JP7189654B2 - Program, apparatus and method for estimating commercial value of real estate - Google Patents

Description

TECHNICAL FIELD The present invention relates to a technology for managerially analyzing a plurality of properties owned by a company.

In recent years, along with SDGs (Sustainable Development Goals) and ESG (Environment Social Governance), "stranded assets" have attracted attention (see, for example, Non-Patent Document 1). Stranded assets refer to assets whose occupancy rates have fallen significantly below their initial expectations due to sudden changes in the surrounding environment, resulting in significant declines in their value.
On the other hand, even for various real estate such as shops and facilities that were established in the past, those that have low commercial value or cannot be expected to improve are called "stranded assets" in the corporate finance version.

There is a "Hough model" that calculates the value of a store (real estate) owned by a company as the probability that the store will attract users from the viewpoint of location (for example, see Non-Patent Documents 2 to 4).
Specifically, if there are two stores that are located at the same distance from the user's home, the user is likely to be attracted to the store with a larger floor area. Also, if there are two stores with the same sales floor area, it is assumed that the user will be attracted to the store closer to home.
However, the Huff model is for calculating an average (expected) suction rate with the home (place of residence) as the starting point, and does not reflect situations that change in real time.

Generally, the number of people entering a store is considered to be affected by the number of people moving around the store (market area). For this reason, companies sometimes adopt a "dominant strategy" as a store location strategy. This is one of the management strategies to concentrate on developing stores in trade areas with high population movement and to block the entry of other companies. In particular, it may be adopted by convenience stores and drug stores that are the same chain stores.

As another conventional technique, for example, there is a technique of counting the number of people staying in an area by collecting the positions of mobile terminals possessed by users (see, for example, Patent Document 1). Generally, using the subscriber ratio of a particular carrier to all carriers, the number of people connected to the carrier's equipment for a relatively large area is multiplied by the subscriber ratio. By doing so, it is possible to estimate the approximate number of visitors (for example, tens of thousands of people). This is commonly referred to as the "augmentation method". For example, when estimating the population movements of tens of thousands of people in the event of a disaster, if the positional movements of 1% of users can be captured, it is possible to estimate the positional movements of hundreds of users.

WO2012/036222

"Stranded assets", [online], [searched on December 21, 2019], Internet <URL: https://sustainablejapan.jp/2016/05/22/strandedasset/18377> “Huff Model”, [online], [Searched on December 21, 2019], Internet <URL: https://www.pasco.co.jp/recommend/word/word036/#anchor01> "Huff Model", [online], [searched on December 21, 2019], Internet <URL: https://business-map.esrij.com/glossary/2021/> "Trade Area Analysis Introduction Guide", [online], [Searched on December 21, 2019], Internet <URL: https://biz.kkc.co.jp/software/am/ef/ef_guide/article18/> “Is Seven “withdrawal and relocation of 1000 stores” the limit of the dominant strategy?”, [onine], [searched on December 21, 2019], Internet <URL: https://www.itmedia.co.jp/business /articles/1910/16/news032.html＞ "Correlation coefficient", [onine], [Searched on December 21, 2019], Internet <URL: https://ja.wikipedia.org/wiki/%E7%9B%B8%E9%96%A2 %E4%BF%82%E6%95%B0>

The company can measure the number of people entering the store in real time for each store. This is the store's own measured and subjective profit index (hereinafter referred to as "explicit profit index"). The number of people entering the store can be counted based on POS (Point Of Sales) data, images captured by an in-store camera, and wireless communication with a smartphone owned by the user. The size of the number of people entering the store leads to the size of sales and profit of the store.
On the other hand, mobile communication carriers can measure the number of people located around each store in real time. Although this cannot be measured by the store itself, it is a potential profit index (hereinafter referred to as a "potential profit index") that looks at the store objectively. A mobile communication carrier can obtain the location of each user by constantly communicating with the mobile terminal owned by the user.

In general, changes in the number of people entering each store (explicit profit index) are potentially affected by the number of people flowing around each store (potential profit index).
When a company has multiple stores, it is not necessarily the case that it should simply maintain stores with a large number of customers and withdraw from stores with a small number of customers.
A store that increases or decreases the number of people entering the store according to the increase or decrease in the number of people flowing around the store can be said to be affected by the number of people around the store. Here, a store where the number of people entering the store does not decrease even though the number of people flowing around the store has decreased can be said to have its own ability to attract customers. On the other hand, if the number of people entering the store does not increase even though the number of people moving around the store has increased, there is a risk that the store may actually become a stranded asset.

