gender, and stay attributes) to each building (facility) based on the total floor area of ​​facilities and buildings on the map, the area allocated to each occupation in each facility and building, and the attributes of each building (facility), and estimates aggregated data of the number of people in meshes smaller than or equal to the unit mesh of the spatial statistical data (S142). As described above, in estimating the "aggregated data of the number of people in meshes smaller than or equal to the unit mesh of the spatial statistical data," the human flow information estimation unit 2103 directly or indirectly ultimately estimates aggregated data for each unit sub-region.

[0238] Finally, the pedestrian flow information estimation unit 2103 estimates the distribution of people with the specified attribute within the analysis area based on aggregated data of the specified attribute for each unit sub-area. More specifically, the pedestrian flow information estimation unit 2103 calculates a pedestrian absorption value, which is the ratio of the number of people with the analysis target attribute for each unit sub-area to a predetermined value (S150). As mentioned above, the "predetermined value" can preferably be the "average pedestrian distribution value," but for example, the system may be configured to calculate the pedestrian absorption value using another value specified by the user.

[0239] Next, the pedestrian flow information estimation unit 2103 returns to step S120 if the distribution of people has not been completed for all of the specified analysis target attributes (N in S160). If the distribution of people has been completed for all of the specified analysis target attributes (Y in S160), and if multiple analysis target attributes have been specified, the pedestrian flow information estimation unit 2103 ultimately calculates a higher-order pedestrian absorption vector value, and the map output unit 2104 displays the map as described later (S170), and the process ends.

[0240] Figure 17 is a conceptual diagram using a map of Osaka City as an example, where the analysis area is specified as a circle (radius 10 km) with Osaka Station as the center.

[0241] Figure 18 is a heatmap showing the human absorption capacity value, with a unit sub-region set to cover the analysis area of ​​Figure 17, and age specified as the analysis target attribute.

[0242] Figure 18(a) shows the human absorption capacity values ​​for each 500m mesh, for all age groups, within a 10km radius area centered on Osaka Station.

[0243] Figure 18(b) shows the human absorption capacity values ​​for each 500m mesh, with an area defined as a 10km radius centered on Osaka Station, and the analysis targeting only males and females aged 15 to 19.

[0244] Figure 18(a) shows that there was no tendency for people to gather at points R11 to R13, but Figure 18(b) shows that there is a tendency for people to gather at those points.

[0245] This shows that a strong force is generated in areas where schools such as universities are located, and it is possible to visualize that different attractive forces act on different attributes.

[0246] Figure 19 is a heatmap showing the human absorption capacity value, with a unit sub-region set to cover the analysis area of ​​Figure 17, and the visitor's place of residence (origin area) specified as the analysis attribute.

[0247] Figure 19(a) shows the human absorption capacity values ​​for every 500m mesh, targeting residents of Miyakojima Ward, Osaka City, Osaka Prefecture, within a 10km radius area centered on Osaka Station.

[0248] Figure 19(b) shows the human absorption capacity values ​​for every 500m mesh, focusing only on visitors from outside the Kinki region, within a 10km radius area centered on Osaka Station.

[0249] While residents tend to be distributed around their own addresses and nearby workplaces, visitors from outside the Kinki region tend to gather in specific locations.

[0250] FIG. 20 is a diagram showing the human absorption power values in a heat map by setting unit partial regions so as to cover the analysis target region in FIG. 17, designating the staying type as the analysis target attribute.

[0251] FIG. 20(a) shows the human absorption power values for each 500 m mesh, with the region centered on Osaka Station and a radius of 10 km set, targeting only those who work within the region.

[0252] FIG. 20(b) shows the human absorption power values for each 500 m mesh, with the region centered on Osaka Station and a radius of 10 km set, targeting only those who "are not away from home and not working" within the region.

[0253] Places with high absorption power for those who "are not away from home and not working" within the region are assumed to be tourist attractions, amusement parks, commercial facilities, etc. Since information can be collected in real time, reference information such as the discovery of tourist spots and the selection of candidate locations for new tourist facilities can be obtained.

[0254] FIG. 21 is a diagram showing the concept of synthesizing human absorption power vectors by setting unit partial regions so as to cover the analysis target region in FIG. 17, and taking the attributes of the people set in FIGS. 18(a), 19(a), and 20(a) as Attribute 1, Attribute 2, and Attribute 3, respectively, as an example.

[0255] The definition of the human absorption power vector value is as described above.

[0256] FIG. 22 is a diagram showing, as an example, the human absorption power values for each 500 m mesh in a heat map for people in their 30s who live outside the Kansai region and "are not away from home and not working" within the analysis target region.

[0257] It becomes possible to visualize the flow of people under conditions combining a plurality of specified attributes, enabling deeper analysis.

[0258] FIG. 23 is a diagram showing an example of analyzing the temporal change in the flow of people by designating a location within the analysis target region, the attributes of people, and the analysis target time zone.

[0259] Furthermore, all attributes are specified for the person.

[0260] Figure 23(a) is a graph showing the increase or decrease in the human absorption capacity value at Umeda Station on the Osaka subway, comparing it to the value one hour ago and the value at present time. The difference between the human absorption capacity value one hour ago and the value at present time corresponds to the "acceleration of human gathering."

[0261] In other words, as shown in Figure 23(b), when the graph in Figure 23(a) is rising upwards, it indicates that a strong pulling force is acting, and conversely, when it is falling downwards, it indicates that a strong force is acting to move the person away from it.

