JP3822523B2 - Distribution prediction apparatus and method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a distribution prediction apparatus and method.
[0002]
[Prior art]
The distribution prediction apparatus 100 currently used will be described with reference to FIG.
[0003]
Statistical data 104 of the temporal transition of the human distribution in the past is accumulated, the human distribution 105 is predicted based on the statistical data 104 and future event data 107, and advertisements and advertisements are based on the prediction results. An example 106 is used to reduce congestion.
[0004]
Specifically, considering the example of the event data 107 of the theme park, by accumulating the statistical data 104 of the time transition of the number of visitors, the distribution 105 of the congestion by day or hour is predicted. Thus, the advertisement / advertisement 106 is efficiently performed based on the prediction result, and it is useful for alleviating the congestion in the facility.
[0005]
However, although the above-described distribution prediction apparatus 100 is expected to have a certain effect as an advertising / advertisement activity or congestion mitigation measure, the accuracy of the prediction is reduced because the target distribution is limited, and the application of the prediction There is a problem that the range is narrow and the effect is limited.
[0006]
In order to solve the above problems, as shown in FIG. 6, based on the statistical data of the temporal distribution of the past distribution accumulated and the event data of the future from a plurality of measured distribution situations. A distribution prediction device 8 that predicts a human distribution has been proposed (Japanese Patent Application No. 2001-314548).
[0007]
This distribution prediction device measures data 1 to 3 of the distribution of people, and based on the accumulated statistical data 4 of the temporal distribution of the past and the event data 7 of the future, the computer 5 Predict. By effectively using the prediction results, the predetermined advertisement device 6 can be used for improving the efficiency of advertisement / advertisement activities and reducing congestion.
[0008]
[Problems to be solved by the invention]
As a specific example of the distribution prediction device 8 described above, when the congestion situation of the current transportation system and the number of visitors in a certain facility are known, the transportation system at the end of the facility is based on the past statistical data. Predict the congestion situation. Based on the prediction result, it is possible to effectively carry out the advertisement / advertisement by performing the advertisement / advertisement in accordance with the preference of the facility user to the transportation facility where congestion is expected.
[0009]
However, the above-described distribution prediction apparatus 8 has a problem that a specific prediction method is not clear.
[0010]
Therefore, the present invention is for solving the above-described problems, and provides a specific prediction method in a distribution prediction apparatus.
[0011]
[Means for Solving the Problems]
According to the first aspect of the present invention, a distribution measuring means for measuring the current state of human distribution, a data storage means for accumulating the past human distribution state, an event name to be held in the future, its location, and its date and time Event data storage means for accumulating future event data that is data including the current person distribution state, the past person distribution state, and the future event data as inputs, and a future event in a specific event A distribution prediction device having a distribution prediction means for predicting a person's distribution status, wherein the distribution prediction means has a distribution prediction circuit that models the flow of the person by circuit theory; The distribution of people in the past is substituted into the distribution prediction circuit as a voltage value, current value, or value of a circuit element constituting the circuit corresponding to the distribution state of the person and the event data. A distribution predicting apparatus which is characterized in that.
[0019]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
[0020]
(First embodiment)
First, before describing the contents of the present embodiment, the basic idea of the present invention will be described.
[0021]
When a person moves from the start point to the end point, the person needs a motive to move. For example, when a popular theme park opens, many people want to go to the theme park, and the more attractive the theme park is, the more people gather. In other words, many people move as the theme park opens. This is similar to a state in which a large amount of current flows when the potential difference between terminals is increased in an electric circuit.
[0022]
Therefore, paying attention to this point, the movement between human subjects is applied to the movement between the terminals of the electric circuit, and the road such as the land route, the sea route, the air route where the person moves is applied to the circuit wiring of the electric circuit. When a human distribution is predicted, the human flow is reproduced by a distribution prediction circuit modeled by circuit theory.
[0023]
FIG. 1 is a diagram showing a distribution prediction circuit 10 according to the first embodiment of the present invention.
[0024]
The distribution prediction circuit 10 is a circuit that predicts a human flow between the event A and the event B.
