JP2003281348A - Market area analyzing system, method, program and recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve the precision of sales prediction at a new store and the prediction of market area space structure by estimating the market area space structure of a conventional store in a business kind and status in a competitive relation by using GIS. <P>SOLUTION: A market area analyzing system 10 comprises: a DB editing part 11; a simulation part 12; a displaying and analyzing part 13; a local area statistic DB 14 storing statistic information about consumers by each area obtained by dividing prepared map data with a designated conditions; and a store DB 15 storing store information including at least a plurality among a store name, the area of a store, the latitude and longitude of a location and a sales result value by each business status. The system 10 designates the object area of market area analysis on the map data and performs simulation to each store included in the designated object area to decide the parameter of a prescribed probability model so as to close to the sales result value stored in the store DB 15 to predict the sales at the new store and the market area space structure by using the probability model the parameter of which is decided. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention relates to a trade area analysis system, method, program, and recording medium, and more specifically,
Existing GIS (Geographic Information)
The present invention relates to a trade area analysis system, method, program, and recording medium for analyzing a trade area between stores in a competitive relationship by using a raster engine or the like as a background map, without depending on the engine.

[0002]

2. Description of the Related Art Map software generally called a GIS map engine has the following basic functions. (1) Enlargement / reduction or switching of scales and scrolling can be performed over a wide area. (2) It can handle point data with latitude and longitude and can be expressed on map data. (3) In the case of the vector engine, vector map data composed of vector format map elements can be arbitrarily layered according to the purpose, and attribute data can be linked to the vector map elements. In addition, vector format map elements can be processed and physical and topographical calculations can be performed. (4) In case of raster engine, there is no layer function,
There is no map element processing or element attribute reference display function,
It has the functions (1) and (2) above, and also has a route search and calculation function.

The map data in vector format is point sequence data that can be treated as data in which individual points are associated with position coordinates. As graphics, points (isolated points), nodes (intersections of lines), lines (lines) ) And a polygon (area and surface defined by a closed curve).
A vector engine is map software that can handle these map data and has basic functions. The vector-type basic map data includes administrative boundaries such as coastlines, lakeshore lines, and town-border boundaries and place names, roads, railways, rivers, and contour lines. These can be easily layered using a vector engine, element attributes can be referenced, map elements can be processed, and topographical and physical calculations such as area and distance calculations can be performed. Vector engines are important when users have their own vector map databases, especially in the field of government. For example, the administration can effectively use the GIS vector engine in the following fields according to its purpose. City planning information system (current land use status, current building status, etc.) Road information management system (current road status, road ledger plan, road ledger, etc.) Fixed asset information management system (lot number map, house diagram + land tax ledger + house tax ledger → Fixed asset survey / evaluation business, fixed asset management business, etc.) Water and sewer management information system (water supply and drainage facilities, sewer facilities, etc.) Disaster prevention information management system (display of disaster information such as fire, earthquake, flood damage, damage calculation, evacuation measures, etc.) Environmental assessment Landscaping / park planning Waste treatment planning

However, the biggest obstacles to the introduction of a vector engine by a general company for marketing purposes are constraints such as high cost for creating a vector map and difficulty in marketing analysis. You must purchase expensive basic vector data such as roads, railways, rivers, place names, etc., even if you prepare the minimum shape as a map. The same applies to the case of local governments.In addition to basic data, the latest topographic maps, lot numbers, and house maps must be created as vector data by each local government, creating a blank map from aerial photographs and aerial laser surveys. It is difficult to make a budget without a local government. Also, the census mesh data and town-by-town data are linked to these polygons by codes, but in the case of the vector engine, the data display can be done by clicking on each polygon Since it is displayed in a card format in, the attribute data cannot be layered directly on the displayed all-polygon map. Analytical display that gives a bird's eye view of the entire area can only be color-coded by hierarchy. This is the shortcoming of vector engines in marketing analysis.

A raster type engine is an engine that handles map data in which the smallest surface is a pixel and the whole is treated as a drawing. Enlargement / reduction switching, scrolling, point display, point management, road distance calculation, etc. It is possible, but it does not have the layer function and the same processing function of the vector map element. The greatest advantage of this raster map is that it is completed as a picture from the beginning, and it is not an object to be processed, but much more map information is drawn than the vector type, so it is extremely inexpensive as a map. is there. Instead, the purpose of analysis is limited, and it is mainly used for logistics / delivery planning, route search, salesman management, and user point database management.

Here, in the case of the above vector engine,
In addition to the basic map, if the user has his or her own vector map database, the vector map databases owned by the user will be overlaid each time according to the purpose, and a map will be created according to the purpose, and the processing will be topographically calculated. However, the marketing analysis function is limited and extremely poor for the above reasons. Geographical information is defined as attributes of map elements (which are coded internally), and census statistical data by mesh and town-character is also defined by being linked to these polygon elements, and user-created data is also an element. Be tied to. The element attribute data is treated as a separate frame, and the attribute data itself is not layered.

In the case of a vector engine, most of the above marketing-like analysis functions are such that the representation of the data itself is obtained through individual map elements (specifically, by clicking points, nodes, lines, polygons, etc.). It is limited to card type that can refer only to information. Polygonal numerical data (for example, population by town-chome character, number of households, etc.) can be displayed in a bird's-eye view or as a graph or by using Microsoft. Whether to display a list using Excel (registered trademark) or the like as a separate table. In addition, the geographic information arithmetic functions that can be performed on them are extremely limited. The mesh is defined by dividing the latitude and longitude into equal parts, so it is a polygon, but unlike the town-choji boundary, it can be handled without depending on the universal coordinate system, but it is It is the current situation that the mesh display is subject to the same restrictions, probably because they use the same module with the same element attribute concept.

As described above, the problems in marketing analysis are common to many systems. Therefore, even if a company or the like introduces the GIS vector engine only for marketing analysis, it is too expensive and cannot be analyzed as expected. This also applies to marketing analysis conducted by the government.

In addition, when users come to selectively use various facilities beyond the resistance of distance from the viewpoint of location problems that are in a competitive relationship, in other words, there is intense competition between the facilities, and each facility Let us consider the location problem when the usage areas and commercial areas of each other overlap and are unstable, and the commercial areas can be expressed only in a probabilistic sense. An example of such a stochastic trade area model is DAVID in 1964.
There is a so-called Huff model proposed by L. HUFF, but if you try to solve this as an electronic GIS problem, over all the selected spatial areas, regarding all stores in the business category in question and all consumers, I have to calculate. It is not possible to take up and calculate only a few specific stores. Because the commercial areas of each store overlap each other,
This is because the effect of any change at any point extends far through the overlapping trade areas. Moreover, actual sales are known for existing stores, and for consumers, the amount of consumption by category is known as the average amount by household surveys and the like. First of all, it is necessary to prove that the actual data itself is explained by the above trade area model by combining all of them.

Here, the above Huff model is used for most G
It is built into an application that uses the IS engine. However, they can only be calculated once according to the model (probability or regression analysis), including what is called simulation. With a model that you set the actual sales of the same or different stores that are competing in the same area that are scattered in many areas,
To what extent can it be explained, and at the same time how the trade area spatial structure at that time is represented by distance and probability,
The empirical process is not incorporated. Therefore, nothing can be said about the degree of certainty of the calculation. This problem is non-linear in nature,
This is a so-called complex phenomenon. Therefore, in order to solve this problem, after constructing the model according to the reality, the overall behavior (consumer behavior) is generated by simulation, the realistic validity of the result is examined, and the model parameters are automatically and manually determined. We have to take a constructive approach in complex systems, changing it and repeating it until the whole phenomenon is explained to a satisfactory level. Only then will it be possible to competitively predict the sales of new stores and the spatial structure of the commercial area. At the same time, this phenomenon
Being a geospatial phenomenon, all results must be structured so that they can be analytically layered on an electronic map.

In the conventional Huff model, the probability that a consumer living in an area R (accessibility) from the i-shop will go to the i-shop (the probability)
yof a consumer at a given
point of distance R from
shop i traveling to the
shop i) is the equation (10) or equation (11) shown below.
It is represented by.

[0012]

[Equation 1]

∝ shown in the above equation (10) is proportional to
Σ shown in 1) represents the sum, and Mi in Expression (10) is
Sales floor area such as retail trade or attraction i
ndex (index of attractiveness) is shown. From this formula, Huff (1964) states that the number of customers visiting from the R point in the i store can be calculated as follows. Ei (R) = Pi (R) × N (R) Equation (12) In the above Equation (12), N (R) = the expe.
cted number of cosumers a
t R that like to travel
to shop i, N (R) = the number
er of consumers at R. Then, if R is summed (integrated), the total number of customers visiting the i store can be obtained.

The above probability is proposed as a shopping trip probability that a consumer goes to a corresponding store in a certain period (how many times in a period N times the number of times of shopping goes to the corresponding store). Until then, the theory of the trade area was a normative theory that was determined only by the physical quantity of the central area (or shopping center), the population, and the distance, but Huff linked the shopping behavior (trip) of consumers to the physical quantity and distance of the central area. Since it proposed a stochastic model, it was accepted as a breakthrough in marketing and geography. Capturing consumer behavior on trips is in line with the flow of policy engineering in America in the 1960s. For example, in formulating a road network maintenance plan, an OD table (Origin-Destination) was devised at that time to estimate and predict interregional road traffic volume.
Table) samples the vehicle drivers by vehicle type at each point on the transportation network, conducts a questionnaire survey of departure points and destinations, estimates the current pattern of trips between the departure points and destinations, and calculates the population of each region.・ The future of industrial traffic, tourism traffic, and shopping traffic will be predicted from the industrial agglomeration, and the interregional future traffic will be predicted from that to make a traffic development plan between positions. It was Huff came up with this idea, and it has been inherited in America and Japan to this day. (England and Japanese geographers advocating the theory of spatial interaction still calculate trade areas based on the OD table above.)

The problem in the Huff's equation from the beginning was how to find the parameter α over distance. It was pointed out by Huff that it would vary by product category. Hough himself pays attention to the consumer's shopping trip and investigates the destination of shopping and the number of periods by the consumer on the departure side.
It is said that observation data was created and calculated by the Fibonacci search method. As a result, although the value of α differs depending on the product category, for example, when shopping for furniture, 2.723,
Huff reported in the same paper that he got 3.191 from shopping for clothes. It is considered that the parameter α is different not only in the product or consumption category but also in the same category depending on the type of industry, the type of business in some cases, and the type of business in each store. In addition, there is a possibility that it will differ depending on the region in which competition is occurring even in the same business category (in other words, the analysis area).
For this reason, making observation values and empirically finding α is difficult and meaningless. Therefore, in the 1980s, the Ministry of International Trade and Industry made an administrative decision to set α to 2.0 for the normative purpose of mediating the large store law at that time. However, in order to make a decision on store investment that determines the fate of a company and to formulate a marketing policy that enhances investment efficiency, it is necessary to obtain α that has the ability to explain reality in any analysis domain. The method of simulation is extremely effective. The same can be said for the power β to be multiplied by the area.

After Huff's suggestion, the probabilities were interpreted in two directions by subsequent practitioners in order to calculate store sales therefrom. One is the direction of reading as trip probability in accordance with Huff's argument. In this case, it was necessary to newly know the total number of trips T that occurred during an individual or household and the average amount of shopping per trip. The other direction is to read that probability indicates the ratio n / N (probability of store selection use) of consumers n who use the store during the period among the collective consumers (households) N at point R. In this case, the average value obtained from household surveys, etc. was used for the period category consumption amount per consumer required for sales calculation. This latter interpretation is popular because of the small amount of information required.