[0262] While not particularly limited, the following calculations can be performed to display "acceleration" in Figure 23(a).

[0263] i) Calculate the human absorption force vector value for each time interval and for a predetermined time interval prior to that interval.

[0264] ii) The following calculation is performed using the current vector human attraction force value current_vector and the previous vector human attraction force value previous_vector. iii)diff_vector = current_vector - previous_vector degree = asin(diff_vector) ×90 / π height = tan(radians(degree))

[0265] Here, asin(·) is the arcsine and tan(·) is the tangent.

[0266] Degree refers to the angle from the previous time point to the current time point in the graph, while height refers to the height from the previous time point to the current time point in the graph.

[0267] However, regarding the graph display in FIG. 23, it is not necessarily limited to such a configuration, and any other display method may be used as long as it is an index indicating the tendency of people to gather and the tendency of people to disperse.

[0268] As shown in FIG. 23, it can be read from the graph that at Osaka Metro Umeda Station, the acceleration in the + direction is high from 9:00 to 10:00, and a nearly constant negative acceleration occurs every hour from after 17:00 until 0:00. The force that moves people in the area at almost the same ratio continues after 17:00. It can be said that it captures that almost the same number of people leave the Umeda Station by boarding the train.

[0269] FIGS. 24 and 25 are diagrams showing examples of analyzing the acceleration of the time change of the flow of people by specifying the location within the analysis target area, the attributes of people, and the analysis target time zone.

[0270] FIG. 24 is a diagram showing the time change of the acceleration of the human absorption force vector value for Kobe Motomachi Station.

[0271] FIG. 24(a) shows the usage status of the station when only visits from outside the Kansai region on the day of analysis are specified.

[0272] FIG. 24(b) shows the usage status of the station when women in their 20s on the day of analysis are specified.

[0273] FIG. 25 is a diagram showing the time change of the acceleration of the human absorption force vector value for Osaka Station.

[0274] FIG. 25(a) shows the usage status of the station when only visits from outside the Kansai region on the day of analysis are specified.

[0275] FIG. 25(b) shows the usage status of the station when women in their 20s on the day of analysis are specified.

[0276] From FIGS. 24 and 25, it is possible to grasp the acceleration of the flow of people at each station, or rather, the congestion situation.

[0277] Furthermore, since it's possible to analyze the attributes of people using each station during specific time periods, this data can be used as a reference for planning commercial facilities, events, and other related activities.

[0278] Figure 26 shows an example of defining the analysis area to include the region where a specific event takes place.

[0279] Here, the "2023 Yodogawa Fireworks Festival" is given as an example of a "specific event."

[0280] In this case, the designated area set up as a viewing area for the fireworks display is shown in an enlarged view in the lower section of Figure 26.

[0281] Figure 27 shows the change in human attractiveness over time when, for the specified area described in Figure 26, all human attributes are specified, and the time period for analysis is specified as a 1-day period including the event time (0:00 to 24:00).

[0282] Figure 27 also shows the change in the human attraction value over time when a day when no event is held (reference day: reference time) is specified.

[0283] As shown in Figure 27, the maximum crowd attraction value at 8 PM on the day of the fireworks display was 14.397. Using the average of the crowd attraction values ​​at the same time over the past month as a reference, the value was 0.8698. Note that the period for calculating the average value for the reference day (reference time period) can be changed depending on the region and the nature of the event, and is not limited to the past month.

[0284] By taking the difference between the attraction value on the day and the average attraction value on the reference day, we can obtain an objective value of 13.5272 as an event effect value, for example.

[0285] This value serves as an indicator of the attractive force generated by the Yodogawa Fireworks Festival in this area. Furthermore, the analysis reveals that the attractive force value of 13.5272 on that day attracted 0.012% of the people within a 10km radius, resulting in 41,298 people being in this area.

[0286] Furthermore, it becomes possible to perform analyses such as determining that the duration of the event effect is determined by the period during which the difference between the attraction value on the day and the attraction value on the reference day exceeds a predetermined value.

[0287] By conducting this type of analysis for this fiscal year, it becomes possible to use it as a reference for planning the guidance of event staff on-site and for management plans of relevant supervisory authorities in subsequent years. (Embodiment 3)

[0288] In the above explanation, the configuration for estimating the distribution and flow of people within a predetermined area mesh was described, based on the premise that the storage device 2300 stores "population dynamics information 2302" which includes information on the number of people located in a predetermined area at least at a predetermined time.

[0289] Here, "specified time" refers to, for example, a first specified time interval, such as every hour, and "specified area" refers to a specific region, for example, a first area mesh unit, such as a 500m mesh (a 500m square). "Specified demographic information" refers to "demographic data," meaning "aggregated data," such as "mobile spatial statistics information (registered trademark)."

[0290] Furthermore, the information service server 2000 was designed to store all data, including pedestrian flow data (including population distribution data at a given time point), building data, facility data, and land data (including information on roads, railways, and road / water areas), after decomposing and converting them into the same unit (regional mesh). A database stored in such a uniform unit was referred to as a "data lake." Therefore, for example, real-time analysis could be performed at any point throughout Japan using the same unit (regional mesh).

[0291] Furthermore, while there are no particular limitations to the same regional mesh as described above, we explained it assuming, for example, that the mesh has sides of 62.5m (referred to as the "estimated regional mesh").