[0025]
The left side of the distribution prediction circuit 10 is an event A circuit 25, and the right side is an event B circuit 26. The event A circuit 25 and the event B circuit 26 are only resistive elements having Rab resistance values. Are connected by an admittance element 19 comprising:
[0026]
In the event A circuit 25, the potential of each terminal is determined by the set of the narrator 13 and the DC voltage source 14, and a pair of the norator 12 and the capacitor 11 is connected to avoid the burning of the current (corresponding to the disappearance of the person). The narrator 13 and the DC voltage source 14 are connected in parallel. In this way, all charges in the circuit are preserved. In addition, when a person moves, a route is determined, and the movement state is limited depending on the type of the route. This is similar to a state in which the circuit (path) is determined and the current flowing is limited by the impedance between the terminals.
[0027]
The narrator 13 and the norator 12 are ideal operational amplifiers. The voltage between the terminals of the narrator 13 is zero, and the flowing current is zero. The voltage between the terminals of the norator 12 and the flowing current are arbitrary.
[0028]
The event B circuit 26 has a similar configuration.
[0029]
Now, the current number of people in event A is Va, the current number of people in event B is Vb, and the admittance of an admittance element (that is, resistance element) 19 between events AB calculated from statistical data of the temporal distribution of the distribution in the past. Is Yab (= 1 / Rab), A (t) is the time change of the predicted increase rate of the event A, and B (t) is the change of the predicted increase rate of the event B.
[0030]
When the voltage value of the DC voltage source 14 at event A is Va × A (t) and the voltage value of the DC voltage source 18 at event B is Vb × B (t), the switches 16 and 17 are closed. A current Iab (t) flows through the admittance element 19.
[0031]
By calculating the current Iab (t) flowing through the admittance element 19, it is possible to predict a human flow between the events AB.
[0032]
When the event A ends, the switch 16 is opened, and the current Iab (t) flowing through the admittance element 19 is calculated to predict the flow of people between the events AB.
[0033]
Similarly, when the event B ends, the flow of people between the events AB is predicted by opening the switch 17.
[0034]
FIG. 2 is a graph showing a specific example showing the distribution of people in the event.
[0035]
Assume that event A is an opening of a certain theme park, and event B is an adjacent area of a theme park, and no event occurs in that area.
[0036]
Currently, there are 0 people in the vicinity of the theme park before the opening of Event A, and if the time change of the predicted increase in the number of people in Event A is A (t), Event B does not occur, so switch 17 is opened. It shall be.
[0037]
At this time, the current Iab (t) flowing through the admittance element 19 is A (t) / Rab, which corresponds to the event start graph of FIG.
[0038]
By opening the switch 16 at the end of the event A, a current Iab (t) corresponding to the event end graph of FIG. 2 flows through the admittance element 19.
[0039]
As a case where the flow of the person can be reproduced from the admittance element 19 of the resistance element in this way, for example, the traffic volume on the road can be considered when the flow of the person is constant. Since the maximum traffic volume is determined, the flow of people is almost constant during congestion, so it can be reproduced with a resistive element.
[0040]
(Second Embodiment)
FIG. 3 is a diagram showing a second embodiment of the distribution prediction circuit 10 according to the present invention, in which the admittance element 19 includes a coil element, and the other circuit configuration is the distribution prediction circuit 10 of the first embodiment. It is the same.
[0041]
A specific example of the distribution of people is shown in the graph of FIG.
[0042]
Assume that event A is an opening of a certain theme park, and event B is an adjacent area of a theme park, and no event occurs in that area.
[0043]
Currently, there are 0 people around the theme park before the opening of the event A, the time change of the estimated increase rate of the event A is A (t), the admittance element 19 is 1 / jωLab, and the event does not occur in the event B Therefore, the switch 17 is opened.
[0044]
At this time, the current flowing through the admittance element 19 corresponds to the graph of FIG. By opening the switch 17 at the end of the event A, a current Iab (t) corresponding to the event end graph in FIG. 4 flows through the admittance element 19.