Conventionally, there have been cases in which multiple stores were opened to occupy a commercial area by a "dominant strategy", but all stores were withdrawn from the commercial area without noticing changes in the commercial area (see, for example, Non-Patent Document 5). ). According to this case, it used to be an attractive trade area with a large population flow, but they were forced to make a drastic change in strategy because they did not monitor changes in the trade area. This means that the more a trade area is occupied, the more the trade area itself changes, making it difficult to notice changes in population flow. Conversely, the attractiveness of the occupied trade area has fallen. This is not a problem with the dominant strategy itself, but rather a lack of understanding of the potential profit index based on the number of people moving around the store.
For this reason, managers of companies adopting a dominant strategy need to be constantly aware of changes in population movement in the trade areas where they operate their stores.
However, even if the manager of a company grasps the changes in the population flow in the trade area, it is difficult for him/her to make management decisions about which stores to maintain and which to withdraw.

On the other hand, the inventors of the present application have found that there is a correlation between the number of people entering the store (exponential profit index) and the number of people around the store (potential profit index), and that there is no correlation in some stores. I wondered.
Specifically, the number of people entering the store and the number of people surrounding the store have a positive correlation in some profitable stores, a negative correlation in some profitable stores, and an uncorrelated store. From the perspective of corporate management, if each store can be judged not only by sales but also by the correlation with the number of people around the store, it will be an appropriate reference for store operation. rice field.

The relationship between the number of people entering the store (explicit profit index) and the number of people around the store (potential profit index) does not simply have to have a positive correlation. Certainly, in the case of a store with a positive correlation, the number of people entering the store changes according to the population flow around the store, but the store may not have the ability to attract customers on its own. On the other hand, in the case of stores with a negative correlation, it is possible that the store has its own ability to attract customers, which is the opposite of the population flow around the store. In addition, in the case of uncorrelated stores, it is possible that the store has its own ability to attract customers regardless of the population flow around the store, or that it is a "stranded asset" in a sense.
In this way, it is important for corporate managers to manage a well-balanced store location portfolio in consideration of the correlation between the potential profit index and the actual profit index.

Therefore, the present invention analyzes the commercial value of real estate according to the correlation between not only the index of actual income that is tangibly obtained from the real estate, but also the index of potential income that exists latently around the real estate. It aims at providing a program, an apparatus, and a method.

According to the present invention, a program that causes a computer installed in an estimation device that estimates the commercial value of real estate to function,
an actual profit index recording means for recording an actual profit index obtained from the real estate at predetermined time intervals;
Potential profit index calculation means for calculating a potential profit index based on the Hough model or the advanced Huff model, according to the number of people located within a predetermined range of the real estate at predetermined time intervals;
a potential profit index recording means for recording a time-series potential profit index;
Correlation determining means for determining a correlation coefficient of the time series of the potential earnings index to the time series of the potential earnings index;
a commercial value determining means for determining the correlation coefficient as the commercial value of real estate ;
to make the computer work
Potential Earnings Index Calculation Means
calculating an individual attraction index by dividing the area or attractiveness of the real estate by the distance between the position of the user and the real estate for each combination of the real estate and the user included in the predetermined range;
For each user, as the probability that the user will be attracted to the real estate, a user attraction index is calculated by dividing the individual attraction index of the real estate by the total sum of the individual attraction indexes of all real estates included in a predetermined range,
For each real estate, the sum of the user attraction indices of all users included in a predetermined range is set as the potential profit index for the real estate.
characterized by making the computer function as

According to another embodiment of the program of the present invention,
The predetermined range is the same commercial area including multiple real estate,
The commercial value determining means also preferably causes the computer to determine the correlation coefficient as the commercial value of the property within the predetermined range.

According to another embodiment of the program of the present invention,
It is also preferable that the potential earnings index calculation means causes the computer to multiply the user attraction index by a potential purchasing power index based on the unique information of the user.

According to another embodiment of the program of the present invention,
It is also preferred that the unique information is profile information about the user, or the distance traveled over a period of time, the purchase history, or the length of time the user is located within a given range.

According to another embodiment of the program of the present invention,
The potential profit index calculating means causes the computer to set the user attraction index of the user based on the real estate to 0 when the distance between the position of the user and the real estate increases over time. It is also preferable to make it work.

According to another embodiment of the program of the present invention,
When the types of the first real estate and the second real estate are similar under predetermined conditions,
The potential earnings index calculating means causes the computer to set the user attraction index of the user based on the second property to 0 when it is determined that the user has stayed at the location of the first property for a predetermined time or longer. It is also preferable to make it work.