[0292] From the above perspective, the entities that reside within or move between estimated geographical meshes do not necessarily have to be limited to "people."

[0293] For example, it is possible to consider a "bus," which is a form of public transportation, as a "moving object."

[0294] Below, we will explain "data lakes" more generally, as a prerequisite for performing real-time analysis using the same unit (estimated area mesh).

[0295] In the following, we will refer to a system that performs real-time analysis of objects moving within or between estimated domain meshes as a domain analysis system, and the server that provides information within the domain analysis system as a "domain analysis server." [Systems based on domain analysis]

[0296] Figure 28 is a conceptual diagram showing the data hierarchy and data processing hierarchy used in the domain analysis system.

[0297] As shown in Figure 28, the "public domain," which is a public and open area, includes, for example, "urban operating systems" and data linkage platforms as "software that supports urban infrastructure."

[0298] In Japan, for example, the following documents from the Cabinet Office are publicly available. Publicly known documents: https: / / www8.cao.go.jp / cstp / stmain / a-whitepaper3_200331.pdf

[0299] Furthermore, as a public data linkage platform, for public transportation such as "buses" as mentioned above, the following data linkages have been realized: Publicly available document: "Guidelines for Dynamic Bus Information Format (GTFS Realtime)" https: / / www.mlit.go.jp / common / 001283242.pdf

[0300] Here, the "GTFS (General Transit Feed Specification) Realtime" format is a data format for transmitting real-time information from public transport operators, continuously updating and transmitting information such as route updates, service status announcements, and vehicle locations. For example, real-time data can be published from a bus location system operated by a public transport company, and data users can use it in combination with "GTFS-JP".

[0301] Currently, the most common method for managing such "dynamic moving object operation information" is the "GTFS Realtime" system described above. However, there are also so-called "dynamic management systems" used by operators to manage vehicle operations. These "dynamic management systems" utilize positioning devices such as GPS (Global Positioning System) and communication devices to determine the vehicle's location and understand its driving status in real time. When linked with a digital tachograph, the driving status can also be understood in real time. To understand vehicle routes, speeds, dwell times, etc., in real time, it is possible to use a "dynamic management system" provided by a specialized company (for example, an "ASP: Application Service Provider").

[0302] Information from digital tachographs and GPS is provided to the operator's office in the following manner: i) Acquisition of vehicle location information, etc. (from digital tachographs, GPS, etc. installed in the vehicle). ii) Transmission of vehicle location information, etc. to a specialized company's "Fleet Management ASP Center" (automatically transmitted via the network of a telecommunications carrier, such as a mobile phone company, using communication equipment installed in the vehicle). iii) Monitoring of the driving status of vehicles managed by the "Fleet Management ASP Center" on a computer in the office.

[0303] For example, if the operator of a domain analysis system has access to data from such a dynamic management system, it will be possible to obtain real-time location information for moving objects such as taxis and delivery trucks, not just buses as described above.

[0304] Alternatively, if each mobile device is equipped with a "communication device that utilizes a mobile phone network," it may be possible to obtain the location information of these mobile devices in real time by acquiring data from the mobile phone network, similar to the methods used in Embodiments 1 and 2.

[0305] Therefore, in the following, as "objects that move" within or between estimated area meshes, in addition to "people who are mobile phone users" as in Embodiment 1 and Embodiment 2, we will also assume, as a typical example, buses whose operation is managed using data in the "dynamic bus information format". However, as mentioned above, as moving objects, we will assume not only "taxis" as public transportation, but also moving objects such as "trucks" and "private cars", including objects from which location information can be collected in real time.

[0306] Conceptually, the "domain analysis system domain" of this embodiment exists on top of the data linkage platform described above.

[0307] As shown in Figure 28, the region analysis system is also expected to utilize data obtained from data linkage platforms for mobile location information, such as the "Docomo Spatial Statistics (registered trademark)" mentioned above, as well as weather data for a specified time period in a specified region obtained from the Japan Meteorological Agency server, and "disaster risk information" obtained from the National Research Institute for Earth Science and Disaster Resilience.

[0308] When such data is stored in the aforementioned "data lake," it is first stored as raw, general-purpose data, and then finalized for a specific time period. As mentioned above, it is then stored in a "system-unified format (for example, a unified estimated area mesh as the unit)."

[0309] As shown in Figure 28, data in a general-purpose data format provided to the "data lake" could, for example, be submitted from a smartphone via a smartphone application. The submitted data could be a subjective evaluation of the congestion situation at a specific facility in a specific location, or it could be data of the congestion situation captured in photos or videos.

[0310] Furthermore, above such a "data lake" lies a layer of data analysis, which corresponds to the data integration analysis processing performed by the domain analysis system mentioned above.

[0311] In particular, while spatial statistics are measured in one-hour increments, the processing described above enables data analysis and visualization in near real-time, for example, in 15-minute increments.

[0312] Furthermore, the data stored in such a "data lake" includes not only real-time location data and attribute data of moving objects (type of object, operating company, etc.), but also location information and attribute information (type of facility, operating entity, etc.) of facilities (hospitals, stations, commercial facilities, etc.) located within a unified estimated area mesh.

[0313] Figure 29 is a functional block diagram illustrating the configuration of the region analysis server 2000'' of Embodiment 3.

[0314] Figure 29 is a diagram that is compared with Figure 12 of Embodiment 2. Therefore, in Figure 29, functional blocks that perform the same function as in Figure 12 are given the same reference numerals, and the explanations are basically not repeated.

