Human movement prediction system, human movement prediction method, and human movement prediction program

Abstract

To provide a person movement prediction system, a person movement prediction method, and a person movement prediction program that predict congestion in a city, buildings, or on floors.SOLUTION: A person movement prediction system 100 includes: a device 400 of collecting area data that shows person movements in a predetermined city area; a device 300 of collecting building data that shows person movements in buildings or at building entrances / exits in a predetermined city; a device 200 of collecting floor data that shows person movements on predetermined floors in the buildings; and a prediction device 500 that predicts the person movements on the basis of the area data, the building data and the floor data.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a human movement prediction system, a human movement prediction method, and a human movement prediction program.

Background Art

[0002] For example, Patent Document 1 (Japanese Patent Application Laid-Open No. 2018-92445) discloses a prediction system and a method thereof capable of making the error of a predicted value as small as possible compared with the prior art.

[0003] The prediction system and method described in Patent Document 1 are a prediction system that calculates a predicted value related to a prediction target that adapts to a prediction during an intended period, and includes a storage device that records a plurality of data used to calculate the predicted value, and a control device that has a predetermined calculation model and applies the plurality of data to the calculation model to calculate the predicted value. The control device changes the calculation model using the data determined based on the information on the time attribute of each of the plurality of data.

[0004] Further, Patent Document 2 (Japanese Patent Application Laid-Open No. 2005-122438) discloses a prediction method, a prediction device, a prediction program, and a recording medium that improve prediction accuracy by processing a plurality of prediction data into high-precision prediction data.

[0005] The prediction method described in Patent Document 2 is a prediction method for calculating prediction data related to future prediction, and includes a prediction procedure for performing prediction by n different prediction methods to calculate n types of prediction data, and a prediction data processing procedure for calculating the average of the n types of prediction data as the prediction data.

[0006] Furthermore, Patent Document 3 (Japanese Patent Application Laid-Open No. 2015-46093) discloses a system, an apparatus, a method, a program, and a recording medium that records the program, which can predict the behavior of a subject earlier and more accurately than the prior art with a simple configuration.

[0007] The behavior prediction system described in Patent Document 3 is a behavior prediction system that predicts the behavior of a subject, and includes signal receiving means for receiving a signal that changes according to the behavior of the subject, and the magnitude of the signal received by the signal receiving means or the magnitude of the change in the signal exceeds a predetermined first threshold value. And behavior prediction means for notifying that the frequency has reached a predetermined second threshold value or more.

[0008] Patent Document 4 (Japanese Patent Application Laid-Open No. 2013-186556) discloses a behavior prediction device that can predict a person's behavior by reflecting individual characteristics.

[0009] The behavior prediction device described in Patent Document 4 inputs video data in which an object is photographed, identifies the person, extracts identification information representing the characteristics of the person, analyzes the behavior and state of the person from the video data, and assigns the identifier assigned to the person. A video analysis unit that outputs the behavior and state as analysis information in association with each other, a person information storage unit that stores the person information, a condition information indicating the situation in which the person is photographed, and an information collection unit that collects the analysis information, and the condition information and the analysis information are associated with each other. An observation information storage unit for storing, a rule generation unit for creating rules regarding the behavior of a person based on the condition information and the analysis information, a rule storage unit for storing the rules, and the condition information and the analysis information collected by the information collection unit for a specific person and the rules created in advance for the person. And a rule collation unit that collates the rules and generates behavior prediction information using the rules based on the collation result.

[0010] Patent Document 5 (Japanese Patent Application Laid-Open No. 2009-151359) discloses a moving object tracking device, a moving object tracking method, a moving object tracking program, and a recording medium on which the moving object tracking program is recorded, which predict the movement of a person in an actual environment and suppress the number of repeated processing times required for the prediction result to stabilize. And achieve both high tracking accuracy and low calculation cost. Disclosed are a moving object tracking device, a moving object tracking method, a moving object tracking program, and a recording medium on which the moving object tracking program is recorded.