However, since the point card DB has become widespread, what has been learned about the purchasing behavior of customers is that even in a large supermarket handling food groceries (hereinafter referred to as SM), for example, in the case of a customer within the nearest 300 m, the monthly visit frequency is 1 50 times a month, from 35 times to 35 times
From ¥ 00 to ¥ 90,000, the purchase price per purchase is 500
This means that from circles to 5,500 yen, all are widely distributed in a form that approximates a normal distribution. As the distance from the store increases, the number of consumers who have intentionally selected becomes larger, so the shape of the distribution deviates from the normal distribution, but nonetheless, these three terms are similarly widely distributed. Individuals have various values, some people choose the store and some do not, and those who choose do not act uniformly,
The consumption behavior draws a fairly distributed normal distribution around the average value. In other words, in areas where various factors and actions from stores are seen and consumers also have various needs, consumer value judgments and selection behaviors are diversified, resulting in a normal distribution. , And human diversity as such a group, in a slight difference,
It enables the existence of many stores. In other words, SM and GMS (General Merchan)
The distress store is placed in a strong competitive location environment, but conversely, many stores can be located. Therefore, even if the Hough model can be applied, the target is not the individual consumers but the consumers as a group, and the average value thereof.

Regarding the average value, what has been found in the point card DB and the customer visitor survey is that, as expected from the Huff model, there are two terms that attenuate with respect to distance. The first is the zone visit ratio. If the area (zone) is zoned along the distance R from the store and the number of consumers (households) in each zone is N (R) and the number of customers from that zone is n (R), then n (R) / N (R) With the distance R

[0019]

[Equation 2]

As described above, α is often around 1.0 in SM and GMS. This means that customers on a certain Sunday are strictly sampled by system extraction, or analyzed by point card receipts, and the area is zoned along the distance R from the store, and the number of households in each zone If one plots the percentage of numbers along the distance R, it is very often demonstrated. This n
(R) / N (R) is conventionally thought to indicate the Hough probability.

[0021] The other is the zone store consumers (households)
The average purchase price over the period, which is not a constant as practitioners have thought, and decays relatively slowly with distance. In the case of large SM, it is 35,000 to 25,500.
Decays to 0 yen. On the contrary, the average purchase price per purchase is not a constant as Huff and others have considered, and in the example of a large SM, it rises with R from 2000 yen to 4000 yen. In other words, the average price per customer cannot be treated as a constant. Therefore, if this probability is used as a probability of coming to the store (trip frequency), then it becomes necessary to model the consumption amount per time as a variable that depends on the distance. Similarly, if we look at this probability as the probability of store selection and use (the ratio of store visits from the perspective of the store), then it becomes necessary to model the average period consumption as a function of distance.

Then, what is the probability of Huff becomes clear when the following is considered. As mentioned above, among the collective consumers living at the distance R, if the average purchase amount at the i store is Si (R), this is empirically a decay function of the distance. The sales Ui (R) of the customer from the R zone of the i store is: Ui (R) = Si (R) × n (R) ... Formula (13). On the other hand, if the consumption amount in the household survey for the merchandise category handled by i store is C and its business category share (described later) is Φi, the total consumption amount of consumers living at R is Φi × C × N. (R). At this time, since it does not have any more information in a general form, it is expected that the probability that the i store can attract the consumption as sales is expressed as Pi (R) as follows. . That is, is the sales of the i store: Ui (R) = Pi (R) × Φi × C × N (R) ... Equation (14). Then, since these two sales are the same, Pi (R) is represented by the following equation (15). Pi (R) = [Si (R) / Φi × C] × [(n (R) / N (R))] Equation (15)

In the above equation (15), the first term on the right side is the customer consumption rate of the zone R in the i store, which is a slow decay function of the distance. The second term is the store selection utilization rate for selecting the i store, which is empirically a sharp attenuation function of the distance. Since the product of the damping functions is also the damping function,
Especially the latter is empirically with the distance R

[0024]

[Equation 3]

Since it is known that the attenuation is performed as shown in FIG. 5, it can be assumed from here that the equation (10) is obtained as in the case of Hough. Therefore, the probability of Huff is: Pi (R) = (consumption rate of customers coming from R in the store) x (selection utilization rate of households living in R) = store selection of residents N (R) in the R zone at i store Consumption probability = Si (R) × n (R) / Φi × C × N (R) ... Interpreted as equation (16). As described above, this interpretation is consistent with the recent knowledge of the point card DB. Instead, the exponentiation of the distance α in this interpretation is not a conventional verification means, and is not known unless simulation is performed in the form of what is the value of α that can explain the actual sales for a large number of stores at once.

SM, G in various geographical areas
In order to theoretically experiment the effect of α on the competitive relationship of MS, for example, α = 0.5 (in this case, the consumer moves through space very easily) to α = 2.0 (in this case, consumption When the simulation calculation is performed by changing the setting up to the point where the loyalty of the person is extremely slow and the position loyalty to the neighboring stores is high), the convergence is fastest around α = 1.0 (that is, the theory of each designated store in the area). The value will be close to actual sales without much modification of the actual sales surface), and the resulting commercial area structure will be close to reality (compared to the ratio of zone households represented by consumption probability contour lines). α =
The further away from 1.0, the slower the convergence, and the more difficult it becomes to converge itself. In other words, the actual sales and commercial area structure of each store cannot be explained. The actual sales surface has been greatly modified, and the resulting trade area structure becomes thin and wide as it approaches α = 0.5, and becomes narrower and higher as it approaches α = 2.0. Thus, the Sunday / holiday zone consumer visit rate is relatively close to the consumption probability).

Competition within the same consumption category in different industries (for example,
There is a possibility that α may be different (drug department of GMS and drug store of GMS), and even in competition between different industries of the same industry (the possibility that SM is larger between SM and GMS). α is specified for each business type (however, since it is specified for each store in the program, it can be specified for each store in some cases). The magnitude of α means the magnitude of mobile mobility of collective consumers, and in the same consumption category, mobile mobility is high for certain business categories (α <1.0) and for certain business categories. Has low mobility (α>1.0; high position loyalty)
It is possible. Even when dealing with such a competition model between different industries, the business type file and the consumer business type share are effectively used.

As described above, the probabilistic model regarding the consumption behavior adopted in this simulation formally corresponds to the Huff model, but the probability is neither the shopping trip probability nor the store selection use probability, but the collective consumption living in the area of the distance R. It is a store selection consumption probability in a person's store. The distance R is originally accessibility, and from the viewpoint of cognitive psychology, it may be a linear space distance, and from the viewpoint of economic cost, it may be a time distance measured along the road. However, in reality, consumers living in the zoned plane are included in the calculation, so there is not much difference between the results.

Next, how to handle the sales floor area M is described. This is originally defined as an attraction index, and as an index of the attractiveness of a store, for example, in addition to the sales floor area showing the degree of integration of the product lineup, qualitative differences in the product lineup, the degree of human service, It was implied from the early days of the model proposal that it would be decided by a combination of factors such as the parking lot area, the concentration of service facilities, and the accessibility of roads around stores. The problem is, in the first place, this model explains what kind of value this attractiveness index for each store that can explain the sales of each store in the actual space, in other words, the attraction of individual store customers Whether it can be done or not, there has been no proof method in the past. This is also sought by a method called simulation. Therefore, the concept of individual store customer attraction coefficient is introduced and defined as follows. Individual store customer attraction force = Individual store floor area x Individual store customer attraction coefficient ・ ・ ・ Equation (17)

Here, the above-mentioned individual store customer attraction force is replaced by the individual store sales floor area in the model. In order for the simulation to succeed, it is necessary to introduce other concepts such as those described below, but one of the central themes of the simulation is that What is the value of this individual store customer attraction coefficient? (In addition, it has been proposed to introduce an exponentiation of the area β, which is the case in the program.
The best way is to modify the area with = 1.0. )

In this Huff model, in order to simulate reality, the interpretation of the probability is changed, the value of α is changed, and the individual customer attraction coefficient is introduced as described above. Different concepts, variables and parameters must be added, and the modules must be structured according to those concepts, variables and parameters. In addition, such a large amount of non-linear calculation was difficult to execute on a conventional personal computer.

[0032]

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned circumstances, and when analyzing the commercial areas of each store in a business category in a competitive relationship using GIS, particularly between stores. To provide a commercial area analysis system, method, program, and recording medium capable of improving the accuracy of sales forecast of a new store and the prediction of the commercial area spatial structure and realizing advanced marketing analysis when there is intense competition. That is the purpose.

More specifically, in order to estimate the sales and the commercial area spatial structure of each store in a business category having a competitive relationship, a new concept and variable are added to the stochastic model (consumer behavior model for selecting consumers). A module configuration for adding and operating a computer using them, and repeatedly performing automatic calculation until it matches the actual data, and empirical verification,
It is an object of the present invention to provide a commercial area analysis system having a module configuration for predicting sales for a new store investment after the verification process and predicting a commercial area spatial structure.

[0034]

According to a first aspect of the present invention, in a trade area analysis system for setting a trade area on map data and analyzing the set trade area, map data prepared in advance is specified under a specified condition. A regional statistical database that stores statistical information about consumers for each of the divided areas, a target area designating unit that specifies a target area for the commercial area analysis on the map data, and a latitude and longitude indicating a store name, a sales floor area, and a location according to business types. , A store database storing store information including at least one of a plurality of actual sales values, and a simulation for estimating a commercial area spatial structure and theoretical sales for each store included in the target area designated by the target area designating unit By performing the calculation, the parameters of the given probabilistic model can be approximated to the actual sales values stored in the store database. It is obtained by; and a simulation means for determining the.

According to a second aspect of the present invention, in the first aspect of the present invention, there is provided a store designating unit for designating a store to be a simulation target from the stores included in the target area, and the simulation unit specifies the store. The parameter of the probabilistic model is determined so as to approach the actual sales value stored in the store database by performing a simulation for each store designated by the means.

According to a third aspect of the present invention, in the first or second aspect of the present invention, the simulation means stores the estimated sales value calculated based on the result of simulation using the probabilistic model, and stores the estimated sales value in the store database. The parameter of the probabilistic model is determined by repeatedly performing simulation until the difference from the actual sales value falls within a preset allowable range with respect to the actual sales value.

A fourth aspect of the present invention is the invention according to any one of the first to third aspects, in which the probabilistic model is the exponentiation β of the sales floor area of a store designated by business category and from the customer residence point to each store. A power of distance α, and a limit distance that represents the distance from the simulation target store included in the target area to the customer's residential area located farthest away,
Customer attraction that represents the level of attraction (attractiveness) that attracts customers at stores, the maximum number of stores that can be selected by business category that customers can select within the limit distance, and the stores of each business category In all consumption categories in the household survey, the consumer business category share, which represents the ratio of collective consumers that can be actually consumed by stores in each business category as an average of one household to the total amount of consumption in all target categories per household, Is defined as a parameter, and the customer attraction force has a customer attraction coefficient for adjusting the size of the customer attraction force with respect to the sales floor area of the store, and a bottom-up coefficient for raising the sales floor area in advance by business type. It is characterized by being calculated by multiplication.