[0315] Furthermore, the hardware configuration of the domain analysis server 2000'' is the same as that described in Figure 3, so we will not repeat the explanation.

[0316] Referring to Figure 29, the storage device 2300 stores mobile body position information 2316, which is acquired from the data linkage platform for mobile body position information and updated at least at a first time interval, preferably at a second time interval that is shorter than that, as described in Figure 28.

[0317] Furthermore, the computing unit 2100 of the area analysis server 2000'' in Embodiment 3 includes a target distribution estimation unit 2101 as a function to be executed according to the program, instead of the pedestrian flow information estimation unit 2103 in Embodiment 2. The main difference between the functional configuration of the area analysis server 2000'' and the functional configuration of the pedestrian flow information service provision server 2000'' in Figure 12 lies in the processing executed by the target distribution estimation unit 2101.

[0318] The analysis condition setting unit 2107' of Embodiment 3 has the function of specifying a "specific area" as the area to be analyzed by the user specifying buildings, administrative districts, etc. on the displayed map, as well as specifying the boundaries of the area by operating a mouse or touch panel. The analysis condition setting unit 2107' also accepts from the user the specification of a "moving object" to be analyzed and its attributes, as well as the facility to be analyzed and its attributes. The analysis condition setting unit 2107' also identifies a set of predetermined unit sub-regions (for example, unit regions of the same shape and area) that cover the area to be analyzed without overlapping each other.

[0319] Here again, although not particularly limited, in this embodiment, as an example, the area to be analyzed is defined as a circle with a predetermined radius centered at a specified point, and the unit sub-region is defined as a rectangular area of ​​the same shape. Therefore, here again, the set of these unit sub-regions is set to cover the area to be analyzed by being adjacent to each other without overlapping, so that in the periphery of the area to be analyzed, the unit sub-regions only need to include, for example, a part of the area to be analyzed.

[0320] Furthermore, while it is desirable for the boundary of the unit sub-region to coincide with the boundary of the estimation region mesh, it is not necessarily limited to this situation. For example, although not particularly limited, the unit sub-region may be set to correspond to multiple estimation region meshes. In this case, the total number of estimation targets in the estimation region mesh included in the unit sub-region (if the estimation target is a person, aggregated data about people, similar to Embodiment 2) becomes the data for the number of estimation targets within the unit sub-region. Alternatively, if the unit sub-region is not aligned to have a multiple relationship with the size of the estimation region mesh, the number of estimation targets within the estimation region mesh may be distributed within the unit sub-region according to the ratio of the area of ​​the estimation region mesh to the area of ​​the overlapping portion, for each overlap between the unit sub-region and the estimation region mesh, and the number of estimation targets in the unit sub-region may be estimated. Note that if the "moving target" is a person, the number of people with specified attributes within the estimation region mesh can be estimated as aggregated data for the unit sub-region, similar to Embodiment 2.

[0321] Furthermore, the storage device 2300 of the area analysis server 2000'' stores, as described later, map information 2314 acquired from an external server (not shown), which includes not only location information of facilities within the area to be analyzed, but also facility attribute information (hereinafter referred to as "facility attribute information"). This map information 2314 is updated with information from the external server at predetermined intervals (longer than the first time interval).

[0322] Therefore, the storage device 2300 of the area analysis server 2000'' further stores demographic information (e.g., "Mobile Spatial Information (registered trademark)"), lifestyle information, and area location information as data acquired from an external server via the network, similar to Embodiment 2. The demographic information is aggregated demographic data for each first time interval (e.g., 1 hour), and contains information on the number of people by age, gender, and place of residence located in each area mesh (e.g., 500m square) within a predetermined area at the first time interval. On the other hand, the predetermined time unit for lifestyle information is a second time interval (e.g., 15 minutes) shorter than the first time interval, and the lifestyle information is information that can associate a person's age, gender, and predetermined classification of a person's life and behavior with a person's occupation. The location information for each predetermined section of the area location information includes the total floor area of ​​facilities within the predetermined area and information on occupations within those facilities.

[0323] The computing unit 2100 of the region analysis server 2000'' also performs the function of a human behavior information estimation unit based on a program, in the same manner as in Embodiment 2, to estimate the number of people with each category of human behavior target attributes located in the region mesh (or preferably estimated region mesh) included in the region to be analyzed, based on demographic information and lifestyle information, at predetermined time intervals (first time interval, or preferably second time interval).

[0324] Furthermore, the storage device 2300 stores information on the type of stay in relation to the demographic data. The demographic information is aggregated demographic data for each first time interval, and contains information on the number of people (for example, age and gender) located in each area mesh within a predetermined area during the first time interval. Here, the attributes of the people include information on the type of stay in which the person is staying within the predetermined area.

[0325] The following will provide a more detailed explanation of the functions of the computing unit 2100, based on the program, and the target distribution estimation unit 2101, which performs the analysis of moving objects (including people) within the target area and the analysis of pedestrian flow information.