According to another embodiment of the program of the present invention,
The commercial value determining means are:
A positive correlation in which the correlation coefficient (-1 to +1) is greater than or equal to the first predetermined threshold,
A negative correlation in which the correlation coefficient is equal to or lower than the second predetermined threshold value (≤ the first predetermined threshold value), or
It is also preferable to have the computer function to determine any non-correlation in which the absolute value of the correlation coefficient (0 to +1) is equal to or less than a third predetermined threshold as the commercial value of the real estate.

According to another embodiment of the program of the present invention,
The commercial value determining means are:
If there is no correlation, determine that the commercial value of the property is low; or
It is also preferred to have the computer function to determine that the commercial value of real estate is low in cases of negative correlation and non-correlation.

According to another embodiment of the program of the present invention,
Real estate is a store,
It is also preferable to make the computer function such that the actual profit index is the number of people entering the store, the sales amount, the profit, the number of products purchased, or the number of products sold.

According to another embodiment of the program of the present invention,
Real estate is a building,
It is also preferable to have the computer function so that the tangible revenue index is the number of people staying in the building, or the sales or profits of all the stores in the building.

According to the present invention, an estimation device for estimating the commercial value of real estate,
an actual profit index recording means for recording an actual profit index obtained from the real estate at predetermined time intervals;
Potential profit index calculation means for calculating a potential profit index based on the Hough model or the advanced Hough model, according to the number of people located within a predetermined range of the real estate at predetermined time intervals;
a potential profit index recording means for recording a time-series potential profit index;
Correlation determining means for determining a correlation coefficient of the time series of the potential earnings index to the time series of the potential earnings index;
a commercial value determining means for determining the correlation coefficient as the commercial value of real estate ;
Potential Earnings Index Calculation Means
calculating an individual attraction index by dividing the area or attractiveness of the real estate by the distance between the position of the user and the real estate for each combination of the real estate and the user included in the predetermined range;
For each user, as the probability that the user will be attracted to the real estate, a user attraction index is calculated by dividing the individual attraction index of the real estate by the total sum of the individual attraction indexes of all real estates included in a predetermined range,
For each real estate, the sum of the user attraction indices of all users included in a predetermined range is set as the potential profit index for the real estate.
It is characterized by

According to the present invention, an estimation method for a device for estimating commercial value of real estate, comprising:
The device
an actual profit index recording unit that records the actual profit index obtained from the real estate at predetermined time intervals;
a potential profit index recording unit that records a time-series potential profit index based on the Hough model or the advanced Huff model calculated according to the number of people located in a predetermined range of the real estate at predetermined time intervals;
a first step of determining the correlation coefficient of the time series of the potential return index to the time series of the potential return index;
a second step of determining the correlation coefficient as the commercial value of the property ;
For the potential return index stored in the potential return index recording unit,
calculating an individual attraction index by dividing the area or attractiveness of the real estate by the distance between the position of the user and the real estate for each combination of the real estate and the user included in the predetermined range;
For each user, as the probability that the user will be attracted to the real estate, a user attraction index is calculated by dividing the individual attraction index of the real estate by the total sum of the individual attraction indexes of all real estates included in a predetermined range,
For each real estate, the sum of the user attraction indices of all users included in a predetermined range is set as the potential profit index for the real estate.
It is characterized by executing

According to the program, device, and method of the present invention, the commercial value of real estate is correlated not only with the actual income index that is tangibly obtained from the real estate, but also with the potential income index that exists latently around the real estate. can be analyzed according to gender.

1 is a system configuration diagram in the present invention; FIG. It is a functional block diagram of the estimation apparatus in this invention. FIG. 10 is a graph showing a positive correlation over time between the potential profit index and the actual profit index; FIG. FIG. 10 is a graph showing a negative correlation over time between an actual profit index and a potential profit index; FIG. 4 is a graph showing the uncorrelated time series of the actual profit index with respect to the potential profit index. It is a table used for calculating a potential profit index. FIG. 4 is an illustration representing a potential earnings index for various embodiments of the present invention;

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

Real estate as commercial assets includes, for example:
(1) The real estate may be multiple "stores" owned by the company.
The profit index for each predetermined time in the store is assumed to be the number of people entering the store, but it may be the sales amount, the profit, the number of products purchased, or the number of products sold. For example, assume a store when the company is a franchise chain.
(2) Real estate may be a “building” owned by the company.
The profit index for each predetermined time period in the building is assumed to be the number of people entering the store, but may be sales or profit. For example, assume a commercial building or hotel operated as a real estate investment fund. In the following, real estate will be explained as "stores."

FIG. 1 is a system configuration diagram in the present invention.

According to FIG. 1, the estimating device 1 receives, in real time, various data explicitly obtained from each store a to c via the network. Such data includes, for example, the number of people entering the store, POS data (sales and profit), the number of purchased products, the number of sold products, etc., and is called an "explicit profit index."
The explicit profit index is an index that the store itself can subjectively obtain by looking at the store from the inside.