According to a fifth aspect of the present invention, in the invention according to any one of the first to fourth aspects, the simulation means is the exponent β of the sales floor area, the exponentiation α of the distance, the limit distance, the consumer business category share, The bottom-up coefficient and the maximum number of stores are designated in advance according to the business format of the store, and the power of the designated sales floor area β, the power of the distance α, the limit distance, the consumer business category share, the bottom-up coefficient, and the maximum number of stores are the probabilities. The model is set, and the customer attraction coefficient is determined by repeatedly performing simulation while correcting the customer attraction coefficient according to the immediately preceding simulation result for each store included in the target area. It was done.

According to a sixth aspect of the present invention, in the invention according to any one of the first to fifth aspects, the simulation means designates the number of simulations for the stores included in the target area, and the designated number of simulations is performed. After executing the, if the difference between the estimated sales value calculated based on the simulation result and the actual sales value stored in the store database is equal to or greater than a predetermined threshold value,
The store showing the difference is extracted, the customer attraction coefficient corresponding to the extracted store is determined at the time when the simulation is completed, and further repeated simulation is performed while changing the limit distance within a specified range. Thus, the limit distance is determined.

According to a seventh aspect of the present invention, in the invention according to any one of the first to sixth aspects, the simulation means designates the number of simulations for the stores included in the target area, and the designated number of simulations is performed. After executing the, if the difference between the estimated sales value calculated based on the simulation result and the actual sales value stored in the store database is equal to or more than a predetermined threshold value,
A store showing the difference is extracted, the customer attraction coefficient corresponding to the extracted store is determined at the time when the simulation is finished, and the power of the distance α is changed within a specified range, and the simulation is repeated. Is performed to determine the exponentiation α of the distance.

According to an eighth aspect of the present invention, in any one of the first to seventh aspects of the present invention, the simulation means designates the number of simulations for stores included in the target area, and the designated number of simulations is performed. After executing the, if the difference between the estimated sales value calculated based on the simulation result and the actual sales value stored in the store database is equal to or more than a predetermined threshold value,
The store showing the difference is extracted, the customer attraction coefficient corresponding to the extracted store is determined at the time when the simulation is finished, and the simulation is further repeated while changing the consumer business category share within a specified range. Is performed to determine the share of the consumer business category.

According to a ninth aspect of the present invention, in the invention according to any one of the first to eighth aspects, the simulation means sets the exponentiation β of the sales floor area and the exponentiation α of the distance to the stores included in the target area. , The customer attraction, the consumer business category share, and the maximum number of stores are specified in advance, and the power of the designated sales floor area β, the power of the distance α, the customer attraction, the consumer business category share, and the maximum number of stores are the probabilistic model. Is set, and the limit distance is determined by repeatedly performing simulation while changing the limit distance within a specified range in accordance with the immediately preceding simulation result for each of the stores. .

According to a tenth aspect of the present invention, in any one of the first to ninth aspects of the present invention, the simulation means, for a store included in the target area, the exponent β of the sales floor area, a limit distance, and Customer attraction, consumer business category share, maximum number of stores are specified in advance, and the specified power of sales floor area β, marginal distance, customer attraction, consumer business category share, maximum number of stores are set in the probability model. ,
It is characterized in that the exponentiation α of the distance is determined by repeatedly performing simulations while changing the exponentiation α of the distance within a specified range in accordance with the immediately preceding simulation result for each of the stores.

The invention of claim 11 is based on claims 1 to 10.
In any one of the inventions, the simulation means preliminarily sets the power β of the sales floor area, the power α of the distance, the customer attraction force, the limit distance, and the maximum number of stores for the stores included in the target area. Designate, set the exponentiation β of the designated sales floor area, exponentiation of distance α, customer attraction force, limit distance, and maximum number of stores in the probabilistic model,
It is characterized in that the consumer business category share is determined by repeatedly performing simulations while changing the consumer business category share within a specified range in accordance with the immediately preceding simulation result for each of the stores.

The invention of claim 12 is defined by claims 1 to 11.
In any one of the inventions, the simulation means stores the estimated sales value calculated based on a result of simulation using the probabilistic model for all the stores included in the target area, and stores the estimated sales value in the store database. The parameter of the stochastic model is determined by repeatedly performing simulation until the average of the difference with the actual sales value falls within a preset allowable range for the actual sales value.

The invention of claim 13 is based on claims 1 to 12.
In any one of the inventions, the simulation means includes at least one or more of a sales floor area, a sales floor efficiency, a sales form, and a location area concentration degree based on the store information stored in the store database for each business category. From the viewpoint, stores are extracted in advance, the extracted stores are classified as different business categories into different business category files, and the parameters of the probabilistic model are determined for each of the classified business category files.

The invention of claim 14 relates to claims 1 to 13.
In any one of the inventions, the simulation means makes a difference between an estimated sales value calculated based on a first simulation result for a store included in the target area and a sales actual value stored in the store database. Is greater than or equal to a predetermined threshold value, the stores showing the difference are extracted, the extracted stores are classified as different business categories into different business category files, and the parameters of the probability model are determined for each of the classified business category files. It is characterized by.

The invention of claim 15 relates to claim 13 or 1.
In the invention of claim 4, the simulation means has a small sales floor area and a sales record in a business category file classified as the different business category before executing a simulation for a store included in the target area. It is characterized in that a business category file to which a high store belongs is selected, and the sales floor area of the store belonging to the selected business category file is multiplied by the bottom-up coefficient in advance.

The invention of claim 16 is defined by claims 1 to 15.
In any one of the inventions, the simulation means performs a simulation on each of the existing stores included in the target area when the sales forecast of a new store in the target area designated by the target area designating means is performed. , The parameter of the probabilistic model, after determining to approach the actual sales value stored in the store database, each of the store name of the new store for which the sales forecast is performed, the sales floor area, the latitude and longitude indicating the location While setting data, the determined parameter is set in the probabilistic model, and a simulation is executed by using the probabilistic model in which the parameter is set to predict sales.

[0050] The invention of claim 17 is based on claims 1 to 16.
In any one of the inventions, the simulation means, based on a simulation result executed for an existing store or a new store included in the target area, a sales estimate or a predicted value of each of the existing store or the new store, It is characterized by generating an analysis result list including at least one of a sales actual value, a difference between an estimated sales value and a sales actual value, a customer attraction coefficient, a customer attraction force, and a limit distance.

According to an eighteenth aspect of the present invention, in the seventeenth aspect of the present invention, the simulation means stores a simulation result executed for an existing store or a new store included in the target area, and stores in the regional statistical database. It is characterized in that a sales estimate or forecast value for each consumption category in the household budget survey is generated for each of the existing store and the new store based on the statistical information.

The invention of claim 19 relates to claims 1 to 18.
In any one of the inventions, the simulation means is characterized in that the simulation is simultaneously executed for each store included in the target area.

The invention of claim 20 is based on claims 1 to 19.
In any one of the inventions, the simulation means uses a stochastic model that formally matches the Hough model.

The invention of claim 21 relates to claims 1 to 20.
In any one of the inventions, statistical information and store information corresponding to the target area designated by the target area designating unit are read from the regional statistical database and the store database, respectively, and based on the read statistical information and store information. The simulation means repeatedly performs simulation so that any one or more of the read statistical information, store information, and analysis result of the executed simulation can be layer-displayed on the map data. It is characterized by having an analysis plane creating means for creating an analysis plane as an edited data group.

The invention of claim 22 is based on claims 1 to 21.
In any one of the inventions described above, the regional statistical database stores statistical information about consumers for each mesh and / or for each town, and at the same time, stores data on household consumption, which is a survey of actual consumption of households and purchases nationwide. When reading statistical information from the regional statistical database, the analysis plane creating means reads household consumption data according to the statistical information to be read, and combines the read household consumption data with the statistical information to edit it. The analysis plane is created based on the edited statistical information.

The invention of claim 23 is based on claims 1 to 22.
In any one of the inventions, an analysis regarding the statistical information stored in the regional statistical database and an analysis regarding the store information stored in the store database is performed on the analysis plane created by the analysis plane creating means. An analysis unit is provided, and any one or a plurality of the statistical information, the store information, and the result analyzed by the analysis unit are layer-displayed on the map data having the target area designated by the target area designation unit. It is characterized by that.

The invention of claim 24 is based on claims 1 to 23.
In any one of the inventions, the simulation means is a store based on a simulation result executed for an existing store or a new store included in the target area on the analysis plane created by the analysis plane creation means. Another consumption probability contour line is generated, and the analysis means displays the consumption probability contour line generated by the simulation means in a layer on the map data, and based on the layer-displayed consumption probability contour line, the existing store or the new store is displayed. The feature is to analyze the trade area structure.

A twenty-fifth aspect of the present invention is characterized in that, in the twenty-fourth aspect of the present invention, the analyzing means can display the consumption probability contour lines in different colors according to the consumption probability.

The invention of claim 26 is based on claims 1 to 25.
In any one of the inventions described above, a polygon area generating means for generating an arbitrary polygon area defined by a user on the map data having the target area designated by the target area designating means, and the generated polygon area It has a polygon database storing attribute information including corresponding statistical information and store information, and the analysis means analyzes the attribute information stored in the polygon database on the analysis plane created by the analysis plane creation means. And then
The analysis result is displayed in layers on the map data.

The invention of claim 27 is based on claims 1 to 26.
In any one of the above-mentioned inventions, the map data is in a raster format or a bitmap format.

The invention of claim 28 is based on claims 1 to 27.
In any one of the inventions described above, the analysis result of the statistical information and / or the store information according to the area divided on the map data is displayed over the entire map data, and the analysis data is displayed. When a desired portion in the above is enlarged and displayed, it is enlarged and displayed in an arbitrary size together with the analysis result as a partial view of a frame different from the map data.

According to a twenty-ninth aspect of the present invention, in a trade area analysis system for setting a trade area on the map data and analyzing the set trade area, a map database storing the map data and the map data are specified under specified conditions. A region statistical database that stores statistical information about consumers for each of the divided regions, and desired map data is selected from the map data stored in the map database, and the target region for the commercial area analysis on the selected map data is selected. A target area specifying means for specifying, a store database storing store information including at least one of a store name, a sales floor area, a latitude and longitude indicating a location, and a sales result value for each business type, and the target area specifying means. Simulation calculation for estimating the commercial area spatial structure and theoretical sales for each store included in the target area And the one in which the parameters of the predetermined probability model, and characterized by having a simulation means for determining to approach the actual sales value stored in the shop database.

According to a thirtieth aspect of the present invention, in a trade area analysis method for setting a trade area on map data and analyzing the set trade area, the map data prepared in advance is consumed for each divided area under a specified condition. Regional statistics database that stores statistical information about employees, store name, sales floor area by business category,
A target area specifying step of specifying a target area for trade area analysis on the map data by preparing in advance a shop database that stores shop information including at least one of a plurality of latitude and longitude indicating a location and sales actual value; By performing a simulation calculation for estimating the commercial area spatial structure and theoretical sales for each store included in the target area, the parameters of the predetermined probabilistic model are approximated to the sales actual value stored in the store database. And a simulation step for determining.

According to a thirty-first aspect of the invention, in the thirtieth aspect of the invention, a store to be simulated is designated from the stores included in the target area, and simulation is performed for each of the designated stores. The parameter of the probabilistic model is determined so as to approach the actual sales value stored in the store database.

The invention of claim 32 is based on claim 30 or 3
In the first aspect of the invention, the difference between the estimated sales value calculated based on the result of simulation using the probability model and the actual sales value stored in the store database is preset for the actual sales value. By repeating the simulation until it is within the allowable range,
It is characterized in that the parameters of the probabilistic model are determined.