[0045]
As a case where the flow of the person can be reproduced from the admittance element 19 of the coil element in this way, when the increase / decrease in the number of events is reflected in the flow of the person as it is, for example, the traffic volume by train or the like can be considered. Assuming that there is no waiting time for trains, the increase or decrease in the number of people between events is reflected in the flow of people as it is.
[0046]
Note that the admittance configuration and the expected rate of increase in the number of events are predicted from statistical data of past events, and the admittance is modeled using a plurality of various circuit elements. This will be described later.
[0047]
(Third embodiment)
FIG. 5 is a diagram showing a third embodiment of the distribution prediction circuit 10 according to the present invention.
[0048]
This is a case where two or more events exist, and an event C is added as compared to FIG.
[0049]
The event C circuit 27 has the same configuration as the event A circuit 25. The current number of people in the event C is Vc, the time change in the predicted number of people increase is C (t), and the admittance of the admittance 24 between the events AC calculated from the statistical data of the temporal transition of the distribution in the past is Yac. To do. In this figure, the admittance between BCs is considered as 0, and wiring between BCs is not performed.
[0050]
The DC voltage source 14 of event A is a voltage value Va × A (t), the voltage value Vb × B (t) of the DC voltage source 18 of event B, and the DC voltage source 23 of event C is a voltage value Vc × C (t). Then, the currents Iab (t) and Iac (t) flow through the admittance elements 19 and 24 by closing the switches 16, 17 and 21.
[0051]
By calculating the currents Iab (t) and Iac (t) flowing through the admittance element 19 and the admittance 24, it is possible to predict a human flow between the events AB and AC.
[0052]
When the event A is completed, the switch 16 is opened, and the currents Iab (t) and Iac (t) flowing through the admittance elements 19 and 24 are calculated, thereby predicting the human flow between the events AB and AC. .
[0053]
Similarly, when the event B and the event C are completed, the flow of people between the events AB and AC is predicted by opening the switches 17 and 21.
[0054]
Similarly, by combining a plurality of events, it is possible to predict the flow of people related to the plurality of events.
[0055]
(Application form)
Next, an application mode of the prediction distribution device 8 incorporating the distribution prediction circuit 10 described above will be described.
[0056]
(1) Configuration of Distribution Prediction Device 8 FIG. 6 is a diagram showing the distribution prediction device 8.
[0057]
Current person distribution status data 1 to 3 and past person distribution statistics and statistical data 4 which is a past person distribution situation consisting of temporal distribution patterns of the distribution number of people are recorded in a computer hard disk or memory. Based on the accumulated past statistical data 4 and the future event data 7 to be held in the future stored in the recording medium, the distribution prediction unit 5 comprising a computer is used to determine the human distribution. An example is shown in which prediction and advertising activities are effectively carried out by the advertisement device 6 such as a speaker or an image display device based on the prediction result. The distribution prediction circuit 10 is built in the distribution prediction unit 5 made of a computer.
[0058]
(2) Contents of Future Event Data 7 As future event data 7, for example, it is assumed that “Toshiba Theme Park” opens at 9:00 on May 1st.
[0059]
And in this event data 7, the time change of the person in the Toshiba theme park in the past (for example, last year or two years ago) is accumulated as the time change pattern of the past person, that is, as the statistical data 4. deep.
[0060]
(3) Contents of Statistical Data 4 of Temporal Change of Past Person Distribution FIG. 8 shows an example of statistical data 4 of the temporal change pattern of past person distribution.
[0061]
It shows the change in the number of visitors before the start and the change in the number of exits at the end when the number of visitors at a certain event at Toshiba Theme Park was measured.
[0062]
The number of visitors increases before the start, and the number of exits increases after the end, so the use of surrounding transportation increases. Therefore, based on the prediction of the crowded area / time, it is possible to carry out advertising / advertisement activities intensively according to the facility user's preference.
[0063]
Also, FIGS. 9 and 10 represent the structure of statistical data for each distribution.
[0064]
9 shows a case where there is no statistical data correlation between the distributions 31 to 33, and FIG. 10 shows a case where there is a statistical data correlation between the distributions 31 to 33.