On the other hand, an unspecified number of people (users) are moving in real time within a predetermined range around each store a to c. The predetermined range around the store may be within a predetermined radius from the center of the store, or may be the prefectural/municipal unit of the address. Moreover, the predetermined range may be a trade area including a plurality of stores (real estate). Furthermore, the predetermined range may be divided into a three-dimensional space such as the number of floors of the building or facility.
Here, the important point is whether or not users who are moving around the store will be attracted to the store.
The number of people flowing around such a store has the potential to enter each store ac, and is referred to as the "potential earnings index."
The potential profit index is obtained by the store itself from the outside, as the number of people around the store increases, the number of people entering the store increases, and as a result, the actual profit index also increases. means no objective index.

According to FIG. 1, each user has a mobile terminal 2 which is in constant communication with the mobile operator equipment. For this reason, the mobile communication service provider associates and accumulates the time and location for each user ID (for example, subscriber ID) possessing the mobile terminal 2 .

The user's position is, for example, as follows.
(1) Terminal Position Position Determined by Mobile Terminal 2 Possessed by User This is latitude and longitude information determined by the mobile terminal 2 itself using a GPS (Global Positioning System).
(2) Base station positioning position of the mobile terminal connected to the base station or access point of the telecommunications carrier The position of the mobile terminal 2 is estimated from the location information of the base station or access point under the control of the mobile terminal 2. may However, this position information has coarse spatial granularity.
These position information may be represented by latitude and longitude or map coordinates, or may be converted into an address name or a map mesh number.

According to FIG. 1(a), the mobile terminal carrier can also derive the number of people located within a predetermined range around the store for each of the stores a to c.
According to FIG. 1(b), in the case of the dominant strategy, it is also possible to derive the number of people located in a "trade area" composed of a plurality of stores a to c.

Here, it is important to note that the number of people moving around the store is the users captured by the communication business equipment of a specific communication carrier, not all of the actual users. In other words, the number of people captured by a particular telecommunications carrier can be counted as far less than the actual number of people staying in that area. That is, not all users' location information can be collected from all carriers.
Here, according to the present invention, it is not necessary to specify the absolute number of people flowing around the store. According to the present invention, which will be described later, it is only necessary to determine the correlation between the number of people entering the store (potential profit index) and the number of people flowing around the store (potential profit index).

FIG. 2 is a functional configuration diagram of an estimation device according to the present invention.

The estimation device 1 estimates the commercial value of real estate.
According to FIG. 2, the estimation device 1 includes an explicit profit index recording unit 101, a potential profit index recording unit 102, a potential profit index calculation unit 11, a correlation determination unit 12, and a commercial value determination unit 13. and These functional configuration units are implemented as programs that cause a computer installed in the apparatus to function. In addition, the flow of processing of these functional components can also be understood as a method of estimating the commercial value of real estate.

[Explicit Profit Index Recording Unit 101]
The explicit profit index recording unit 101 records an “explicit profit index” explicitly obtained from the real estate (shop) at predetermined time intervals. The explicit profit index is, for example, the "number of people entering the store" received from each store.

[Potential profit index calculator 11]
The potential earnings index calculation unit 11 calculates a "potential earnings index" according to the number of people located within a prescribed range based on the real estate (shop) at predetermined time intervals. The potential profit index is, for example, "the number of people moving around the store".

The explicit profit index recording unit 101 and the potential profit index recording unit 102 record "dynamically" by shortening a predetermined time (for example, one minute).
According to the present invention, the user's location is not a static address or a snapshot location, but a dynamic user's location that changes from moment to moment. As a result, the attractiveness of each store changes from moment to moment with respect to the population flow around the store.
A specific calculation example of the potential profit index will be described later.

[Potential Earnings Index Recording Unit 102]
The potential profit index recording unit 102 records a potential profit index corresponding to the number of people located within a predetermined range of the real estate at predetermined time intervals.

[Correlation determination unit 12]
The correlation determining unit 12 determines the correlation coefficient of the time series of the potential earnings index with respect to the time series of the potential earnings index.

A “correlation coefficient” is an index that measures the strength of a linear relationship between two random variables (see Non-Patent Document 6, for example). The correlation coefficient takes real values from -1 to +1. When the correlation coefficient is positive (the correlation coefficient is greater than or equal to the first predetermined threshold), the two random variables are often directly proportional, and when the correlation coefficient is negative (the correlation coefficient is greater than or equal to the second predetermined threshold), the two random variables are often inversely proportional. Also, when the correlation coefficient is uncorrelated (the absolute value of the correlation coefficient is equal to or less than the third threshold), the two random variables are independent (irrelevant). The correlation coefficient is +1 or -1 when two random variables are in a linear relationship.