[0011] The moving object tracking device described in Patent Document 5 is a device that tracks the state of a moving object, such as its three-dimensional position and size, using an image input device such as one or more cameras. It includes means for acquiring an image captured by the imaging means at each time, means for creating a silhouette image by extracting only the region in the acquired input image where the object to be tracked appears, a three-dimensional environment information database that stores in advance the three-dimensional structure of the real world, the internal and external parameters of the imaging means arranged in the real world, and the action history distribution of the target object in the environment, means for estimating the probability distribution of the target state at each time using the information in the three-dimensional environment information database and the silhouette image, target state distribution storage means for storing the probability distribution of the target state estimated by the estimation means at each time and the target state having the maximum probability, means for calculating the target state having the maximum probability at each time from the probability distribution of the target state stored in the target state distribution storage means, means for calculating the action history distribution from the action history distribution stored in the three-dimensional environment information database based on the target state calculated by the target state calculation means, and updating the target state distribution of the previous time stored in the target state distribution storage means to the target state distribution of the current time and updating the target state having the maximum probability calculated by the target state calculation means, and updating the action history distribution stored in the three-dimensional environment information database using the target state calculated by the action history distribution calculation means.

Prior Art Documents

Patent Documents

[0012]

Patent Document 1

Patent Document 2

Patent Document 3

Patent Document 4

Patent Document 5

Summary of the Invention

Problems to be Solved by the Invention

[0013] The prediction devices described in Patent Documents 1 to 3 are basically for performing prediction calculations. In the behavior prediction devices of Patent Documents 4 and 5, they only detect and predict personal movements. However, it has been difficult to realize behavior prediction in large cities. In particular, it has not been possible to solve the problem of what should be done to eliminate the congestion level in the city. Furthermore, along with the problems in the city, it has not been possible to solve the problems of congestion inside buildings and congestion on floors.

[0014] The main object of the present invention is to provide a human movement prediction system, a human movement prediction method, and a human movement prediction program that perform congestion prediction in cities, buildings, and floors. Another object of the present invention is to provide a human movement prediction system, a human movement prediction method, and a human movement prediction program that perform congestion prediction in cities, buildings, and floors and notify of improvements.

[0015] (1) A human movement prediction system according to one aspect includes an area data collection device that shows the movement of people in a predetermined urban area, a building data collection device that shows the movement of people at the entrance and exit of a building or building in a predetermined city, a floor data collection device that shows the movement of people on a predetermined floor in the building, and a prediction device that visualizes the movement of people or predicts the number of staying people from the area data, building data, and floor data.

[0016] In this case, the movement of people can be visualized or the number of staying people can be predicted from the area data, building data, and floor data. As a result, congestion information in the city, building, and floor can be appropriately presented, and congestion avoidance can be performed. Note that the building data collection device may be a security gate in the building. If there is no security gate, the number of people getting on and off the elevator or imaging devices such as cameras at the entrance and exit may be used instead. Note that it is desirable to predict the floor data in real time, and the movement of people or the number of people staying may be predicted from the area data and the building data.

[0017] (2) The human movement prediction system according to the second invention is a human movement prediction system according to one aspect. The prediction device may predict the area data and building data of the day by multiple regression analysis using one or more pieces of past data recorded in the area data collection device and the building data collection device.

[0018] In this case, the prediction device can estimate the congestion and contact frequency of people using one or more pieces of past data.

[0019] (3) The human movement prediction system according to the third invention is the human movement prediction system according to the second invention. When the area data Z1 of the day is the data at the same time of the previous day as X1 and the data at the same time of one week ago as Y1, and a1, b1, c1 are predetermined constants, it is predicted by the following mathematical formula 1. [Formula 1] Z1 = a1X1 + b1Y1 + c1

[0020] In this case, prediction data can be output in the prediction system. In particular, the average error can be kept within about 10% or less.

[0021] (4) The human movement prediction system according to the fourth invention is the human movement prediction system according to the second invention. When the building data Z2 of the day is the data at the same time of the previous day as X2 and the data at the same time of one week ago as Y2, and a2, b2, c2 are predetermined constants, it is predicted by the following mathematical formula 2. [Formula 2] Z2 = a2X2 + b2Y2 + c2

[0022] In this case, predictive data can be output from the prediction system in question. In particular, it can be kept within an average error of about 10% or less.

[0023] (5) The human movement prediction system according to the fifth invention, in the human movement prediction system that follows a scenario, the floor data collection device may include a mobile terminal of a person, a USB beacon communication unit that can communicate with the mobile terminal and is provided in plurality on the ceiling, floor, tabletop or wall surface of the floor, and a control unit that determines the relationship between the plurality of USB beacon communication units and the mobile terminal.