[0326] As a prerequisite, the lifestyle information estimation unit 2110 of the computing unit 2100 estimates the number of people for each classification of human behavior located in the area mesh included in the analysis area at each first time interval (or preferably a second time interval), based on demographic information and predetermined lifestyle information (behavioral information and occupational information), and performs human behavior information estimation. When the moving object to be analyzed is a person, the mobile object distribution estimation unit 2115 of the target distribution estimation unit 2101 estimates aggregate data for each estimated area mesh within the analysis area by distributing the number of people estimated by the lifestyle information estimation unit 2110 to the estimated area mesh, based on the estimation results of the lifestyle information estimation unit 2110 and the type and size of the location (building, facility, road, transportation, etc.) corresponding to the classification of human behavior for each "area mesh (e.g., 500m mesh)" or preferably "estimated area mesh (e.g., 62.5m square)".

[0327] As described above, the moving object distribution estimation unit 2115, when the moving object being analyzed is a person, directly or indirectly, ultimately estimates aggregated data for each unit sub-region. At this time, the human flow information estimation unit 2103 estimates the distribution of human attributes (including gender, age or age group, occupation, type of stay, etc.) specified by the user for each unit sub-region within the analysis target area, based on the classification of human behavior and aggregated data estimated in the estimated region mesh within the analysis target area.

[0328] While not particularly limited, it is desirable that, similar to Embodiment 1, the lifestyle information estimation unit 2110 not only performs estimations in a first time unit based on demographic information and predetermined lifestyle information, but also estimates the number of people for each category of human behavior located in the area mesh included in the area to be analyzed for each second time unit which is shorter than the first time unit.

[0329] On the other hand, the mobile object distribution estimation unit 2115 of the target distribution estimation unit 2101 calculates, based on the mobile object location information 2316, the number of mobile objects with the attributes of the target of analysis that are staying in each estimated area mesh at predetermined time intervals (each first time interval, or more preferably each second time interval). Then, for each unit sub-area within the target area of ​​analysis, the mobile object distribution estimation unit 2115 estimates the distribution of the number of mobile objects with the attributes specified by the user.

[0330] On the other hand, the facility distribution estimation unit 2116 of the target distribution estimation unit 2101 calculates, based on map information 2314, the number of facilities with the attributes of the target of analysis that exist in each estimated area mesh at predetermined time intervals (each first time interval, or more preferably each second time interval). Then, the facility distribution estimation unit 2116 estimates the distribution of the number of facilities of the type and attributes specified by the user for each unit sub-area within the target area of ​​analysis.

[0331] Here, corresponding to the "human absorption value" in Embodiment 2, the "mobile body absorption value" is defined as an indicator that objectively represents the "tendency for mobile bodies to gather" in each area, and the "mobile body absorption value" is defined for the attributes of a specified mobile body. Then, by combining the types and attributes of multiple mobile bodies and the types and attributes of facilities, the "mobile body absorption vector value" is defined as an indicator that represents "the tendency for the mobile bodies under analysis to gather with respect to the combined attributes." (Motor vehicle absorption capacity value)

[0332] For example, if the area to be analyzed is 10,000 m 3 Let's assume a flat area of ​​land where 1,000 moving entities (including people) are located. Now, let's say we divide this area into 1,000 unit sub-regions.

[0333] Conceptually, the land within the area is empty, and if it were divided into 1,000 sections, which unit sub-area would be 10m²? 3) The same conditions apply, 10m 3 The number of moving objects per unit is 1 (or person). The value obtained by simply dividing the total number of moving objects of a specified type and attribute within the analysis area by the number of units in the unit sub-area is called the "mean moving object distribution value." When moving objects are distributed evenly across the analysis area and there is no bias in the distribution within the area, the number of moving objects located in the unit sub-area is equal to the "mean moving object distribution value."

[0334] However, if a specific facility is located in a particular area, and mobile objects tend to gather at that facility, then a bias will occur in the distribution of mobile objects.

[0335] This phenomenon can be understood as a state in which a bias occurs in the distribution of moving objects because a "force that attracts moving objects" is generated by a "specific facility".

[0336] Therefore, a "moving object absorption force value" is defined for each unit sub-region as a value that represents the extent to which a "force that attracts moving objects" is acting on a specified location or point within the area under analysis. (Moving object absorption capacity value) = (Estimated number of moving objects of a specified type and attribute per unit subdomain) / (Average moving object distribution value)

[0337] Therefore, the sum of the "mobile body absorption force values" within the area under analysis will always be equal to the number of area divisions.

[0338] In other words, the sum of the "number of moving entities of specified types and attributes estimated for a unit subdomain" in the numerator over the analyte domain equals the "number of moving entities of specified types and attributes estimated for the analyte domain," while the denominator is equal to (number of moving entities of specified types and attributes estimated for the analyte domain) / (number of domain divisions).

[0339] Therefore, the following relationship holds true.

[0340] a) If the moving object absorption force value is the average moving object distribution value, then a force is acting to absorb (attract) the average moving object.

[0341] b) If the moving object absorption force value is greater than the average moving object distribution value, the force absorbing (attracting) the moving object is strong.

[0342] c) If the moving body absorption force value is smaller than the average moving body distribution value, the force with which the moving body absorbs (attracts) is weak.

[0343] Furthermore, the "facility absorption capacity value" can be defined for each unit sub-region as follows: (Facility absorption capacity value) = (Estimated number of facilities of the specified type / attribute for a unit subdomain) / (Average facility distribution value)

[0344] Herein lies the following: (Average facility distribution value) = (Total number of facilities of specified type and attribute within the analysis area) / (Number of units in the sub-area)

[0345] In other words, although the time scale over which the location of facilities changes is much longer than that of moving objects, the number of facilities changes over time as they accumulate in certain locations or go out of business. This trend is defined as the "facility absorption capacity value." The interpretation of the absorption (attraction) trend for the "facility absorption capacity value" is the same as for the "moving object absorption capacity value."