<Stores with positive correlation>
FIG. 3 is a graph showing a positive correlation over time between the potential profit index and the actual profit index.
According to FIG. 3, the left vertical axis represents the number of people entering the store (explicit profit index), the right vertical axis represents the number of people flowing around the store (potential profit index), and the horizontal axis represents time. represents progress. As is clear from FIG. 3, the actual profit index changes in a similar manner to the potential profit index over time.

<Stores with negative correlation>
FIG. 4 is a graph showing a negative correlation over time between the potential profit index and the actual profit index.
Although FIG. 4 also has the same axes as FIG. 3, the actual profit index changes inversely with respect to the potential profit index over time.

<Stores with no correlation>
FIG. 5 is a graph showing the uncorrelated time series of the actual return index with respect to the potential return index.
Although FIG. 5 also has the same axes as FIGS. 3 and 4, the actual profit index changes with the passage of time with no relation to the potential profit index.

[Commercial value determination unit 13]
The commercial value determination unit 13 determines the correlation coefficient as the commercial value of the real estate within a predetermined range for each real estate. The predetermined range is a predetermined radius around the real estate (store), an area where the real estate is located, or a trade area consisting of a plurality of real estates.
Moreover, the commercial value determining unit 13 may specifically determine the commercial value of real estate as one of the following.
(1) If the correlation coefficient is greater than or equal to the first predetermined threshold, the correlation is positive.
(2) When the correlation coefficient is equal to or less than a second predetermined threshold (≦first predetermined threshold), the correlation is negative.
(3) When the absolute value (0 to +1) of the correlation coefficient is equal to or less than the third predetermined threshold, it is determined as no correlation.

The commercial value may be determining which store to maintain and which store to withdraw in one commercial area based on a dominant strategy. For example, there are the following determination methods, but they are not limited to these depending on management decisions.
(Determination method 1)
For example, if there is no correlation, we determine that the commercial value of the property is low.
Stores with positive correlation and negative correlation -> target for maintenance Stores with no correlation -> target for withdrawal (decision method 2)
Decide that the commercial value of real estate is low if there is negative correlation and no correlation Stores with positive correlation -> to be retained Stores with negative correlation and no correlation -> to withdraw

In the case of FIG. 3 described above, the store is determined to have a "positive correlation" in which the number of people entering the store changes according to the number of people flowing around the store.
In the case of FIG. 4 described above, the store is determined to have a "negative correlation" in which the number of people entering the store changes, contrary to the number of people flowing around the store. The store is assumed to have a customer attraction completely different from the potential profit index, or, in the case of a store with a dominant strategy, to be withdrawn.
In the case of FIG. 5 described above, the store is determined to be "uncorrelated" in which the number of people entering the store changes regardless of the number of people flowing around the store. The store is assumed to have a customer attraction completely different from the potential profit index, or, in the case of a store with a dominant strategy, to be withdrawn. In particular, among multiple stores based on a dominant strategy, a store that is determined to be "uncorrelated" can be said to be a potential "stranded asset" based on corporate finance for the population around the store.

For the dominant strategy, the potential earnings index is based on the number of people moving around the store. The crowd of people flowing around the store changes dynamically from moment to moment. Each store is located to attract that crowd. According to the present invention, the commercial value of a store (real estate), which is how attractive each store is to the crowd of people around the store, is determined by the correlation of the actual profit index to the potential profit index. can be determined as

<<Specific calculation example of the potential earnings index>>

FIG. 6 is a table used to calculate the potential earnings index.

According to FIG. 6, numerical values used for calculating the potential profit index are recorded for each store.
Area/Attractiveness: Index used for Hough model belonging to the store User ID: User with a mobile terminal located in a predetermined area around the store Distance: Straight line distance or route on the map between the store and each user Distance User-specific information: Specific information that belongs to each user Perspective over time: The distance between the store and each user changes over time
Stay history: Whether or not the user stayed at the same type of real estate before a predetermined period User attraction index: Probability q that each user is attracted to the store
Potential profit index: Users who flow around the store are attracted to the store
exponent that would be

FIG. 7 is an illustration representing a potential earnings index for various embodiments of the present invention.