[0024] In this case, it is possible to grasp the real-time congestion of human movement and the contact history between people.

[0025] (6) The human movement prediction system according to the sixth invention, in the human movement prediction system that follows a scenario, the prediction device may include a recording device, and may estimate prediction data from the data recorded in the recording device.

[0026] In this case, since the prediction device using floor data, building data, and area data can estimate prediction data regarding human congestion and contact from the data recorded in the recording device, it is possible to make suggestions to reduce human congestion and contact.

[0027] (7) The human movement prediction system according to the seventh invention, in the human movement prediction system related to a scenario, the prediction device further includes a notification device that notifies of congestion avoidance in a predetermined urban area and a building or a building, and the notification device may perform at least any one of guideline notification of congestion avoidance, notification of formulating company attendance rules, and congestion notification to transportation facilities. Note that the notification device includes real-time visualization and dashboarding of past data.

[0028] In this case, the notification device can appropriately perform various types of notifications. Particularly when congestion is predicted, it can call on people to avoid such congestion. As a result, it becomes possible to take measures against diseases such as infectious diseases.

[0029] (8) A method for predicting the movement of people following other aspects includes an area data collection process for indicating the movement of people in a predetermined urban area, a building data collection process for indicating the movement of people at the entrances and exits of buildings or structures in a predetermined city, a floor data collection process for indicating the movement of people on a predetermined floor in a building, and a prediction process for visualizing the movement of people or predicting the number of staying people from the area data, building data, and floor data.

[0030] In this case, it is possible to visualize the movement of people or predict the number of staying people from the area data, building data, and floor data. As a result, it is possible to appropriately present congestion information within the city, building, and floor, and congestion avoidance can be carried out.

[0031] (9) A program for predicting the movement of people following still other aspects includes an area data collection process for indicating the movement of people in a predetermined urban area, a building data collection process for indicating the movement of people at the entrances and exits of buildings or structures in a predetermined city, a floor data collection process for indicating the movement of people on a predetermined floor in a building, and a prediction process for visualizing the movement of people or predicting the number of staying people from the area data, building data, and floor data.

[0032] In this case, it is possible to visualize the movement of people or predict the number of staying people from the area data, building data, and floor data. As a result, it is possible to appropriately present congestion information within the city, building, and floor, and congestion avoidance can be carried out.

Brief Description of the Drawings

[0033]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Embodiments for Carrying Out the Invention

[0034] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following description, the same parts are denoted by the same reference numerals. Their names and functions are also the same. Therefore, detailed descriptions thereof will not be repeated.

[0035] (Embodiment) FIG. 1 is a schematic diagram showing an example of a human movement prediction system 100 according to an embodiment.

[0036] (Prediction System 100) As shown in FIG. 1, a human movement prediction system (hereinafter referred to as a prediction system) 100 includes a floor data collection device 200, a building data collection device 300, an area data collection device 400, and a prediction device 500. Details of each of the above configurations will be described later. Also, as shown in FIG. 1, the prediction device 500 includes a recording device 510, a notification device 520, and a display device 530.

[0037] (Floor Data Collection Device 200) Next, FIG. 2 is a schematic diagram showing an example of the floor data collection device 200, and FIG. 3 is a schematic diagram for explaining an example of the communication state between the USB beacon communication unit 250 and the mobile terminal 230.

[0038] As shown in FIG. 2, the floor data collection device 200 includes a recording unit 210, a control unit 220, a mobile terminal 230, a plurality of USB beacon communication units 250, and a display unit 290. The control unit 220 includes a processing unit 221 inside. The recording unit 210 may record and hold the performance information 201 of each employee and the movement history information 202 of each employee.

[0039] As shown in FIG. 2, in the present embodiment, the control unit 220 can communicate with the mobile terminal 230. Further, the mobile terminal 230 can communicate with a plurality of USB beacon communication units 250. Note that the control unit 220 may be able to communicate with the plurality of USB beacon communication units 250. In this case, the control unit 220 causes the mobile terminal 230 to read the unique information of the plurality of USB beacon communication units 250 and recognize the position.