[0346] Furthermore, as described above, it is preferable to use the "average mobile distribution value" or "average facility distribution value" as the denominator for the definition of "mobile body absorption capacity value" and "facility absorption capacity value." However, the system may also be configured to calculate the "mobile body absorption capacity value" and "facility absorption capacity value" using other values ​​as predetermined values, for example, as specified by the user. (Absorption force vector value)

[0347] Furthermore, since we have defined the "mobile body absorption value" and the "facility absorption value" (hereinafter collectively referred to as the "target absorption value") for each attribute, for example, using the "mobile body absorption value" for at least one attribute and, if necessary, the "facility absorption value" for at least one attribute, we define the "target absorption vector value" as follows, for a "target absorption vector" whose components are the target absorption values ​​for each attribute.

[0348] Furthermore, in the "Target Absorption Vector," the "Mobile Absorption Value" is a required element, while the "Facility Absorption Value" is an element specified by the user. This is because, under the time span of a typical analysis, the "Mobile Absorption Value" has statistical significance. (Definition of the target absorptive force vector value (=target absorptive force value) in the case of order 1) When the target absorption force value of attribute 1 is x, then (the target absorption force vector value of degree 1) = |x| (Definition of the target absorption force vector value for order 2)

[0349] Let x be the target absorption value for attribute 1, and let y be the target absorption value for attribute 2. (Value of the symmetric absorption force vector of degree 2) = |x·y| / (x 2 +y 2 ) 1 / 2 (Definition of the target absorption force vector value for order 3)

[0350] Let x be the target absorption value for attribute 1, y be the target absorption value for attribute 2, and z be the target absorption value for attribute 3. (Value of the target absorption force vector of degree 3) =|x·y·z| / ((x·y) 2 +(y·z) 2 +(z·x) 2 ) 1 / 2

[0351] The same definition can be applied to cases with higher degrees.

[0352] Furthermore, the calculation method for higher-order target absorption force vector values ​​is not necessarily limited to the above formula; other formulas are acceptable as long as they satisfy the following conditions 1 and 2.

[0353] Condition 1: When calculating the "force that attracts the object" for an "object" that corresponds to a combination of two or more attribute values, if the object absorption force value for any one of the attributes is 0, the (object absorption force vector value) will be 0. This is because an object absorption force value of 0 for an attribute means that the object does not possess that attribute.

[0354] Condition 2: In combinations of two or more attribute values, the target absorptive vector value increases monotonically as the target absorptive value corresponding to each attribute increases.

[0355] Condition 3: The target absorption force vector value is non-negative. The value may be normalized to fall within a predetermined range. (Specific examples of target absorptive force value and target absorptive force vector value)

[0356] Figure 30 is a conceptual diagram showing the first example of the target absorption capacity value.

[0357] Figure 30(a) shows the location information of internal medicine clinics within the divided unit sub-regions (attribute 1) superimposed on the map, while Figure 30(b) shows the target absorption capacity value obtained from the number of elderly residents in the divided unit sub-regions (attribute 2) superimposed on the map.

[0358] By calculating the target absorptive capacity value and the target absorptive capacity vector value for each of the subjects in Figure 30(a) and Figure 30(b), it is possible to visualize an indicator of whether or not there is a shortage of internal medicine services per elderly person within the area of ​​analysis.

[0359] The map output unit 2104 can also display these target absorption force vector values ​​superimposed on the map.

[0360] Figure 31 is a conceptual diagram showing a second example of the target absorption capacity value.

[0361] Figure 31(a) shows the distribution of "restaurants (number of establishments) located in divided sub-regions within the domain" (attribute 1) superimposed on a map; Figure 31(b) shows the distribution of "railway stations (number of stations) located in divided sub-regions within the domain" (attribute 2) superimposed on a map; and Figure 31(c) shows the target absorption capacity value obtained from "the number of people staying in each divided sub-region within the domain per hour" (attribute 3) superimposed on a map.

[0362] By calculating the target absorptive capacity value and then the target absorptive capacity vector value based on each of these distributions, it is possible to appropriately analyze the extent to which public transportation (railways) has an impact on restaurants.

[0363] The map output unit 2104 can also display these target absorption force vector values ​​superimposed on the map.

[0364] Furthermore, for example, by calculating target absorption vector values ​​using target absorption values ​​obtained by counting the number of buses per hour using GTFS real-time data for each sub-region into which the analysis target area has been divided, and target absorption values ​​obtained from the number of people staying in each sub-region per hour, and then overlaying and displaying these values ​​on a map, it becomes possible to visualize the bus routes that are used most frequently per hour.

[0365] Although one embodiment of the present invention has been described above, the present invention is not limited to the embodiments described above, and any modifications, improvements, etc. that can achieve the objectives of the present invention are included in the present invention.

[0366] For example, the series of processes described above can be executed by hardware or by software.

[0367] In other words, it is sufficient for the information processing system to have the functionality to execute the series of processes described above as a whole, and the type of functional blocks used to realize this functionality is not particularly limited to the examples in Figures 1, 2, through 12. Furthermore, the location of the functional blocks and databases is not particularly limited to those in Figures 2 through 12, and can be arbitrary. For example, at least some of the functional blocks and databases necessary for executing various processes may be delegated to a user terminal, etc. Conversely, the functional blocks and databases of the user terminal may be delegated to a server, etc.