The invention of claim 33 is based on claims 30 to 3.
In the invention of any one of 2 above, the probabilistic model is a power β of a sales floor area of a store designated for each business type, a power α of a distance from a customer residence point to each store, and a simulation target included in the target region. Limit distance that represents the distance from the store to the customer's residential area that is located furthest away, and the attraction that attracts the customer at the store (attractiveness)
Of the customer attraction that represents the size of the store, the maximum number of stores that can be selected by the customer within the limit distance that can be selected by business category, and the total consumption category in the household survey that the stores of each business category deal with With respect to the consumption amount of all target categories per household, it is defined as a parameter with the consumer business category share that represents the ratio that a collective consumer can actually consume at a store of each business category as an average of one household. The product is calculated by multiplying the sales floor area of the store by a customer attraction coefficient for adjusting the magnitude of the customer attraction force and a bottom-up coefficient for raising the sales floor area in advance by business type. Is.

The invention of claim 34 is based on claims 30 to 3.
In the invention of any one of 3 above, the exponent β of the sales floor area, the exponentiation α of the distance, the marginal distance, the consumer business category share, the bottom-up coefficient, and the maximum number of stores are designated in advance according to the business category of the store, The specified sales floor area exponentiation β, distance exponentiation α, marginal distance, consumer business category share, bottom-up coefficient,
While setting the maximum number of stores in the probabilistic model, while repeatedly correcting the customer attraction coefficient according to the immediately preceding simulation result for each of the stores included in the target area, by repeatedly performing simulation, the customer attraction coefficient is calculated. It is characterized by making a decision.

The invention of claim 35 is based on claims 30 to 3.
In the invention of any one of 4 above, a sales estimation value calculated based on the simulation result after designating the number of simulations for a store included in the target area and executing the simulation of the designated number of times When the difference from the actual sales value stored in the store database is equal to or more than a predetermined threshold value, the store showing the difference is extracted, and the customer attraction coefficient corresponding to the extracted store is determined when the simulation ends. In addition to the above, the limit distance is determined by repeating the simulation while changing the limit distance within the specified range.

The invention of claim 36 is based on claims 30 to 3.
In any one of the inventions of 5, the number of simulations is designated for the stores included in the target area, the estimated number of times of simulations is performed, and then the estimated sales value calculated based on the simulation result, When the difference from the actual sales value stored in the store database is equal to or more than a predetermined threshold value, the store showing the difference is extracted, and the customer attraction coefficient corresponding to the extracted store is determined when the simulation ends. And the power α of the distance is determined by further repeating the simulation while changing the power α of the distance within a specified range.

The invention of claim 37 is based on claims 30 to 3.
In the invention according to any one of 6 above, a sales estimation value calculated based on the simulation result after designating a simulation count for a store included in the target area and executing the simulation of the designated count, When the difference from the actual sales value stored in the store database is equal to or more than a predetermined threshold value, the store showing the difference is extracted, and the customer attraction coefficient corresponding to the extracted store is determined when the simulation ends. In addition, the consumer business category share is determined by further repeating the simulation while changing the consumer business category share within the specified range.

The invention of claim 38 is based on claims 30 to 3.
In the invention of any one of 7 above, for the stores included in the target area, the power β of the sales floor area, the power α of the distance, the customer attraction, the consumer business category share, and the maximum number of stores are designated in advance. , The specified sales floor area exponentiation β, distance exponentiation α, customer attraction, consumer business category share, maximum number of stores are set in the probabilistic model, and the limit is set for each of the stores according to the immediately preceding simulation result. It is characterized in that the limit distance is determined by repeatedly performing simulation while changing the distance within a specified range.

The invention of claim 39 relates to claims 30 to 3.
In any one invention of 8, the power of the sales floor area β, the power of the distance α, the customer attraction force, the consumer business category share, and the maximum number of stores are designated in advance for the shops included in the target area. , Power of the specified sales floor area β, limit distance, customer attraction, consumer business category share, maximum number of stores is set in the probabilistic model, and the power of the distance is calculated for each of the stores according to the immediately preceding simulation result. It is characterized in that a power α of the distance is determined by repeatedly performing simulation while changing α within a specified range.

The invention of claim 40 is based on claims 30 to 3.
In the invention of any one of 9 above, for a store included in the target area, the exponent β of the sales floor area, the exponentiation α of the distance, the customer attraction force, the limit distance, and the maximum number of stores are designated in advance, and The specified sales floor area exponentiation β, distance exponentiation α, customer attraction, consumer business category share, maximum number of stores are set in the probabilistic model, and the consumer business category is set for each of the stores according to the immediately preceding simulation result. It is characterized in that the consumer business category share is determined by repeatedly performing simulation while changing the share within a specified range.

The invention of claim 41 relates to claims 30 to 4.
In the invention of any one of 0, sales estimation values calculated based on results of simulation using the probabilistic model for all stores included in the target area, and sales actual values stored in the store database. The parameter of the probabilistic model is determined by repeatedly performing a simulation until the average of the differences between the sales actual values falls within a preset allowable range.

The invention of claim 42 is based on claims 30 to 4.
In any one of the first aspect of the invention, based on the store information stored in the store database for each business type,
Stores are extracted in advance from the viewpoint of including at least one or more of sales floor efficiency, sales form, and location area concentration degree, the extracted stores are classified as different business types into different business type files, and each of the classified business type files is described above. It is characterized by determining the parameters of the probabilistic model.

The invention of claim 43 is based on claims 30 to 4.
In any one invention of 2, the difference between the estimated sales value calculated based on the first simulation result for the store included in the target area and the actual sales value stored in the store database is predetermined. When the difference is equal to or more than a threshold value, the stores showing the difference are extracted, the extracted stores are classified as different business categories into different business category files, and the parameters of the probability model are determined for each of the classified business category files. It is a thing.

The invention of claim 44 relates to claim 42 or 4
In the third aspect of the invention, before the simulation is performed on the stores included in the target area, a business category to which a store having a small sales floor area and a high sales record belongs in the business category files classified as the different business categories It is characterized in that a file is selected and the sales floor area of the store belonging to the selected business format file is multiplied by the bottom-up coefficient in advance.

The invention of claim 45 relates to claims 30 to 4.
In the invention of any one of 4 above, when the sales forecast of a new store in the target area is performed, the parameters of the probabilistic model are stored in the store database by performing a simulation for each of the existing stores included in the target area. After determining to approach the stored sales actual value, the store name of the new store for which the sales forecast is performed, the sales floor area, and the latitude and longitude data indicating the location are set, and the determined parameters are set as described above. It is characterized in that it is set in a probabilistic model and a simulation is executed by using the probabilistic model in which the parameters are set to forecast sales.

The invention of claim 46 is based on claims 30 to 4.
In the invention of any one of 5 above, based on a simulation result executed for an existing store or a new store included in the target area, a sales estimate or a predicted value, a sales actual value of each of the existing store or the new store, It is characterized in that an analysis result list including at least one of the difference between the estimated sales value and the actual sales value, the customer attraction coefficient, the customer attraction force, and the limit distance is generated.

According to a 47th aspect of the present invention, in the 46th aspect of the present invention, a simulation result executed for an existing store or a new store included in the target area and statistical information stored in the regional statistical database are provided. On the basis of,
It is characterized in that a sales estimate or a forecast value for each consumption category in a household budget survey is generated for each of the existing store and the new store.

The invention of claim 48 is based on claims 30 to 4.
The invention of any one of 7 is characterized in that a simulation is simultaneously executed for each store included in the target area.

The invention of claim 49 is based on claims 30 to 4.
The invention according to any one of 8 above is characterized in that a stochastic model that formally matches the Huff model is used.

The invention of claim 50 is based on claims 30 to 4.
In the invention of any one of 9 above, statistical information and store information according to the target area are read from the regional statistical database and the store database, respectively, and a simulation is repeatedly executed based on the read statistical information and store information. Of the read statistical information, store information, and analysis results of the executed simulation,
It is characterized in that an analysis plane is created as a data group processed and edited so that any one or a plurality of them can be layer-displayed on the map data.

The invention of claim 51 is based on claims 30 to 5.
0. In the invention of any one of 0, analysis on the statistical information stored in the regional statistics database and analysis on the store information stored in the store database are executed on the analysis plane, and the statistical information, the store Any one or more of the information and the result of the analysis and execution is displayed in layers on the map data.

The invention of claim 52 is based on claims 30 to 5.
In the invention of any one of 1, the consumption probability contour line for each store is generated on the analysis plane based on a simulation result executed for an existing store or a new store included in the target region, The consumption probability contour line is displayed on the map data as a layer, and the analysis of the commercial area structure of the existing store or the new store is performed based on the layer-displayed consumption probability contour line.

The invention of claim 53 is based on claims 30 to 5.
In the invention of any one of 2 above, an analysis result of statistical information and / or store information corresponding to an area divided on the map data is displayed over the entire map data, and the analysis result is displayed on the map. When a desired portion of the data is enlarged and displayed, it is characterized in that it is enlarged and displayed in an arbitrary size together with the analysis result as a partial view separate from the map data.

The invention of claim 54 is based on claims 1 to 29.
4. A computer for executing the function as the commercial area analysis system according to claim 1, or claim 3.
It is a program for causing a computer to execute the trade area analysis method described in any one of 0 to 53.

The invention of claim 55 is a computer-readable recording medium in which the program according to claim 54 is recorded.

[0089]

1 is a diagram for explaining a configuration example of a commercial area analysis system according to an embodiment of the present invention, in which 1 is a personal computer (PC) and 10 is a commercial area analysis system. The commercial area analysis system 10 has a DB editing unit 11, a simulation unit 12, and a display / analysis unit 13, and these units are mutually linked through the creation and delivery of data files. In addition, the commercial area analysis system 10 is a regional statistics DB that stores, as a database, regional statistical information 14a such as census mesh statistics and town claw statistics, and link data 14b such as household survey data.
14, store DB 15 storing store information 15a according to business type, attribute information 1 of polygon area defined by the user
It has a user polygon DB 16 in which 6a is stored and a map DB 17 in which map data 17a in raster format or bitmap format is stored in advance. This trade area analysis system 1
0 is installed in the internal or external storage device of the PC 1 and executed by the PC 1. The user can appropriately select and use desired map data from the map DB 17 when performing the trade area analysis.

The DB editing unit 11 defines a target area for trade area analysis (hereinafter referred to as an analysis area), reads various DBs,
It has an analysis area definition module 11a for providing the function of an input interface (I / F) for creating an analysis plane, and an analysis plane file creation module 11b for providing the functions of DB reading and editing and analysis plane file creation, The simulation unit 12 includes an input I / F module 1 for providing a simulation calculation input I / F function.
2a, analysis plane file read, simulation calculation of existing store, consumption probability contour line calculation of existing store, numerical table result write, analysis plane drawing file write (both existing store and new store) simulation calculation to provide functions It has an execution module 12b, a new store sales forecast, and a new store sales forecast module 12c for providing the function of an I / F for calculating the consumption probability contour line of the new store.
The analysis unit 13 is a startup module 13a that provides functions for software startup and drawing / analysis file reading,
It has a display / analysis module 13b for providing functions of analysis condition instruction on the drawing surface, the same analysis execution, and the same result screen display operation, and an output module 13c for providing an output function to printing and CSV files.