[0040] Next, as shown in FIG. 3, the plurality of USB beacon communication units 250 are disposed in the ceiling space. The unique ID (number, etc.) of the installed USB beacon communication unit 250 is recorded in the recording unit 210. Therefore, the control unit 220 can recognize which USB beacon communication unit 250 is communicating with the mobile terminal 230 from the information in the recording unit 210.

[0041] Specifically, the USB beacon communication unit 250 is powered by USB and communicates with the mobile terminal 230 via Bluetooth (registered trademark). As a result, it is possible to recognize near which of the plurality of USB beacon communication units 250 the mobile terminal 230 is present. On the other hand, the control unit 220 may return the position information obtained based on the radio field intensity information between the USB beacon communication unit 250 and the mobile terminal 230 to the mobile terminal 230. The control unit 220 transmits the acquired position information of the mobile terminal 230 to the prediction device 500 as floor data.

[0042] (Building Data Collection Device 300) FIG. 4 is a schematic diagram showing an example of the building data collection device 300. As shown in FIG. 4, the building data collection device 300 includes a security gate 310 of the building and a control unit 320. The control unit 320 transmits, as building data, the number of people per time unit, for example, per minute, who have passed through the security gate 310 of the building, to the prediction device 500.

[0043] (Area Data Collection Device 40) FIG. 5 is a schematic diagram showing an example of the area data collection device 400.

[0044] The area data collection device 400 includes an area mesh detection device 410 and a control unit 420. The area mesh detection device 410 meshes (divides) a predetermined area and detects the number of mobile terminals 430 present in the division. In this case, a predetermined application software is downloaded to the mobile terminal 430, and the number can be detected based on the GPS signal. The control unit 420 transmits the number of mobile terminals 430 in the area to the prediction device 500. Although not shown, the area data collection device 400 may acquire weather information, precipitation, temperature, and solar radiation amount from the website of the Japan Meteorological Agency.

[0045] (Operation of the prediction device 500) The prediction device 500 shown in FIG. 1 performs congestion prediction based on the floor data from the floor data collection device 200, the building data from the building data collection device 300, and the area data from the area data collection device 400. The prediction device 500 predicts the congestion information in the area data using the following formula 1. [Formula 1] Z1 = a1X1 + b1Y1 + c1

[0046] Here, Z1 represents the predicted area data, X1 represents the area data at the same time of the previous day, and Y1 represents the area data at the same time of one week ago. a1, b1, and c1 are predetermined constants.

[0047] In the present embodiment, the prediction device 500 of the prediction system 100 sets the constant a1 to be 0.1 or more and 3.0 or less, the constant b1 to be 0.3 or more and 5.0 or less, and the constant c1 to be in the range of 1000 or more and 2000 or less. Note that the constants a1, b1, and c1 are not limited to the above numerical values and are calculated according to the model. That is, it is desirable to arbitrarily determine the constants from the multiple regression analysis according to the model and perform the prediction.

[0048] Further, the prediction device 500 predicts congestion information in building data using the following formula 2. [Formula 2] Z2 = a2X2 + b2Y2 + c2

[0049] Here, Z2 represents predicted building data, X2 represents building data at the same time on the previous day, and Y2 represents building data at the same time one week ago. a2, b2, and c2 are predetermined constants.

[0050] In the present embodiment, the prediction device 500 of the prediction system 100 sets the constant a2 to be 0.3 or more and 5.0 or less, the constant b2 to be 0.1 or more and 0.3 or less, and the constant c2 to be in the range of 3 or more and 10 or less. Note that the constants a2, b2, and c2 are not limited to the above numerical values and are calculated according to the model. That is, it is desirable to arbitrarily determine the constants from multiple regression analysis according to the model and perform prediction.

[0051] The prediction device 500 of the prediction system 100 performs the above prediction, and in a situation where congestion information may occur, it can notify the administration or government, enterprises, real estate developers, restaurants and commercial tenants, public transportation and air / ship / railway companies, or residents and travelers of the prediction of the occurrence of congestion from the notification device 520 of the prediction system 100. Furthermore, the display device 530 can visualize the congestion situation.

[0052] For example, the administration or government can use it for guidelines or policy decisions to avoid the three Cs, and enterprises can use it for office layout or attendance rules, etc. Furthermore, real estate developers can use it for building operation or population control, restaurants and commercial tenants can use it for operations to avoid the three Cs, public transportation and air / ship / railway companies can provide real-time congestion predictions, and residents and travelers can shift their travel times.