[0368] Furthermore, a single functional block may consist of hardware alone, software alone, or a combination of both.

[0369] When a series of processes are executed by software, the programs that make up that software are installed on a computer or other device from a network or storage medium.

[0370] The computer may be a computer that is built into dedicated hardware.

[0371] Furthermore, a computer can be any computer capable of performing various functions by installing various programs, such as a server, a general-purpose smartphone, or a personal computer.

[0372] Such recording media containing programs may consist not only of removable media (not shown) distributed separately from the main unit to provide programs to users, but also of recording media provided to users, etc., that are pre-installed in the main unit.

[0373] In this specification, the step of describing a program to be recorded on a recording medium includes not only processes that are performed chronologically in that order, but also processes that are not necessarily performed chronologically, but are executed in parallel or individually.

[0374] Furthermore, in this specification, the term "system" refers to an overall system composed of multiple devices.

[0375] The embodiments disclosed herein are illustrative of configurations for specifically carrying out 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 rather than by the description of the embodiments, and modifications within the literal scope and equivalent meaning of the claims are intended. [Explanation of symbols]

[0376] 2 Network, 100 users, 200 smartphones, 1000 people flow analysis system, 2000 information service provision server, 2100 computing unit, 2103 people flow information estimation unit, 2104 map output unit, 2106 area setting unit, 2107 analysis condition setting unit, 2108 area location information estimation unit, 2110 lifestyle information estimation unit, 2300 storage device, 2302 demographic information, 2304 traffic route information, 2306 lifestyle information, 2308 occupational employment information, 2310 area location information, 2312 people flow path information, 2314 map information, 3000 lifestyle information user terminals, 5000.1~5000.M data provision server.

Claims

1. A storage device that stores predetermined statistical information including the number of multiple objects located in each of multiple predetermined area meshes and information on the attributes of the objects, The system includes a computing device that performs an analysis of the distribution of the target within the area to be analyzed, The object includes at least a first object that resides within the predetermined region mesh and is capable of moving between the predetermined region meshes, The predetermined statistical information includes dynamic statistical information in a series of first time units, which includes the number of first objects located in each predetermined area mesh and information on the first attributes of the first objects. The aforementioned computing device is An analysis condition setting means that accepts the designation of a predetermined analysis target area and analysis target attributes through user operation, and identifies a set of unit sub-regions that cover the said analysis target area without overlapping each other, The system includes a target distribution estimation means that estimates aggregated data for each unit sub-region within the analysis target region by distributing the number of the first targets to the unit sub-regions within the region mesh based on the dynamic statistical information, The area analysis device includes a target distribution estimation means which estimates the distribution of the first target attributes within the analysis area based on the aggregated data, and calculates a first target absorption value which is the ratio of the first target attributes for each unit sub-area to a first predetermined value.

2. The region analyzer according to claim 1, wherein the first predetermined value is an average value obtained by dividing the first target of the attributes to be analyzed within the region to be analyzed by the number of unit subregions present within the region to be analyzed.

3. The aforementioned subject further includes a second subject which is a facility installed within the predetermined area mesh, The predetermined statistical information includes facility distribution information that includes the number of the second object located in each predetermined area mesh and information on the second attribute of the second object, The attribute to be analyzed includes at least one of the first attributes and at least one of the second attributes, The aforementioned target distribution estimation means is i) Based on the facility distribution information, the number of the second target is distributed to the unit sub-regions within the area mesh, thereby estimating the number of the second target for each unit sub-region within the area to be analyzed. ii) For each of the second attributes, estimate a second target absorption value which is the ratio of the number of the second target attributes for each unit sub-region to a second predetermined value, iv) A region analyzer according to claim 1 or 2, which calculates a target absorptive vector value having the first target absorptive value and the second target absorptive value as components.

4. The region analyzer according to claim 3, wherein the second predetermined value is an average value obtained by dividing the second target of the attributes to be analyzed within the region to be analyzed by the number of unit subregions present within the region to be analyzed.

5. The region analyzer according to claim 1 or 2, wherein the first target is a person.

6. The region analysis apparatus according to claim 1, further comprising a map output means for displaying the first target absorption value superimposed on a map of the region to be analyzed.

7. The region analysis apparatus according to claim 3, further comprising a map output means for displaying the target absorption force vector values ​​superimposed on a map of the region to be analyzed.

8. A storage device that stores predetermined demographic information, including the number of people located in each of a plurality of predetermined area meshes and information on the attributes of the people, in a series of first time units; The system includes a computing device that performs analysis of pedestrian flow information within the aforementioned analysis target area. The aforementioned computing device is An analysis condition setting means that accepts the designation of a predetermined analysis target area and analysis target attributes through user operation, and identifies a set of unit sub-regions that cover the said analysis target area without overlapping each other, A human behavior information estimation means that estimates the number of people with each category of human behavior and the target attribute located in the area mesh included in the analysis area at predetermined intervals, based on the demographic information and predetermined lifestyle information, The human flow information estimation means includes a means for estimating aggregated data for each unit sub-region within the analysis target region by distributing the number of people estimated by the human behavior information estimation means to the unit sub-regions within the region mesh, based on the estimation results of the human behavior information estimation means and the type of place and size of the place corresponding to the classification of human behavior for each region mesh, The human flow information estimation means estimates the distribution of people with the target attribute within the analysis area based on the aggregated data, and calculates a human absorption capacity value which is the ratio of the number of people with the target attribute for each unit sub-area to a predetermined value.