As the file format other than the map handled in this embodiment, for example, both CSV format and TXT format can be handled. Here, the statistical information stored in the regional statistics DB 14 is read and edited by the DB editor 11 in advance only for the data in the analysis area.
The analysis operation in can be performed extremely agilely. In addition, the DB editing unit 11, the simulation unit 12,
Each module configuration in the display / analysis unit 13 is not limited to the above configuration, and may be any module configuration as long as the function as the commercial area analysis system 10 can be executed.

Here, in the regional statistics DB 14, for example,
Census town census statistics, census mesh statistics (eg 5
(00m mesh, 1km mesh, etc.), establishment statistical mesh data, industrial statistical mesh data, commercial statistical mesh data, etc. There are various kinds in the store DB 15, but for example, a supermarket directory nationwide version CDRO
M version (commercial industry), NTT industry-specific town page data (address-latitude / longitude conversion required), user-owned customer D
B, store DB, etc. (with position latitude and longitude), etc., and the store DB 15 used in the simulation unit 12 is
It is possible to use store data of various business types according to the purpose. This data item requires the store name, the size of the sales floor area, the sales amount, and the latitude and longitude indicating the position.
Of these, the sales do not have to exist for all the stores, and can be simulated if some stores are known. The link data 14b linked to the above-mentioned regional statistical information includes, for example, the annual household survey report FD version, the single-person household income and expenditure survey report (because it is a printed matter, it is necessary to scan), and the national consumption survey report C.
There are DROM version etc. Further, the raster map used for the background can be arbitrarily selected according to the user's preference, but basically, for example, the latest version of the Increment P Company MAPDK National Edition, East Japan Edition, or West Japan Edition, etc. To use. One is required for each PC, and a partial print license is attached only as long as it is distributed to the same company.

The DB editing unit 11 reads the data of the range of the analysis area defined in advance when reading the statistical mesh data or the statistical data of the town-chou character from the local statistical DB 14,
In addition, a link file such as household survey data is read and edited, an analysis plane is created, and a file for layering the analysis plane is created on a raster map (a plurality of each scale) that has been cut out in advance.

The simulation unit 12 reads the store information from the store DB 15 corresponding to various business categories in the analysis plane created by the DB editing unit 11 so that the reality can be explained. In order to improve the prediction accuracy by
The simulation is executed by introducing the concept and variables as described later. The display / analysis unit 13 is used to draw the result as a layer on a raster map as a background drawing, and the arithmetic / analysis operation of geographical information and the like is freely performed and displayed on the screen of the PC 1.

The display / analysis unit 13 freely performs an operation and analysis operation of geographical information and the like in the analysis plane for layering on the raster map created in the DB editing unit 11, and displays the result on the screen of the PC 1. indicate. At the same time, the consumption probability contour lines for each store calculated by the simulation unit 12 can be analyzed and displayed. The display / analysis unit 13 is a DB editing unit 1.
If the calculation of 1 or the simulation unit 12 is executed in advance by the provider side of the commercial area analysis system 10, it can be used by the user side as independent software to be displayed / analyzed on the analysis plane. In this case, the country's mesh statistical data or town-choji statistical data are limited to those available at that time, but store data and customer DB
As for the polygon data defined by the user, the latest polygon data can always be imported and used on the screen of the PC 1.

In the present invention, the analysis plane is defined as vector data, but no other map engine or development kit is used to generate itself. The background map is a raster map, and when the analysis area is relatively small, it may be used by cutting it from a user's favorite electronic map, for example, as a BMP file. When the user analyzes over many areas, map data corresponding to the analysis area is cut out from a predetermined map DB (for example, MAPDK of Increment P Co., etc.) at any time and used as a BMP file. The trade area analysis system 10
It has a map data cutting function for that purpose and a function of indicating the latitude and longitude of an arbitrary point. At this time, when one analysis area becomes wide, the above-mentioned predetermined map DB and its raster engine are used, and only the maximum area (each maximum area is visually judged as the purpose of analysis,
Inputs to the DB editing unit 11 as an input.

Further, the display / analysis unit 13 analyzes geographical information such as descriptive statistics, index statistics, specialization coefficient, etc. from the viewpoint of area marketing problem, and also develops various user store data to store-mesh. Performs related analysis, and also performs zone analysis using the polygon generation function,
From these, merchandising policies, sales policies, city planning, etc. can be formulated in a market-friendly manner. In general, demographic factors (for example, age, gender, family composition, income, wealth situation, vocational industry, etc.) tend to have different products to buy, so the characteristics of the market can be displayed on a map using a mesh or town claw statistics. Analyzing and displaying the products and services that match the market characteristics will dramatically increase the market suitability of the policy. In addition, the products gradually penetrate the market according to the life cycle, such as market introduction period, growth period, and saturation period. A more appropriate regional sales policy can be set up by grasping the market penetration rate of the product in mesh or town-character format.

In another method of the GIS marketing analysis, the market may be subdivided by region / zone type. First, select appropriate items from the census items, perform mesh area classification, and classify into several large classification areas and more small classification zones. Next, develop various identified store data and perform zoning with a clearer character. Next, assuming that there is survey data regarding what products and services were purchased in the past for each region / zone type, the region / zone type is used as an indicator to identify the degree of penetration and its penetration. It can be indexed from sex. If a company or the like makes clear from the customer profile what type of region / zone the penetration rate is high and which type has a low penetration rate, the region / zone to be targeted in the future can be seen.

Further, the display / analysis unit 13 can solve the "weak competitive location problem" from the viewpoint of location problem. This is because the location of facilities is not very competitive and users are more likely to use nearby things (eg fast food, convenience stores, bank and non-life insurance company branch and dispenser placement issues, For public things, promptly solve hospitals, emergency medical centers, post offices, post offices, schools, libraries, community centers such as public halls, police stations, police boxes, fire stations, branch offices, branch offices, etc.). You can This kind of location problem simply occurs in the point buffer (circle) of location candidate points, or in the Voronoi division (method of dividing the market at the middle line between facilities), etc. It is often resolved by whether or not there is.

Simulation unit 12 in the present invention
Aims to substantially correct the shortcomings of the sales forecast calculation that lacks the process of explaining the reality by the conventional stochastic model, and to solve the "strong competitive location problem" where there is intense competition between facilities over a wide range. .

Here, an example of a system configuration that the commercial area analysis system 10 can take is as follows according to its purpose of use. (1) For geographic statistical data, mesh statistics or town-chome statistical data are used, combined with household survey data as a link file, and based on these, geographic information analysis, store information analysis, and user-created polygon information analysis on an electronic map. In the case of, the combination of the DB editing unit 11 and the display / analysis unit 13 is used. (2) For geographical statistical data, use mesh statistics or town-chome statistical data, combine household survey data, etc. as a link file, and introduce a new concept into a probabilistic model as described below on the analysis plane based on these. Then, the calculation of the commercial area spatial structure and the calculation of the attractiveness (customer attraction for individual stores) using the sales and consumption contours by store are repeated until they match the actual sales of each store in the analysis plane. , Furthermore, after that, the same prediction calculation is performed for the new store in the analysis plane, the consumption probability contour lines of those results are displayed on the map, and when geographical information analysis and store information analysis are also performed on the electronic map,
It is a combination of the DB editing unit 11, the simulation unit 12, and the display / analysis unit 13.

As the usage pattern of the commercial area analysis system 10, for example, the user side rents the commercial area analysis system 10 for a certain period of time, or the system provider side explains in advance about the analysis area the user side wants to analyze. After performing a simulation analysis and then performing a sales forecast and a trade area structure analysis when the user side is interested in the location point of any scale, display the result file and the display / analysis for analyzing the file. It is possible to use the part 13 as a set for sale.
In the former case, the user side can scrutinize a wide area in the whole country, and in the latter case, the user side has a merit that the marketing analysis is immediately started from the day of delivery, together with a certain conclusion regarding a specific store investment.

Here, the commercial area analysis system 10 according to the present invention is, for example, SM, discount store, G
Although a food-centered business category such as MS is taken up as an example, if there is store data such as store name, business area, address, etc. and sales data for some of the stores, a simulation described below. Can be implemented in any competitive format. In addition, the trade area analysis system 10 uses a Hough model as a representative example for a probabilistic model related to a competitive location problem to show the ability to explain reality and at the same time to predict sales of a new store and predict the trade area spatial structure. It changes the interpretation of the type Hough model, introduces new concepts and variables, and provides a modular structure that follows those concepts and variables.

In order to execute the calculation by the module configuration as described above, a dedicated analysis plane is created in the PC 1, and all mesh grid points or meshes are further subdivided on the analysis plane, and the mesh data is The method is to calculate a probabilistic model on all subdivided grid points and repeat it until it fits the actual data (for example, sales record of individual store and structure of commercial area). In addition, the analysis plane created in this way is displayed on the raster map in turn, and the geographical information analysis operation can be performed on the display screen of the PC 1 for display.

First, in order to put the probabilistic model into practical use, the concept of "limit distance" based on observation facts is introduced.
This is due to the fact that the distribution of customers at any store is always accompanied by a marginal distance, which is a vanishing line. Empirical facts indicate that this marginal distance depends on the type of business and size of the store, and at the same time on the density of the urban space. Even for large stores such as GMS, the limit distance is short inside a big city, and even for medium and small stores, the limit distance can be extended to the suburbs. This limit distance is set as an input item for each business type file together with the power of the distance α.
However, after the initial simulation, if it is found that there is a particularly strong store in the analysis area, it is decided whether to classify the store into another business category file or move to a simulation in which the limit distance is gradually increased for that store. You can choose.

Next, the sales floor area is an index showing the degree of attractiveness or attraction of attracting customers, but if the same business format is true, the larger the sales floor area, the better the product lineup, and the consumers. It can be said that the recognition level of will increase. If so, if the same location space, that is, the same analysis area, the sales of stores with the same sales floor space
The values should be similar, and so should the m ² efficiencies. However, for example, SM
If you look at the store database of, you can see that there are enormous variations. In general, the m ² efficiency is higher for SMs with smaller sales floor areas, and gradually decreases as the sales floor area increases. GMS, which includes clothing and household goods in addition to grocery items, is also an extension of this trend.
The m ² efficiency of S is much smaller than that of SM.

On the other hand, the empirical fact that the above-mentioned limit distance is shorter as the sales floor area is smaller and longer as the sales floor area is larger is added in the same business category. Consider how to handle these seemingly contradictory reality with the same Huff model. This is because from the consumer's perspective, all small SMs and GMSs that are within the limit distance are targets for selection, and in order to simulate reality, they have to be handled in the same model. .. The same thing can be said, for example, for drug stores, DIY products, casual wear, etc. Depending on how to set up a problem, competition between different industries and different business formats must be handled by the same model.

Therefore, in the simulation unit 12, the structure of the data is checked in advance in the store DB 15, and not only the product composition handled and the sales method but also the sales floor area and m ²
You can categorize shops with similar efficiencies and define them as different business categories, or look at the first calculation result in a simulation run and see an abnormal value that is far from the estimated value and the actual value even in the same file. We introduced the concept of "business format file", which allows you to arbitrarily define a store that indicates as a different business format. And, in order to raise the sales floor area (suction power) of a store belonging to a business format file that is compact and highly efficient, it is possible to set a "bottom raising coefficient" as an input so that it can be raised a fixed number of times.

In a store with such a business format file,
The power α of the distance in the model can be specified for each business category file in consideration of the possibility that it differs for each business category.