[0065]
For example, considering the distribution A of people in event A and the distribution B of people in event B, if there is no correlation between statistical data, the influence of distribution A and distribution B on each other is not considered. When there is a correlation, the influence of the distribution A and the distribution B on each other is considered.
[0066]
(4) Distribution Prediction Method FIG. 7 is a flowchart showing the distribution prediction method. This processing content is executed by a program stored in the prediction processing unit 5.
[0067]
This program measures the distribution of the current person, and based on this measured value and the statistical data 4 of the temporal change pattern of the distribution of the past person and the future event data 7, the distribution resulting from the event is calculated. It is to be predicted.
[0068]
Specifically, when the current traffic situation and the number of visitors in a facility are known, the traffic situation at the end of the facility is predicted based on past statistical data.
[0069]
(4-1) Distribution Status Measurement In step 1, the current human distribution status is measured. The current distribution status of people refers to, for example, the number of people per unit area in a predetermined event or the number of people in a certain facility.
[0070]
As a method for measuring the distribution of people, there are the following methods.
[0071]
The first method measures the number of people in the facility in real time by measuring the number of people entering and leaving the facility.
[0072]
The second method installs a monitoring camera on the road, recognizes a car that has passed the road, and measures the amount of the car that has passed.
[0073]
The third method measures the traffic volume from road traffic information such as traffic jams and traffic regulations edited and processed at the VICS Center.
[0074]
In the fourth method, the traffic volume is measured from a beacon transmitter installed on a highway / general road in the VICS area.
[0075]
In the fifth method, since the mobile phone company can derive the location information by triangulating from the location of the antenna and base station closest to the caller in the mobile phone, the distribution of mobile phone holders can be obtained using this location information. Measure.
[0076]
The sixth method measures the distribution of terminal holders using a portable information terminal equipped with global positioning system (GPS) technology.
[0077]
The seventh method measures the distribution of people watching TV from real-time TV audience ratings.
[0078]
(4-2) Prediction of distribution status In step 2, the future distribution status of a person is predicted. The prediction is performed by the distribution prediction circuit 10 described above. As an example, the flow of people between event A and event B is predicted using the distribution prediction circuit 10 of the first embodiment.
[0079]
The prediction method is demonstrated based on the flowchart of FIG.
[0080]
In step 11, the current spatial distribution state described above is measured. This is information such as where and how many people are. Specifically, how many people exist per unit area at the entrance of event A and event B. That is, Va and Vb described above are measured.
[0081]
In step 12, a past pattern similar to the current distribution pattern is extracted.
[0082]
In step 13, an event in which an event similar to the event to be predicted has occurred is further extracted from the extracted past pattern.
[0083]
Specifically, a temporal change A (t) in the past distribution of people when the event A is performed, and a temporal change B (t) in the past distribution of people when the event B is performed. Are extracted respectively. Further, Yab of the admittance element 19 is also extracted.
[0084]
In step 14, event A and event are obtained by substituting Yab, Va, A (t), Vb, and B (t) obtained as described above into distribution prediction circuit 10 to obtain current Iab (t). The flow of people between B can be predicted.
[0085]
(4-3) Use of prediction result In step 3, the prediction result is used.
[0086]
That is, based on this prediction result, advertisements / advertisements adapted to the facility user's preference are notified by voice using a speaker, which is one of the advertisement devices 6, or displayed on an image display device.
[0087]
In addition, based on the prediction result, in a transportation system where congestion is expected, the fact that it is crowded is notified by voice using a speaker, or displayed on an image display device. Thereby, congestion can be avoided.
[0088]
【The invention's effect】
As described above, according to the present invention, the present distribution state is measured, and the future person distribution is predicted by modeling based on circuit theory based on the past person distribution state and future event data. Thus, the distribution situation can be easily predicted, and the prediction result can be used for advertising / advertisement activities and congestion reduction.
[Brief description of the drawings]
FIG. 1 is a diagram showing a first embodiment of a distribution prediction apparatus according to the present invention.
FIG. 2 is a graph for explaining distribution prediction according to the present invention.
FIG. 3 is a diagram showing a second embodiment of a distribution prediction apparatus according to the present invention.