[0053] The verification of the prediction device 500 of the prediction system 100 will be described below. First, the floor data, building data, and area data will be described.

[0054] (Floor data) FIG. 6 is a schematic diagram showing an example of installing the floor data collection device 200 on a predetermined floor of a building. As shown in FIG. 6, the floor data collection device 200 is arranged at the floor data collection position. Hereinafter, a company occupies a predetermined floor of the building.

[0055] FIG. 7 is a diagram showing an example of the floor data of the company. The vertical axis represents the total congestion level (congestion occurrence frequency), and the horizontal axis represents the date and time.

[0056] In FIG. 7, a state where the ratio of the number of staying employees determined based on the area and the number of seats for a predetermined position exceeds 50% is defined as congestion. Then, the occurrence frequency of congestion at each predetermined position is shown in a graph every 5 minutes.

[0057] As shown in FIG. 7, since congestion in the company occurs randomly, monitoring using the real-time floor data collection device 200 is essential.

[0058] FIGS. 8 and 9 are diagrams showing other examples of the floor data transmitted from the floor data collection device 200 in the company. FIG. 8 is a schematic diagram showing the congestion occurrence positions in the company. The arrowed square part (position) indicates that congestion is likely to occur.

[0059] Next, FIG. 9 is the floor data of the company. As shown in FIG. 9, the vertical axis represents the capacity at each position (point), and the horizontal axis represents the date. Also, the congestion occurrence frequency is indicated by shading. As a result, it is possible to identify where congestion is likely to occur.

[0060] The floor data collection device 200 sends the floor data in FIGS. 7 to 9 and / or the floor data to the prediction device 500.

[0061] (Building data example) FIG. 10 is a diagram showing an example of building data of a building in which a company is located.

[0062] The vertical axis of FIG. 10 indicates the number of people in the building, and the horizontal axis indicates time. As shown in FIG. 10, in the building data of the company, the number of people in the building increases from 11:00 and from 14:00 to 16:00 on weekdays. Also, on Saturdays (Sat) and Sundays (Sun), there is no increase in the number of people in the building and it is constant. The building data collection device 300 sends the building data of FIG. 10 to the prediction device 500.

[0063] (Area data example) FIG. 11 is a schematic diagram showing an example of a mesh in which the area data collection device 400 acquires area data.

[0064] FIG. 11 shows an example of a mesh in the surrounding area of the company. In FIG. 11, meshes are set every 62.5 m in an area with a radius of 500 m centered on the company. Note that the numerical setting of the meshes is arbitrary, and other arbitrary numerical settings such as every 1 m, every 5 m, every 10 m, every 50 m, etc. may be provided.

[0065] FIG. 12 is a diagram showing an example of area data

[0066] In FIG. 12, the vertical axis indicates the number of people staying in the area (congestion level), and the horizontal axis indicates time. The number of people staying peaks at 8:00, 12:00, and 18:00 on weekdays. The area data collection device 400 sends the area data of FIG. 12 to the prediction device 500.

[0067] (Prediction processing of the prediction device 500 of the prediction system 100) The prediction device 500 records the floor data, building data, and area data sent from the floor data collection device 200, the building data collection device 300, and the area data collection device 400 in the recording device 510. FIG. 13 is a diagram for explaining a model implemented by the prediction device 500 of the prediction system 100. As shown in FIG. 13, the prediction device 500 of the prediction system 100 performs modeling based on area data, building data, and floor data, and constructs models of the area data and the building data. Furthermore, the prediction device 500 of the prediction system 100 may also notify the floor responsible person about the congestion possibility in the floor data from past data using the notification device 520 and the display device 530.

[0068] (Correlation of Floor Data, Building Data, and Area Data) Next, the correlations of the floor data, the building data, and the area data will be described. FIG. 14 is a diagram showing the daily congestion levels of the floor data, the building data, and the area data, and FIG. 15 is a diagram showing the congestion levels of the floor data, the building data, and the area data by time period.

[0069] As shown in FIG. 14, the daily congestion levels show that the building data and the area data are somewhat similar, and as shown in FIG. 15, the congestion levels by time show that the floor data and the building data are similar.

[0070] FIGS. 16 and 17 are diagrams showing an example of the conversion process of the prediction device 500.