Next, even in a store in the same business format file defined in this way, the sales force of each individual store is, of course, the sales floor area M or the floor-up sales floor area d × M.
Since there is a difference more than that shown by, "Individual store customer attraction coefficient"
The concept of (hereinafter referred to as γ coefficient) was introduced, and it was defined as the following expression (1). Customer attraction = Sales floor area (M or d × M) × Individual store customer attraction coefficient (= γ coefficient) Equation (1) Therefore, if the Huff model equation is used, the following Equations (2) and (3) are obtained. Like

[0111]

[Equation 4]

Here, β in the above equation (2) is a power exponent for the area, which is also set as an input for each business type file. However, in most cases β = 1.0
Then the simulation ends successfully.

Furthermore, in order to simulate reality,
It is necessary to introduce some concepts and variables. First, the "maximum number of stores that can be selected by business type" counted from the residential points that consumers living at each point on the analysis plane can select within the limit distance
The display / analysis unit 13 displays the store point flag and confirms it on the screen to determine. This is an input item.

Next, "consumer business category share" is introduced.
This is the average value per household for the period consumption consumed by collective consumers in all the categories in the household survey data targeted by the stores in each business type file being dealt with. It indicates what percentage of the total amount of consumption of all categories that the store in the business category deals with. This is within a certain range depending on the analysis domain, that is, by the sticking situation of the stores, and by what kind of stores are ignored in the construction of the competition model, and whether or not single households are taken into consideration or ignored. Changes. At this time, the appropriate value can be estimated by comparing the estimated sales of each store with the actual sales in one calculation. This is an input item and is set for each business type file. In addition, if a store that shows a particularly strong sales performance than that estimated from theory is found from the results of the simulation, it is possible to change this to another business format file and then change the business category share.

As described above, it is possible to set parameters such as the power of customer attraction β, the power of distance α, the limit distance, the bottom-up coefficient, the business category share, and the maximum number of selectable stores for each business category file DB. Suction coefficient γ (where
The initial value of the γ coefficient is automatically set to a predetermined value, but is set to 1.0 by default. ) Is variable, it is possible to calculate competition between stores of different sizes from small SM to GMS, or competition between different industries of the same consumption category under the same model. . In addition, in order to be able to handle more complex situations, options such as a process that makes the limit distance a variable, a process that makes the power of the distance α a variable, or a process that makes the consumer business category share a variable are included. Is. Furthermore, as for these parameters defined for each business type file, there are prepared options that can be set for each individual store of the database file.

Here, the calculation formula is shown as the following formula (4). At any point in the analysis area, the collective consumer selects all selectable stores within the limit distance by business type, and selects and consumes the stores with the probability indicated by the Hough model. The sum of the selected consumption probabilities of all the stores selected at this time is standardized to be 1. This saves all consumption, ensuring consistency in sales forecasts for new stores.

Here, for simplicity, SM and GMS are used.
It is assumed that there is a business category file (actually, these are further divided into several files), and one consumer whose distance from the store of interest for each business category is R1 and R2 is within the limit distance for each business category. The maximum number of stores is selected, and the sum of all selected consumption probabilities becomes 1 for each point.

[0118]

[Equation 5]

Next, the consumer business category share of each business category is Φs
m and Φgms, and the SM and GMS assortments correspond to the period consumption amount per household (on household survey data), Csm and Cgms, respectively, and the maximum consumer Φsm in each business category. × Csm, Φgm
Only s × Cgms is consumed. Sales Ui brought by the consumer at a distance R1 from the SM store i to which he is interested
(R1) is represented by the following formula (5).

[0120]

[Equation 6]

A in the above equation (5) is a normalization constant,
It is calculated at each point by the following equation (6).

[0122]

[Equation 7]

Therefore, the theoretical total sales estimation or prediction in the store is expressed by the following equation (7).

[0124]

[Equation 8]

Here, Pi (R
1) takes values of various probabilities from 0.00 to 1.00 depending on the position and value of R1. Further, n (R1) is R
The density of the number of households in a small area represented by one point and having a probability P, ΔR1 is the area of the small area, and n (R1) Δ
R1 = N (R1) is the number of households in a small area represented by the R1 point and having a probability P.

The specific calculation is not integration, but all bands from 1.00 to 0.05 (however, this minimum probability can be changed) along the width of the band-shaped contour probability contour line. Done. This Ui is the theoretical estimated or predicted sales of the i store and is compared with the actual sales of the i store.

The above calculation is performed by software as follows. First, the probability of each store is calculated on all mesh grid points in the analysis area or on subdivided grid points obtained by further dividing the mesh into n equal parts, and the equal probability contour lines are calculated. The suction sales are calculated for each band of the nine probability contour lines having the consumption suction probability of 0.1 to 0.9. For example, for probability 0.1, 0.05-
The probability of the area surrounded by the band of 0.15 is 0.1, and the total number of households in this area is calculated. If the sales within this obi are added from 0.1 to 0.9, it becomes the total sales of the category of the store. (However, the method of dividing this probability band can be changed.)

Further, the simulation section 12 is designed so that the simulation can be performed in the following five ways. (1) The γ coefficient is determined by simulation calculation.
That is, until all the actual sales of all the stores existing in the analysis plane or all the stores specified in the convergence calculation and the estimated sales of the stores calculated by the model are within the error% specified in advance, γ It is calculated repeatedly while modifying the coefficient in a certain way. The γ coefficient is a variable determined by calculation and is not an input item. in this case,
The limit distance is specified for each business type file. (2) First, the γ coefficient is calculated to some extent by designating the number of simulations for determining the γ coefficient, and the γ coefficient is relatively large or small,
For stores where the difference between the estimated value and the actual value still shows a large difference, for those stores, the γ coefficient is fixed there, and this time, the limit distance of those stores is changed by the specified increment / decrement. Then, the simulation is repeated to bring the difference between the estimated sales and the actual sales close to the specified error range. That is, the simulation is performed in the first half to determine the γ coefficient, and in the second half to determine the limit distance for the newly designated store, and for the other stores to determine the γ factor. (3) The sales floor area or the γ coefficient is fixed from the beginning or in the middle of the simulation, and for the specified business category file, the limit distance is specified to be variable, and the simulation is performed to determine it. (4) The sales floor area or the γ coefficient is fixed from the beginning or in the middle of the simulation, and with respect to the specified business category file, it is specified that the exponentiation α of the distance is made variable.
Perform a simulation to determine it. (5) The sales floor area or the γ coefficient is fixed from the beginning or in the middle of the simulation, and for the designated business category file, the business category share is designated as a variable, and the simulation is performed to determine it.

In the following, the case of the above (1) executed to determine the γ coefficient will be mainly described. The maximum allowable value (%) of the ratio (%) of the difference between the theoretical value and the actual value of the store sales to the actual sales (when it becomes less than that, the convergence calculation is finished) is an input item, and the "convergence condition for the store"". In addition, the sales floor area (hence the γ coefficient) calculated one time ago is corrected by the percentage of γ.
Input the "sales floor area correction ratio" that specifies whether to calculate the coefficient. In addition, there is an input item that specifies the percentage of the average sales-estimation error of all stores specified for the convergence calculation to end the simulation, and it is specified as the "convergence condition regarding the end". In addition, separately from this, the number of times the simulation is terminated can be specified by an input item.

Here, the empirical value of the customer attraction coefficient (γ coefficient) in the competition for SM and GMS by simulation reflects the strength of each store in the case of GMS, and is 0.50 to 1.1 in most cases. It converges in the range of 50. Once the analysis area is determined, the customer attraction is relatively determined within that range, so the γ coefficient should take any reference value, but since it starts from the actual sales floor area, it will converge within the above range. Will be done.

The output result of the simulation is not limited to the theoretical sales-actual sales comparison table of the existing store, the customer attraction coefficient (γ coefficient), and the sales forecast value of the new store. Layer drawing on the raster map in the form of "consumption probability contour lines" (20 lines in 5% increments by default from 0.05 to 1.00, but can be changed arbitrarily) in the form of a convergence area calculated for each existing store It is also possible to show it on the analyzed plan view. This is done by delivering the file to the display / analysis unit 13. By displaying data of various mesh DBs in the mesh on this analysis plane drawing and adding it using the "addition / subtraction module", it is possible to grasp the number of consumers by attribute of each store by consumption probability. . As a result, each store will be able to establish a target-specific sales policy.

Further, if various geographical information analyzes are used together, the characteristics of the commercial area of each existing store can be grasped from another angle. This expands the estimated sales into sales estimates by category and if you use a window that displays the amount of sales that can be classified by item and consumption category by the same probability band, you can compare this with the actual sales by category. By doing so, it will be possible to analyze the strengths and weaknesses of the business of own stores by category, and it will be possible to utilize it for the floor plan of new stores.

[0133] In addition to the display on the map, for example, CSV
As a separate table of formats, a table is created with the following items. (1) Store type, theoretical sales, actual sales, ratio of difference (%), sales floor area, bottom-up coefficient, customer attraction coefficient (γ)
Coefficient), customer attraction, marginal distance, category consumption, business category share, distance power (α), area power (β) (2) Population by consumption probability zone by store, number of households with 2 or more persons, number of single-person households , Theoretical sales amount, ratio of the theoretical sales amount (%) (3) Population by gender by age group, probability by population, and number of households with two or more households by age group Value, the number of households multiplied by the probability, the number of single households by age group, the number of single households multiplied by the probability (4) The population by occupation, and the population multiplied by the probability by store probability band (5) Number of households by type of residence, number of households by size of residence, by probability band by store

FIG. 2 is a diagram showing an example of a theoretical total sales list according to the present invention. In the figure, 20 is a theoretical (estimated) total sales list, and the theoretical total sales list 20 is a store name 21, Business type 2
2, theoretical monthly sales 23, all departments monthly sales 24, monthly sales relative error 2
5, directly managed sales surface 26, customer attraction coefficient 27, customer attraction power 2
8 has a limit distance 29 and a store number 30. All department monthly sales 24 are subdivided into the sales results of food monthly sales, clothing monthly sales, and other monthly sales by the department of the product handled by each store. You can also do it. This theoretical total sales list 20
Is calculated as a result of executing the simulation as described above.

Here, the business format 22 is gms (sales area 6000 m ² or more), s depending on the size of the directly-managed sales surface 26.
m1 (floor space 1000~1999m ^2), sm2 (floor space 2000~5999m ^2), sm3 (sales floor area 4
50-999 m ² ) etc. The limit distance 29 is
Determined according to this business category 22, and in the case of the present embodiment, g
7000m for ms, 3500m for sm1, 50 for sm2
00m is specified respectively. Customer attraction coefficient 27
Is determined by the simulation executed in the simulation unit 12, and the customer attraction force 28 is determined based on the customer attraction coefficient 27. Furthermore, theoretical monthly sales 2
3 is calculated by the formula (7) described above, and the monthly sales of all departments 2
Compared with 4.

Here, the monthly quotient relative error 25 indicates the ratio of the difference between the theoretical monthly quotient 23 and the total monthly quotient 24 with respect to the total monthly quotient 24 of all the departments, and this ratio is within a predetermined allowable range. The simulation is repeatedly executed while correcting the parameters such as the customer attraction coefficient so as to enter. In this way, the parameters of the probabilistic model for accurately calculating the theoretical monthly quotient 23 in the theoretical total sales table 20 are determined, and the probabilistic model that has determined the parameters is used to predict the sales of the new store and the structure of the commercial area. By conducting the analysis, more accurate marketing analysis becomes possible.