FIG. 4 is a graph for explaining distribution prediction according to the present invention.
FIG. 5 is a diagram showing a third embodiment of a distribution prediction apparatus according to the present invention.
FIG. 6 is a diagram showing a distribution prediction apparatus.
FIG. 7 is a flowchart for explaining distribution prediction according to the present invention.
FIG. 8 is a graph for explaining past statistical data.
FIG. 9 is a conceptual diagram for explaining past statistical data.
FIG. 10 is a conceptual diagram for explaining past statistical data.
FIG. 11 is a flowchart of a prediction method.
FIG. 12 is a diagram for explaining the concept of a conventional distribution prediction apparatus.
[Explanation of symbols]
1 Current distribution A
2 Current distribution B
3 Current distribution C
4 Distribution Statistical Data 5 Distribution Prediction Unit 6 Advertising / Advertising 7 Future Event Data 8 Congestion Mitigation 10 Distribution Prediction Circuit 11 Capacitor 12 Norator 13 Narrator 14 DC Voltage Source 15 Capacitor 16 Switch 17 Switch 18 DC Voltage Source 19 Admittance 20 Capacitor 21 Switch 23 DC voltage source 24 Admittance

Claims (6)

A distribution measuring means for measuring the current distribution of people;
Data storage means for accumulating past human distribution status;
Event data storage means for accumulating future event data that is data including the name of the event to be held in the future, the location of the event, and the date and time of the event,
In a distribution prediction apparatus comprising: a distribution prediction unit that predicts a distribution situation of a future person in a specific event using the distribution situation of the current person, a distribution situation of the past person, and the future event data as input. ,
The distribution prediction means includes
A distribution prediction circuit that models the flow of the person by circuit theory; and a voltage value, a current value, or a circuit corresponding to the current person distribution state, the past person distribution state, and the event data. A distribution prediction apparatus characterized by substituting into the distribution prediction circuit as a value of a circuit element to be configured to predict a human distribution. The distribution prediction circuit includes:
It has a distribution prediction circuit for each event,
The distribution prediction apparatus according to claim 1, wherein the distribution prediction circuits for each event are connected by an admittance element, and a human distribution is predicted by a current value flowing through the admittance element. The admittance element is
The distribution prediction apparatus according to claim 2, wherein the distribution prediction apparatus is configured using one or a plurality of circuit elements corresponding to a distribution situation in the past. The distribution of previous human and distribution number of people in the past, the distribution predicting apparatus according to at least one of claims 1 to 3, characterized in that the change pattern that indicates a temporal change of the distribution. A distribution measuring means for measuring the current distribution of people;
Data storage means for accumulating past human distribution status;
Event data storage means for accumulating future event data that is data including the name of the event to be held in the future, the location of the event, and the date and time of the event,
In a distribution prediction apparatus comprising: a distribution prediction unit that predicts a distribution situation of a future person in a specific event by using the distribution situation of the current person, the distribution situation of the past person, and the future event data as input. In the distribution prediction method,
The distribution prediction method is:
A distribution prediction circuit that models the flow of the person by circuit theory; and a voltage value, a current value, or a circuit corresponding to the current person distribution state, the past person distribution state, and the event data. A distribution prediction method characterized by substituting into the distribution prediction circuit as a value of a circuit element to constitute and predicting a human distribution. A distribution measuring means for measuring the current distribution of people;
Data storage means for accumulating past human distribution status;
Event data storage means for accumulating future event data that is data including the name of the event to be held in the future, the location of the event, and the date and time of the event,
In a distribution prediction apparatus comprising: a distribution prediction unit that predicts a distribution situation of a future person in a specific event by using the distribution situation of the current person, the distribution situation of the past person, and the future event data as input. The distribution prediction method is realized by a computer program,
The program of the distribution prediction method is:
A distribution prediction circuit that models the flow of the person by circuit theory; and a voltage value, a current value, or a circuit corresponding to the current person distribution state, the past person distribution state, and the event data. A program for a distribution prediction method, which realizes a function of predicting a human distribution by substituting into a distribution prediction circuit as a value of a circuit element to constitute.

Images