[0071] As shown in FIG. 16(a), the prediction device 500 calculates the moving average of the staying population over time from the area data and the building data. Here, the prediction device 500 normalizes the values of the area data and the building data. Next, as shown in FIG. 16(b), a two-stage conversion process is performed. Here, the prediction device 500 first moves the data before 11 o'clock in the building data two hours earlier and moves the data after 15 o'clock two hours later. Then, next, the prediction device 500 performs an inversion process on the building data at the position with a value of 0.80 for the data from 9 o'clock to 17 o'clock.

[0072] Next, as shown in FIG. 17, the prediction device 500 calculates the average congestion time transition of building data and floor data on weekdays. Here, the prediction device 500 divides the values of the area data and the building data by the maximum value and normalizes them. The time transition regarding the congestion of the building data and the floor data shows the same tendency, but the coefficient of variation shows 6.2% for the building data, while the floor data shows 48.0%. From this, it was found that the floor data needs to be estimated from real-time data. That is, when showing the area with 95% reliability as a band, the width becomes large. That is, it was found that the floor data has large fluctuations.

[0073] Finally, the prediction device 500 performs a prediction regarding the area data. The prediction device 500 creates a future congestion model based on the past congestion situation, using the day of the week and the time as parameters. FIG. 18 is a diagram in which the prediction device 500 performs a prediction regarding the area data and the building data and conducts verification.

[0074] In the present embodiment, as described above, the prediction device 500 created a model using multiple regression analysis. Note that it is not limited to multiple regression analysis, and other arbitrary SARIMA (Seasonal Auto Regressive Integrated Moving Average) models, Dynamic Linear Model (DLM), or RNN (Recurrent Neural network) models may be used.

[0075] The prediction device 500 calculates the predicted area data using the above formula 1 and compares the root mean square error (RMSE) values of each model. As a result, when the prediction device 500 calculated the predicted area data from the area data at the same time of the previous week and the area data at the same time of the previous day, it was able to achieve a prediction with high accuracy within an average error range of about 3%. In addition, in the rectangular frame in FIG. 18(a), some irregularities occurred.

[0076] Further, as shown in Fig. 18(b), the prediction device 500 calculated prediction building data using the above formula (2) and compared the root mean square error (RMSE) values of each model. As a result, when the prediction device 500 calculated the prediction building data from the building data at the same time of the previous week and the building data at the same time of the previous day, it was able to achieve a prediction with high accuracy within an average error range of about 5%. Note that in the rectangular frame in Fig. 18(b), some irregularities occurred.

[0077] In the present embodiment, the prediction system 100 corresponds to the "prediction system", the operation of the prediction system 100 corresponds to the "prediction method", and the program of the prediction system 100 corresponds to the "prediction program". The area data collection device 400 corresponds to the "area data collection device", the operation of the area data collection device 400 corresponds to the "area data collection process", and the program of the area data collection device 400 corresponds to the "area data collection process". The building data collection device 300 corresponds to the "building data collection device", the operation of the building data collection device 300 corresponds to the "building data collection process", and the program of the building data collection device 300 corresponds to the "building data collection process". The floor data collection device 200 corresponds to the "floor data collection device", the operation of the floor data collection device 200 corresponds to the "floor data collection process", and the program of the floor data collection device 200 corresponds to the "floor data collection process". The prediction device 500 corresponds to the "prediction device", the recording device 510 corresponds to the "recording device", the notification device 520 corresponds to the "notification device", the mobile terminal 230 corresponds to the "mobile terminal", the USB beacon communication unit 250 corresponds to the "USB beacon communication unit", and the control unit 220 corresponds to the "control unit".

[0078] While a preferred embodiment of the present invention is as described above, the present invention is not limited thereto. It will be understood that various embodiments may be made without departing from the spirit and scope of the present invention. Further, in this embodiment, the actions and effects according to the configuration of the present invention are described, but these actions and effects are merely examples and do not limit the present invention.