The potential profit index calculator 11 has the following as a specific calculation example of the potential profit index.
<Basic Potential Earnings Index Based on the Hough Model>
<Basic Potential Earnings Index Based on Advanced Huff Model>
<Potential profit index based on user's direction>
<Potential Earnings Index Based on User Stay History>
<Potential profit index based on user's unique information>

<Basic Potential Earnings Index Based on the Hough Model> (See Figure 7(a))
We extend the Hough model used in marketing to adopt a "dynamic Hough model" based on the location of users moving around the store.
A potential profit index based on the “dynamic Hough model” is calculated as follows (see, for example, Non-Patent Document 2).
(S1) Calculate an individual attraction index by dividing the "area" of the real estate by the "distance" between the position of the user and the real estate for each combination of the real estate and the user included in the predetermined range.
Individual Attraction Index: (Area of store i)/(Distance between location of user j and store i)
(S2) User attraction index p(t) obtained by dividing the individual attraction index of the real estate by the total sum of the individual attraction indexes of all real estates included in a predetermined range, as the probability that the user will be attracted to the real estate for each user. ,i,j) (probability that user j is attracted to store i) is calculated.
User attraction index of store i =
(area of store 1) / (distance between location of user 1 and store 1)
/{(area of store 1)/(distance between location of user 1 and store 1)+
(area of store 2) / (distance between location of user 1 and store 2) +
(area of store 3)/(distance between location of user 1 and store 3) +...}
(S3) For each property, the sum of the user attraction indices of all users included in a predetermined range is set as the potential profit index q(i,t) (attraction of store i at time t) in the property.
q(i,t)=Σj ₌ ^1N p(t,i,j)
The potential profit index is an index that estimates the potential attraction force that will enter the real estate from the number of people moving around the dynamically changing real estate.
It should be noted that the potential profit index is an index obtained by simply adding up the user attraction index, and is not a probability.

Incidentally, when a company owns a plurality of stores in a trade area, the potential profit index of the entire trade area is calculated by summing the potential profit indices of each store.
Q(t)=Σi ₌ ^1M q(i,t)

<Basic Potential Earnings Index Based on Advanced Hough Model> (See Figure 7(b))
We extend the Advanced Huff Model used in marketing to adopt a "dynamic Advanced Huff Model" based on the location of users moving around the store.
While the Huff model described above uses the sales floor area to calculate the potential profit index, the dynamic "advanced Huff model" uses "attractiveness" (see, for example, Non-Patent Document 3). Also, the distance between the user's position and the store is not the straight line distance on the map, but the "distance distance".

(S1) Calculate an individual attraction index by dividing the "attractiveness" of the real estate by the "distance distance" between the location of the user and the real estate for each combination of the real estate and the user included in the predetermined range.
Individual Attraction Index: (Attractiveness of store i)/(Distance between location of user j and store i)
(S2) For each user, as the probability that the user will be attracted to the real estate, the "user attraction index" p is obtained by dividing the individual attraction index of the real estate in question by the total sum of the individual attraction indexes of all real estates included in a predetermined range. Calculate (t, i, j) (the probability that user j will be attracted to store i).
User attraction index of store i =
(attractiveness of store 1) / (location of user 1 and distance from store 1)
/ {(Attractiveness of store 1) / (Distance between user 1's position and store 1) +
(Attractiveness of store 2) / (Distance between location of user 1 and store 2) +
(Attractiveness of store 3) / (Distance between user 1's position and store 3) +
...}
(S3) For each property, the sum of the user attraction indices of all users included in a predetermined range is set as the potential profit index q(i,t) (attraction of store i at time t) in the property.
q(i,t)=Σj ₌ ^1N p(t,i,j)

<Potential profit index based on user's direction> (see FIG. 7(c))
The potential earnings index calculation unit 11 calculates the potential earnings index using the dynamic Hough model or the dynamic advanced Hough model described above, and the distance between the position of the user and the real estate increases over time. In this case, the user attraction index of the user based on the real estate is set to zero.

According to FIG. 7(c), the traveling direction of the user 1 is heading toward the store B. As shown in FIG. At this time, since at least the user 1 is away from the store A, it should be considered that the probability of entering the store A is almost zero. In spite of this, when the attraction force is calculated including the store A, a rational user attraction index cannot be calculated.
Specifically, when the distance from the location of the user to the store decreases at regular intervals (for example, one minute), the user is not included in the user attraction index of the store.
Distance between the position of user i and store A at time t d(t,i,A)=|p(i,t)−p(A)|
Using the indicator function, the user attraction index to the store is calculated, limited to the approaching users.
p(t,i,j)×1{d(t,i,A)−d(t−Δt,i,A)<0}
As a result, the total sum of user attraction indices is calculated by limiting to users approaching each store.
q(i,t)=Σp(t,i,j)×1{d(t,i,A)−d(t−Δt,i,A)<0}
In this way, by also considering the user's traveling direction, it is possible to calculate the user attraction index of the store that is in line with the actual situation.