FIG. 3 is a diagram showing an example of a commercial area analysis-regional map list according to the present invention. In the figure, 40 is a regional map list, and the regional map list 40 is a numerical display screen selection 4
1, screen display method selection 42 of whole view, partial view display grid number 4
3. It has an entire area map list 44 as input items. Here, the numerical display screen selection 41 is the local statistics DB1.
This is an item for selecting whether to display the numerical data acquired from 4 or the like on the whole view or the partial view. The screen display method selection 42 of the overall view is an item for selecting whether to match the map display to the overall view or to display a scroll screen. The partial view display grid number 43 is an item for designating the grid numbers for each of longitude and latitude when the partial view is displayed in another frame. The entire area map list 44 displays a list of currently available raster map data (bmp format), and the user selects and uses desired map data from the list. In this example, "\ bmp \ Chofu 3km.bmp" is selected.

FIG. 4 is a diagram showing an example of a commercial area analysis-all shops and characteristic table according to the present invention. In the figure, 50 is all shops and characteristic table.
Selected in the group selection column 51 for selecting a group classified by business category according to the sales floor area such as MS and SM, the store display column 52 for displaying the stores belonging to the selected group, and the group selection column 51 It has a characteristic table 53 for displaying a list of characteristic information about each store belonging to the group. First, the user selects a desired group in the group selection column 51, and selects the stores to be displayed on the map data among the stores belonging to the selected group in the store display column 5
Select by checking the 2 check box.
Further, the characteristic table 53 includes, for example, a store number, a store name, a store location, a directly managed suction sales surface (customer suction power), a directly managed suction coefficient (customer suction coefficient), a directly managed sales surface, a business category, a monthly average sales amount, and a food sales composition. It is a list of various data relating to stores such as ratio, clothing sales composition ratio, other sales composition ratio, m ² sales (monthly), data on the number of parking lots and buildings, number of employees, latitude and longitude.

FIG. 5 is a diagram showing an example of a trade area analysis-contour line list according to the present invention. In the figure, 60 is a contour line list, and the contour line list 60 is 1.00 for a band-like probability contour line. The probability contour line selection column 61 for selecting the contour line band to be displayed on the map data in the range of to 0.05, the store selection column 62 for selecting the store for displaying the probability contour line on the map data, and the screen for displaying the probability contour line ( It has a screen designation 63 for designating a whole view or a partial view). At this time, it is possible to specify the line color and line thickness for the probability contour lines to be displayed on the map data.
It is also possible to specify the color of the probability contour line for each store.

FIG. 6 is a diagram showing an example of a trade area analysis-item list according to the present invention. In the figure, 70 is an item list, and the item list 70 is a linear combination designation 71, index creation 72,
It has an item list 73 as an item. Here, the item list 73 is a list of items of statistical information stored in the regional statistical DB 14, and the user selects the statistical information (numerical data) to be displayed on the map data by checking the check box. . In this example, No. 27
"2 or more general households" is selected, and the numerical data corresponding to this item is displayed on the map data in the specified mesh unit or town chome character unit. Linear combination designation 71
Creates a linear combination (sum) or product of the items checked in the check boxes. Also, index production 7
In item 2, the item checked in the check box is divided in units of lattices to create an index.

FIG. 7 is a diagram showing another example of the trade area analysis-item list according to the present invention. In this example, in addition to the items shown in FIG. 6, No. 25 “the number of single households aged 30-64” and No. 26 “the number of single households aged 65 or older” are selected. Numerical data corresponding to the item is displayed on the map data in a specified mesh unit or a town-kanji character unit. As described above, as long as the items are registered in the item list 70, all or some of them may be designated, and any item may be appropriately set according to the analysis purpose of the user.

FIG. 8 is a view showing an example of a trade area analysis-contour line area setting screen according to the present invention. In the figure, 80 is a contour line area setting screen, and the contour line area setting screen 80 is
Inter-contour line area display selection field 81, a contour line value specification field 82 for designating the value of the contour line (consumption suction probability) to be displayed on the map data from the nine contour lines of 0.1 to 0.9, displayed on the map data It has a list 83 of store names. Here, when the contour line area display selection column 81 is checked, the contour line area is displayed on the map data. As this display method, the ratio (%) inside the contour line designated in the contour line value designation column 82 is displayed. It can be calculated and displayed, or can be displayed by multiplying the display item by the area ratio, or can be displayed by multiplying the display item by the area ratio and the probability. Furthermore, the statistical information registered in the item list 70 shown in FIG.
No. 27 Consume by performing consumption amount calculation by consumption category at the store selected in the store name list 83 based on the condition "number of general households for two or more persons" and the analysis result by the consumption probability contour line for each store You can make sales estimates or forecasts by category.

FIG. 9 shows the area map list 40 shown in FIG.
In the figure which shows an example of the map data selected by
Reference numeral 90 is map data, the map data 90 is mesh-divided in a grid pattern, 91 is a grid number in the longitude direction, and 92 is a grid number in the latitude direction.

FIG. 10 is a map data 9 showing the probability contour lines of each store selected in the store selection fields shown in FIGS.
It is a figure which shows an example at the time of displaying on 0, and 93 is a probability contour line in the figure. In the map data 90, all the stores shown in FIG. 4 and the stores with store numbers 3700, 4267, and 4919 selected in the characteristic table 50 are displayed on the map data 90, and the store with the same store number is selected in FIG. By doing so, store numbers 3700, 4267,
Probability contour lines corresponding to each of the 4919 stores are displayed. Here, in order to simplify the description, the store for which the probability contour lines shown in FIG. 5 are displayed is designated as the same store as in FIG. 4, but the store selection in FIG. 4 and the store selection in FIG. It doesn't have to match.

FIG. 11 is a diagram showing an example in which numerical data corresponding to the item selected in the item list shown in FIG. 6 is displayed on the map data 90. In the figure, 94 is numerical data. is there. The numerical data 94 is numerical data corresponding to No. 27 “Number of general households for two or more persons” selected in the item list 70 shown in FIG. 6, and is displayed in mesh units.

FIG. 12 is a diagram showing an example in which numerical data corresponding to the item selected in the item list shown in FIG. 7 is displayed on the map data 90. In FIG. 12, 95 is numerical data, Reference numeral 96 is a detailed data list regarding the numerical data 95. This numerical data 95 is the item list 7 shown in FIG.
Numbers corresponding to No. 25 “Number of single households aged 30-64”, No. 26 “Number of single households aged 65 and older”, No. 27 “Number of general households of 2 or more” selected in No. 0 It is data and is displayed in mesh units.
At this time, by clicking the desired mesh data, the detailed data 96 is displayed in another frame.

FIG. 13 shows the map data 90 shown in FIG.
In the figure which shows an example when it is enlarged and displayed as a partial view,
In the figure, 97 is a partial view. This example shows a case where the area around the shop of shop number 4919 is enlarged and displayed as a partial view in a frame different from the map data 90.
7 is enlarged and displayed in a window (4 × 4 mesh) separate from the map data 90, and numerical data 95 is written in the partial view 97.

The display / analysis unit 13 in the commercial area analysis system 10 of the present invention has a function of directly writing the statistical data or the like in a polygonal area such as a mesh unit or a town-character unit over the entire screen. The number of data items to be written at one time is a selection formula, and can be selected as many times as the user can visually distinguish each time the data items are displayed. By the way, this function is not found in the existing GIS engine. In the existing GIS engine, for example, data is linked to each polygon, and when the polygon is clicked, the data content of the polygon is opened in a separate window in a card format. This method is called a one polygon card display method. When looking at the whole from a bird's-eye view, a method is adopted in which polygons are color-coded according to layers for specific 1 or 2 terms. In the present invention, all the selected plural data are directly written and displayed in the polygon, and the shape and color of the font of the character can be selected. And the data can be calculated by defining an arithmetic expression on the display screen. Therefore, it is possible to analyze the entire analysis region in a bird's-eye view with raw data or data that has been arbitrarily processed.

However, for example, 50 km square or 1
When a wide area such as a square of 00 km is to be analyzed, the polygon area becomes too small to directly check the written data. The partial view 97 shown in FIG. 13 is a function for making it possible to confirm the data in such a case. In this example, when the user clicks a point on the map data 90 where numerical data or the like is to be confirmed, the screen size and the number of meshes specified as default (4 × in this example)
In 4), a partial view 97 centering on that point opens in a frame separate from the overall view. This partial view 97 is a copy of that part from the overall view into a partial view frame, and numerical data 95 is written to this partial view 97. Even if numerical data and the like are included in the overall view, writing is performed again in the partial view 97. By doing so, various arithmetic processes such as addition and subtraction can be performed on the partial view 97. This operation is apparently executed instantaneously, so when the partial view 97 is opened, the numerical data 95 is included.

As another method of designating the range of the partial view, the area in the overall view where the numerical data is desired to be viewed is
If you drag from the upper left to the lower right, the map data of the range is copied to the enlarged partial drawing frame, it opens in a separate frame from the overall drawing, then the numerical data is written in that partial drawing and the operation is performed. Be subject to. Also, store data can be enlarged and displayed using this partial view function.

Also, the partial view function can be used for normal zoom-in,
The advantage that is different from zoom out is that you can enlarge the point you want to see while putting the whole diagram in front and perform data calculation. In the case of zooming in and out, the positional relationship in the overall view is unknown, and the partial analysis of the entire view cannot be performed as desired. This partial view feature is also not found in existing GIS engines.

FIG. 14 is a diagram showing an example of a calculation result of a trade area analysis-category consumption amount calculated based on the area setting between contour lines shown in FIG. 8. In the figure, 100 is a category consumption amount calculation. The result. This category is determined based on, for example, the consumption category classified on the household survey data based on the “Household Survey Annual Report” that the Statistics Bureau of the Ministry of Internal Affairs and Communications carries out every year. This “Household Survey Annual Report” is a sample survey of household consumption for average households (excluding agriculture, forestry and fishery households, and single-person households) across the country. The calculation result 100 of the consumption amount by category is calculated for each store. In this example,
As the statistical information registered in the item list 70 shown in FIG. 6, based on the condition No. 25 “the number of households with one person aged 30-64” and the analysis result by the consumption probability contour line for each store, It shows the result of executing the consumption amount calculation for each consumption category in the store selected in the store name list 83 of the area setting screen 80 to estimate or predict the sales amount for each consumption category.

FIG. 15 is a flow chart for explaining an example of a processing procedure for designating an analysis area, reading map data and regional statistical data, processing and editing, and creating analysis plane data to which the present invention is applied. First, a map data file prepared in advance is set (step S1), and position information of each map data and latitude / longitude per km at that position are set (step S2). Next, a larger analysis area (calculation area) that encloses the largest map data is specified (step S3), and the regional statistical DB to be analyzed
Is designated (step S4), and what kind of analysis is to be performed on the designated regional statistics DB is selected (step S4).
5). Next, the map data file is read (step S6), and the regional statistical data in the analysis area is read from the regional statistical DB (step S7). Furthermore, it is determined whether or not to create the simulation analysis plane data (step S8), and when the analysis plane data is created (YES), the simulation calculation analysis plane data is created (step S9) and is displayed on the raster map. An analysis plane is created as a file to be layered (step S10). When the analysis plane data is not created (in the case of NO), the analysis plane as a file to be layered on the raster map is created (step S11).