9. The human flow analyzer according to claim 8, wherein the predetermined value is an average value obtained by dividing the number of people with the attribute to be analyzed within the area to be analyzed by the number of unit sub-areas present within the area to be analyzed.

10. The human behavior information estimation means estimates the number of people for each category of human behavior located in the area mesh included in the analysis target area, based on the demographic information and predetermined lifestyle information, for each second time unit shorter than the first time unit. The human flow analysis apparatus according to claim 9, wherein the human flow information estimation means estimates the aggregated data for each estimation region mesh obtained by dividing the region mesh.

11. The aforementioned attributes to be analyzed include multiple personal attributes, The human flow analysis apparatus according to any one of claims 8 to 10, wherein the human flow information estimation means estimates a human absorption capacity value for each of the plurality of human attributes, which is the ratio of the number of people of the analysis target attribute for each unit sub-region to a predetermined value, and calculates a human absorption capacity vector value whose components are the human absorption capacity values ​​for each of the plurality of human attributes.

12. The human flow analysis apparatus according to claim 8, further comprising a map output means for superimposing the human absorption value onto a map of the area to be analyzed.

13. The human flow analysis apparatus according to claim 11, further comprising a map output means for displaying the human absorption force vector values ​​superimposed on a map of the area to be analyzed.

14. The analysis condition setting means accepts, by user operation, the specification of the analysis time period for the analysis target area and the specification of the analysis location within the analysis target area. The human flow analysis apparatus according to claim 11, wherein the human flow information estimation means estimates the change over time of the human absorption value and the human absorption vector at the analysis target location during the analysis time period.

15. The analysis condition setting means accepts, by user operation, the specification of the analysis time period and reference time period for the analysis target area, and the specification of the analysis handling location within the analysis target area. The human flow analysis apparatus according to claim 11, wherein the human flow information estimation means estimates the change over time of the human absorptive capacity value and the human absorptive capacity vector at the analysis target location during the analysis time period and the reference time period.

16. The aforementioned analysis time period is the time period during which a predetermined event is held at the aforementioned analysis location. The aforementioned reference time period is the time period during which the predetermined event is not being held at the location to be analyzed. The human flow analysis apparatus according to claim 15, wherein the human flow information estimation means calculates information indicating the human attraction force due to a predetermined event at the analysis target location by comparing the changes over time of the human absorptive force value and the human absorptive force vector during the analysis period and the reference period.

17. The human flow analysis apparatus according to claim 16, wherein the human flow information estimation means calculates information indicating the duration of the effect of human attraction due to a predetermined event at the analysis target location by comparing the changes over time of the human absorption value and the human absorption vector during the analysis period and the reference period.

18. The storage device stores the demographic information, the lifestyle information, the area location information, and the type of stay information. The aforementioned demographic information is aggregated demographic data for each first time interval, and is information on the number of people by attribute located in each area mesh within the predetermined area during the first time interval. The human flow analysis device according to claim 8, wherein the attributes of the person include information on the type of stay in which the person stays in the predetermined area.

19. The human flow analysis device according to claim 18, wherein the type of stay information includes attributes of residents to the predetermined area, attributes of workers to the predetermined area, and attributes of persons other than residents and workers.

20. The human flow information estimation means allocates the number of people based on the demographic information to each location within the predetermined area according to the attributes of the area location information corresponding to the type of stay information, wherein if the type of stay is a resident, the allocation is preferential to the attribute of residential; if the type of stay is a worker, the allocation is preferential to business facilities; and if the type is neither a resident nor a worker, the allocation is preferential to short-term stay locations and entertainment facilities.

21. A first data provision server provides predetermined demographic information, including the number of people located in each of a plurality of predetermined area meshes and information on the attributes of the said people, in a series of first time units. A second data provision server provides lifestyle information in predetermined time units, including information related to people's lives and behaviors, A third data provision server provides area location information including information on the type and size of each predetermined section located within a predetermined area, The system includes a human flow analysis device, and the human flow analysis device is A storage device that receives data from the first data provision server and stores predetermined demographic information, including the number of people located in each of a plurality of predetermined area meshes and information on the attributes of those people, in a series of first time units. The system includes a computing device that performs analysis of pedestrian flow information within the aforementioned analysis target area. The aforementioned computing device is An analysis condition setting means that accepts the specification of a predetermined analysis target area and analysis target attributes through user operation, and divides the analysis target area into predetermined unit sub-regions, A human behavior information estimation means that estimates the number of people with each category of human behavior and the target attribute located in the area mesh included in the analysis target area at predetermined intervals, based on the demographic information and lifestyle information from the second data provision server, The human flow information estimation means includes a means for estimating aggregated data for each unit sub-region within the analysis target region by distributing the number of people estimated by the human behavior information estimation means to the unit sub-regions within the region mesh based on the estimation results of the human behavior information estimation means and the region location information received from the third data provision server, and the type of location and size of the location corresponding to the classification of human behavior for each region mesh. A human flow analysis system comprising: a human flow information estimation means which estimates the distribution of people with the target attribute within the analysis area based on the aggregated data, and calculates a human absorption capacity value which is the ratio of the number of people with the target attribute for each unit sub-area to a predetermined value.

Images