<Potential Earnings Index Based on User's Stay History> (See FIG. 7(d))
When the types of the first real estate and the second real estate are similar under a predetermined condition, the potential earnings index calculation unit 11 determines that the user has stayed at the first real estate for a predetermined time or more. , the user attraction index of the user based on the second real estate is set to zero.

For example, assume that both the first store and the second store are "restaurants". Here, the probability that the user will enter the second store after staying at the first store is extremely low. This is because the user is highly likely to have already eaten and eaten during his/her stay at the first store, and is less likely to enter the second store even if the user is located near the second store. In such a case, even if a user staying at the first store is located near the second store, the probability that the user will be attracted to the second store when calculating with the Hough model is Should be 0.

Similarly, for example, a user (driver) of a car who has finished refueling at a first gas station should have zero attraction to the second gas station even if there is a second gas station in the direction of travel. is. Even if a car with a full tank of gasoline comes near another gas station, the probability of entering that gas station is extremely low.

Thus, not only the user's location, but also the user's stay history is considered to calculate the potential earnings index.
A function that outputs a set of types of stores that user j visited during a time interval T(t)=(tT, t) that is a certain period of time before current time t is expressed as follows. Suppose that when a store is input, the type of the store is output.
sy (T(t),j)
For example, when store A is a restaurant, sy(A)=“restaurant” is output.
If the user j visits a store of the same type as the store A during T(t), the indicator function that is 1 is as follows.
[1-1 {sy(A)∈sy(T(t),j)}]
As a result, 0 is output when user j stays at the same type of store within a certain period of time, and 1 is output otherwise.
Then, the user attraction index of store A is calculated as follows.
q(i,t)=Σp(t,i,j)×1{d(t,i,A)−d(t−Δt,i,A)<0}
×[1−1{sy(A)εsy(T(t),j)}]

<Potential Earnings Index Based on User's Unique Information> (See FIG. 7(d))
The potential profit index calculator 11 preferably multiplies the user attraction index by a potential purchasing power index based on the user's "unique information".
The potential purchasing power index of user j in past period d is defined as follows.
pbp(j,d)
The potential purchasing power index is determined based on the unique information, and the unique information is the profile information of the user, the distance traveled in a predetermined period, the purchase history, or the length of time the user is located in a predetermined range. There may be.

For example, a user who travels a longer distance between holidays is considered to have higher spare funds and activity, and a higher potential purchasing power index. Conversely, users who travel shorter distances between holidays are considered to have lower living resources, less activity, and a lower potential purchasing power index. Of course, the user's potential purchasing power index may also be determined based on the usage history of credit cards and the balance of deposits in a bank account. Also, it may be assumed that the longer a user stays in an area such as a luxury residential area or a luxury hotel, the higher the potential purchasing power index of the user.
q(i,t)=Σj ₌ ^1Np (t,i,j)×pbp(j,d)

As described in detail above, according to the program, device and method of the present invention, the commercial value of real estate is not only the actual profit index that is tangibly obtained from the real estate, but also the potential It can be analyzed according to the correlation with the potential return index that exists.

According to the present invention, it is possible to constantly monitor the correlation between the potential earnings index that changes moment by moment around the real estate and the actual earnings index obtained from the real estate. For the manager of a company, it is possible to provide information for managerial judgment as to which store should be maintained and which store should be withdrawn from each store included in a market area based on, for example, a dominant strategy.

For the various embodiments of the present invention described above, various changes, modifications and omissions within the spirit and scope of the present invention can be easily made by those skilled in the art. The foregoing description is exemplary only and is not intended to be limiting. The invention is to be limited only as limited by the claims and the equivalents thereof.

1 estimation device 101 explicit profit index recording unit 102 potential profit index recording unit 11 potential profit index calculation unit 12 correlation determination unit 13 commercial value determination unit 2 portable terminal 3 store

Claims (12)