Explanation of Reference Numerals

[0079] 100 Prediction system 200 Floor data collection device 230 Portable terminal 220 Control unit 250 USB beacon communication unit 300 Building data collection device 400 Area data collection device 500 Prediction device 510 Recording device 520 Notification device

Claims

1. An area data collection device for indicating the movement of people in a predetermined urban area, A building data collection device for indicating the movement of people at the entrances and exits of buildings or structures in the predetermined city, A floor data collection device for indicating the movement of people on a predetermined floor in the building, A prediction device for visualizing the movement of the people or predicting the number of staying people from the area data, the building data, and the floor data, The prediction device predicts the time transition of the number of staying people in the urban area using the data of the time transition of the number of staying people in the building, The prediction of the time transition of the number of staying people in the urban area, The prediction of the movement of people is performed based on data obtained by shifting the data before 11:00 among the time transitions of the number of staying people in the building two hours earlier, shifting the data after 15:00 two hours later, and further inverting the data of the time transitions of the number of staying people in the building from 9:00 to 17:00 symmetrically with respect to a straight line that is 0.8 times the peak value. A prediction system for the movement of people.

2. The prediction device predicts the area data and building data of the current day by multiple regression analysis using one or more pieces of past data recorded in the area data collection device and the building data collection device. The prediction system for the movement of people according to Claim 1.

3. The area data Z1 of the current day is predicted by the following mathematical formula 1, where the data at the same time of the previous day is X1, the data at the same time of one week ago is Y1, and a1, b1, and c1 are predetermined constants. The prediction system for the movement of people according to Claim 2. [Formula 1] Z1 = a1X1 + b1Y1 + c1

4. The building data Z2 of the current day is predicted by the following mathematical formula 2, where the data at the same time of the previous day is X2, the data at the same time of one week ago is Y2, and a2, b2, and c2 are predetermined constants. The prediction system for the movement of people according to Claim 2. [Formula 2] Z2 = a2X2 + b2Y2 + c2

5. The floor data collection device The mobile terminal of the person, A USB beacon communication unit that is communicable with the mobile terminal and is provided in plurality on the ceiling or floor of the floor, A control unit for determining the relationship between the plurality of USB beacon communication units and the mobile terminal. The prediction system for the movement of people according to Claim 1.

6. The prediction device includes a recording device, The prediction system for the movement of people according to Claim 1, which estimates prediction data from the data recorded in the recording device.

7. The prediction device further includes a notification device that notifies of congestion avoidance in the predetermined urban area and the building or the building. The notification device performs at least one of the following: notification of congestion avoidance guidelines, notification of formulation of company commuting rules, and notification of congestion to transportation facilities. The human movement prediction system according to claim 1.

8. A method for predicting human movement executed by a human movement prediction system including an area data collection device, a building data collection device, a floor data collection device, and a prediction device, An area data collection step of indicating the movement of people in a predetermined urban area, A building data collection step of indicating the movement of people at the entrance and exit of a building or a building in the predetermined city, A floor data collection step of indicating the movement of people on a predetermined floor in the building, A prediction step of visualizing the movement of the people or predicting the number of staying people from the area data, the building data, and the floor data, The prediction step predicts the time transition of the number of staying people in the urban area using the data of the time transition of the number of staying people in the building. The prediction of the time transition of the number of staying people in the urban area is as follows. The data before 11:00 in the time transition of the number of staying people in the building is shifted 2 hours forward, the data after 15:00 is shifted 2 hours backward, and further, the data from 9:00 to 17:00 in the time transition of the number of staying people in the building is based on the data that is line-symmetrically inverted with a straight line that is 0.8 times the peak value. A method for predicting human movement performed.

9. A human movement prediction program for causing a computer of a human movement prediction system including an area data collection device, a building data collection device, a floor data collection device, and a prediction device to execute the following processing: An area data collection process for indicating the movement of people in a predetermined urban area, A building data collection process for indicating the movement of people at the entrance and exit of a building or a building in the predetermined city, A floor data collection process for indicating the movement of people on a predetermined floor in the building, A prediction process for visualizing the movement of the people or predicting the number of staying people from the area data, the building data, and the floor data, The prediction process predicts the time transition of the number of staying people in the urban area using the data of the time transition of the number of staying people in the building. The prediction of the time transition of the number of staying people in the urban area is as follows. A human movement prediction program that shifts the data before 11:00 in the time series of the number of people staying in the building by 2 hours forward, shifts the data after 15:00 by 2 hours backward, and is based on data obtained by inverting the data of the time series of the number of people staying in the building from 9:00 to 17:00 symmetrically with respect to a straight line that is 0.8 times the peak value.

Images