A program that operates a computer installed in an estimation device that estimates the commercial value of real estate,
an actual profit index recording means for recording an actual profit index obtained from the real estate at predetermined time intervals;
Potential profit index calculation means for calculating a potential profit index based on the Hough model or the advanced Huff model, according to the number of people located within a predetermined range of the real estate at predetermined time intervals;
a potential profit index recording means for recording a time-series potential profit index;
Correlation determining means for determining a correlation coefficient of the time series of the potential earnings index to the time series of the potential earnings index;
a commercial value determining means for determining the correlation coefficient as the commercial value of real estate ;
to make the computer work
Potential Earnings Index Calculation Means
calculating an individual attraction index by dividing the area or attractiveness of the real estate by the distance between the position of the user and the real estate for each combination of the real estate and the user included in the predetermined range;
For each user, as the probability that the user will be attracted to the real estate, a user attraction index is calculated by dividing the individual attraction index of the real estate by the total sum of the individual attraction indexes of all real estates included in a predetermined range,
For each real estate, the sum of the user attraction indices of all users included in a predetermined range is set as the potential profit index for the real estate.
A program characterized by causing a computer to function as The predetermined range is the same commercial area including multiple real estate,
2. The program according to claim 1, wherein the commercial value determination means causes the computer to determine the correlation coefficient as the commercial value of the real estate within a predetermined range. 3. The program according to claim 1, wherein the potential profit index calculation means causes the computer to multiply the user attraction index by a potential purchasing power index based on the user's unique information. 1. The unique information is profile information about the user, or the distance traveled in a predetermined period, the purchase history, or the length of time the user is located within a predetermined range, which causes the computer to function. 4. The program according to any one of 3 . The potential profit index calculating means causes the computer to set the user attraction index of the user based on the real estate to 0 when the distance between the position of the user and the real estate increases over time. 5. The program according to any one of claims 1 to 4 , which is caused to function. When the types of the first real estate and the second real estate are similar under predetermined conditions,
The potential earnings index calculating means causes the computer to set the user attraction index of the user based on the second property to 0 when it is determined that the user has stayed at the location of the first property for a predetermined time or longer. 6. The program according to any one of claims 1 to 5 , which is caused to function. The commercial value determining means are:
A positive correlation in which the correlation coefficient (-1 to +1) is greater than or equal to the first predetermined threshold,
A negative correlation in which the correlation coefficient is equal to or lower than the second predetermined threshold value (≤ the first predetermined threshold value), or
Claims 1 to 6 , characterized in that the computer functions to determine any one of non-correlations in which the absolute value (0 to +1) of the correlation coefficient is equal to or less than a third predetermined threshold as the commercial value of the real estate. The program according to any one of Claims 1 to 3. The commercial value determining means are:
If there is no correlation, determine that the commercial value of the property is low; or
8. The program of claim 7 , causing a computer to determine that commercial value of real estate is low in cases of negative correlation and non-correlation. Real estate is a store,
9. A computer according to any one of claims 1 to 8 , wherein the actual profit index is the number of people entering the store, sales, profits, the number of products purchased, or the number of products sold. The program described in Section. Real estate is a building,
9. The method according to any one of claims 1 to 8 , characterized in that the tangible profit index is the number of people staying in the building, or the sales or profits of all stores in the building. program as described. An estimating device for estimating commercial value of real estate, comprising:
an actual profit index recording means for recording an actual profit index obtained from the real estate at predetermined time intervals;
Potential profit index calculation means for calculating a potential profit index based on the Hough model or the advanced Huff model, according to the number of people located within a predetermined range of the real estate at predetermined time intervals;
a potential profit index recording means for recording a time-series potential profit index;
Correlation determining means for determining a correlation coefficient of the time series of the potential earnings index to the time series of the potential earnings index;
a commercial value determining means for determining the correlation coefficient as the commercial value of real estate ;
Potential Earnings Index Calculation Means
calculating an individual attraction index by dividing the area or attractiveness of the real estate by the distance between the position of the user and the real estate for each combination of the real estate and the user included in the predetermined range;
For each user, as the probability that the user will be attracted to the real estate, a user attraction index is calculated by dividing the individual attraction index of the real estate by the total sum of the individual attraction indexes of all real estates included in a predetermined range,
For each real estate, the sum of the user attraction indices of all users included in a predetermined range is set as the potential profit index for the real estate.
An estimation device characterized by: A method for estimating a commercial value of real estate, comprising:
The device
an actual profit index recording unit that records the actual profit index obtained from the real estate at predetermined time intervals;
a potential profit index recording unit that records a time-series potential profit index based on the Hough model or the advanced Huff model calculated according to the number of people located in a predetermined range of the real estate at predetermined time intervals;
a first step of determining the correlation coefficient of the time series of the potential return index to the time series of the potential return index;
a second step of determining the correlation coefficient as the commercial value of the property ;
For the potential return index stored in the potential return index recording unit,
calculating an individual attraction index by dividing the area or attractiveness of the real estate by the distance between the position of the user and the real estate for each combination of the real estate and the user included in the predetermined range;
For each user, as the probability that the user will be attracted to the real estate, a user attraction index is calculated by dividing the individual attraction index of the real estate by the total sum of the individual attraction indexes of all real estates included in a predetermined range,
For each real estate, the sum of the user attraction indices of all users included in a predetermined range is set as the potential profit index for the real estate.
A method for estimating a device, characterized in that:

Images