FIG. 16 is a flow chart for explaining an example of a processing procedure in various settings for a simulation to which the present invention is applied. First, the purpose of the simulation (whether to calculate the existing store sales from the beginning, to take over the intermediate results, to predict the new store sales,
Etc.) is designated (step S21), and the number of times of repeated simulation calculation is designated (step S22). Next, the mesh division number N is designated (step S23), the business category file DB is designated (step S24), and the bottom-up coefficient is set for each designated business category file (step S2).
5). Next, the maximum selectable number of stores, the limit distance, the exponentiation α of the distance, and the exponentiation β of the sales floor area are set for each business type file (step S26). Further, the category consumption amount and the consumer business category share are set for each business category file (step S27).

Next, it is designated whether this simulation is an existing store convergence calculation or a new store sales forecast (step S
28) In the case of the new store sales forecast, the unique number of the new store to be predicted is set in addition to the unique number of the existing store (step S29). In the case of the existing store convergence calculation, the unique number of the existing store for which the convergence calculation is performed is set for each business type (step S30).

FIG. 17 is a flowchart for explaining another processing procedure in various settings for simulation to which the present invention is applied, and is a continuation of the flowchart shown in FIG. First, step S3 shown in FIG.
After setting the unique number of the existing store to be converged at 0 for each business type, what is used as a variable, the γ coefficient from start to finish, the limit distance from the beginning or the middle, the power of the distance α, the business form share, Etc. are designated (step S31), and an allowable limit% of the theory-actual sales difference by store is set (step S3).
2). Next, in the case where the γ coefficient is used as a variable in step S31, the correction% of the sales floor area in the store outside the allowable limit for each time
Is specified (step S33), and when the limit distance is used as a variable, a correction distance (meter or the like) for each limit distance in a store outside the allowable limit is specified (step S34), and the exponentiation α of the distance is used as a variable. In this case, the correction value for each power of the distance α in a store outside the allowable limit is specified (step S
35) When the business category share is used as a variable, the corrected% value for each business category share in the store outside the allowable limit is designated (step S36). Further, the allowable upper limit% of the simple average of the differences by stores is set as the convergence determination condition for the entire convergence calculation (step S37).

FIG. 18 is a flow chart for explaining the processing procedure of the existing store simulation to which the present invention is applied. First, the analysis plane data for simulation is read (step S41), the business condition file is read (step S42), and each data input for simulation is read (step S43). Next, for the γ coefficient, the limit distance, the power of the distance α, or the business category share, the simulation is executed using the set value for the first time and the modified value for the second time and thereafter (step S44), and the existing store consumption probability contour line is obtained. Is generated (step S45), and theoretical sales of existing stores are calculated (step S46). Next, the individual store individual convergence determination is performed (step S47), and if converged, the convergence determination based on the average difference of all existing stores is performed (step S48), and if converged, various numerical results are written to the file (step S49). . Step S47,
If it does not converge in step S48, the process returns to step S44 to repeat the process.

Next, the convergence result analysis plane data file in which the various numerical results have been written in the above step S49 is saved (step S50), a new store sales forecast store DB file is created (step S51), and the drawing layer analysis plane is created. A file is created (step S52).

Here, the following items are written in the numerical result file as a list. (1) Store type, theoretical sales, actual sales, ratio of difference (%), sales floor area, bottom-up coefficient, customer attraction coefficient (γ)
Coefficient), customer attraction, marginal distance, category consumption, business category share, distance power (α), area power (β) (2) Population by consumption probability zone by store, number of households with 2 or more persons, number of single-person households , Theoretical sales amount, ratio of theoretical sales amount by consumption probability (%) (3) Population by gender by age group, by probability band by store, numerical value obtained by multiplying the population by probability, two or more persons by household head age group Estimated number of households, numerical value obtained by multiplying the estimated number of households by probability, number of single-person households by age group, numerical value obtained by multiplying the number of single-person households by probability (4) Population by occupation, by population Numerical value multiplied by probability (5) Number of households by type of residence, number of households by size of residence, by probability band for each store. Here, some probability contour lines are included in a 500m mesh depending on the location. Calculate the area enclosed by the mesh line and distribute the number of mesh households by area Calculated to. In the estimation simulation store DB file, the sales floor area of the existing store is replaced with the customer attraction.

FIG. 19 is a flow chart for explaining the processing procedure of the new store sales forecast simulation to which the present invention is applied. First, step S51 shown in FIG.
Information about the new store is input to the new store sales prediction store DB file created in step S61. In particular, how much the γ coefficient of the new store is superior to the γ coefficient of the existing competitive store in the commercial area is set by subjective judgment. The setting parameters for new store sales forecast set in steps S21 and S29 are read (step S62),
The convergence result analysis plane data file of the existing store saved in step S50 shown in FIG. 18 is read (step S
63), and reads the store DB file for forecasting new store sales to which the new store information was input in step S61 (step S).
64). Next, the sales forecast calculation of the existing store and the new store is performed (step S65), the consumption probability contour line data of the existing store and the new store is generated (step S66), and various numerical results regarding the existing store and the new store are written in a file. (Step S
67). Further, a drawing layer analysis plane file is created (step S68).

FIG. 20 is a flow chart for explaining an example of the processing procedure of analysis plane drawing and analysis operation to which the present invention is applied. Here, the display / analysis unit 13 has a function for performing geographical information analysis, and if data is created in the DB editing unit 11 or the simulation unit 12, it can be treated as independent software. The display / analysis unit 13 is composed of a plurality of modules described later.

First, the display / analysis unit 13 is activated (step S71), the drawing analysis plane file is read (step S72), the consumption probability contour line data for each store is read (step S73), and sales for each consumption category are calculated. Consumption data by category for reading (step S7
4). Next, select the raster map file as the background map (step S75), layer the analysis plane file,
Consumption probability contour lines by store are displayed (step S76).
Next, it is judged whether or not the shape of the consumption probability contour line of the target store (existing store or new store) is appropriate from an empirical rule (step S7).
7) If it is determined to be valid (in the case of YES), display / analysis operation according to various modules is performed (step S7).
8). Here, an example of the functions and processing procedures of the various modules will be described below. Here, the module configuration in the commercial area analysis system 10 is not limited to these, and the function as the commercial area analysis system 10 may be realized by another module configuration. Also,
When it is determined in step S77 that it is not appropriate (NO
In the case of), the process returns to step S21 or step S61 described above, the setting is changed, and the process is performed again.

(Calculation Execution Module) Module 1 has a function of performing calculation under various selection conditions, calculation conditions and display conditions, and writing out an analysis plane for displaying the results together with a background map. It is constantly activated and outputs results in a very short time. It is a module consisting of various calculation subroutines and result display subroutines.

(Analysis Module Concerning Regional Statistics DB 14) Module 2: A function that selects the items to be written in the mesh and the statistical value calculation target, and directly writes the numerical value into each mesh. Select by selecting the check box in the column window (multiple selections are possible). The maximum number of selections is determined by the analyst by looking at the results depending on the ease of viewing. -Module 3: having a function of selecting a background map (overall view) of an analysis area divided by scale and range. -Module 4: It has a function of selecting whether to display the mesh statistic value in the entire area of the selected whole view or to enlarge and display it in the partial view. The data writing display in the mesh can be selected from the selected whole view simultaneous display and the partial view display which is partially enlarged and displayed as a separate view. The partial view can be selected by dragging and specifying the range, or by clicking and displaying a point to be seen in detail. In the latter case, the number of meshes to be displayed can be designated as the default. The display size can be specified as desired for both the whole area display and the partial view display.

Module 5: Has a function of designating the screen size of the whole view and partial view. Module 6: It has a function of designating the font (for example, shape, size, thickness, color, etc.) of the writing data in the mesh. Module 7: With respect to the data written in the mesh (a plurality of items can be specified), when the mesh is clicked, the mesh is added, and when the mesh is clicked again, the addition / subtraction function is performed. -Module 8: It has a function of specifying the number of meshes in each of the vertical and horizontal directions of the partial view that is clicked and displayed for each point. -Module 9: has a function of giving a definition formula for creating an index of mesh data and a definition formula of the four arithmetic operation processing function values and instructing display. Here, as an example of an index created from the results of the national census, the ratio of managerial profession workers, the ratio of professional and technical profession workers, the blue-collar ratio, the primary industry employment rate, the public employment rate, 1 life stage household ratio, teenager population ratio, elderly population ratio, labor force ratio, single-person household ratio, home ownership ratio, condominium ratio, large housing ratio, ratio of commuting to school in local municipalities, ratio of inflowing population for 5 years, etc. is there.

Module 10: Has a function of instructing to display the selected mesh statistical value, the sum of the defined indexes and function values, the average, the standard deviation, etc. of the whole view or partial view in a separate window. For each item on the regional statistics DB 14 or for the created index, the total mesh total value, average value, standard deviation, etc. of the numerical values of each range in the overall view or partial view are calculated and displayed in a new window. In the case of an index, the total value is not given, and the average value is the value obtained by summing the denominator and the numerator separately. -Module 11: has a function of instructing the calculation and display of the deviation value of the mesh statistical value and the 1/2 deviation value. Calculate the normal deviation value using the standard deviation or the deviation value using the 1/2 standard deviation (the standard deviation is often larger than the average value, etc.) by the following formula (8), Display within each mesh and save the results. Mesh deviation value = [10 × (mesh statistical value−average value of whole or partial view / standard deviation (or 0.5 × standard deviation)] + 50 Equation (8)

Module 12: Has a function of indexing the mesh statistical value with the average value of the range of the whole view or partial view as 100 and instructing the display. Regional statistics DB1
For each item above 4 or for the created index, the mesh total value and the average value in the range of the whole view or partial view are calculated, and the value of each mesh is indexed and displayed in each mesh. In the case of an index, the total value is not given, and the average value is the value obtained by summing the denominator and the numerator separately. -Module 13: has a function of calculating the specialization coefficient of various share indexes created from mesh statistical values and instructing display. For each item in the regional statistics DB 14 or for the created share index, an area is arbitrarily dragged and designated by an input means such as a mouse on the screen, and a specialization coefficient that sets the value of the entire area to 1 Is calculated by the formula (9) shown below and displayed in each mesh. Mesh specialization coefficient for item a of item a = (mesh value of item a / mesh value of item b) / (total view or partial view total value of item a / total view or partial view total value of item b)・ Formula (9)

Module 14: has a function of designating the radius of a point buffer at an arbitrary point to be clicked and displayed, and instructing to calculate and display a statistical value and a calculated value within a circle. Module 15: Has a function of automatically plotting contour lines connecting the same level with respect to the selected data or index. -Module 16: has a function of instructing Voronoi division (a method of dividing the market at the middle line between facilities). If this is specified, the statistical value is displayed within the Voronoi dividing line. Module 17: Has a function of instructing to display the closed population estimation value by mesh age group for each year. In addition,
These estimated values are calculated in the trade area analysis system 10 using the birth rate, mortality rate, spouse rate, and the like.

Module 18: Has a function of instructing to display the characters written in the mesh in different colors according to the selected mesh statistical value or the hierarchy of the defined index or function value. In this example, the hierarchy can be designated from 7 to 20 levels. Module 19: has a function of searching for meshes in the upper 10% and lower 10% (these% can be specified) of the mesh statistical value or index value and instructing the characters written in the mesh to be displayed in different colors . Module 20: has a function of instructing to calculate a scatterplot of two terms between mesh statistical values and display it in another window. Module 21: has a function of displaying a color-coded table for each layer based on the scatter diagram of the two terms between mesh statistic values in a different frame, and instructing the characters written in the mesh to be color-coded based on the table. -Module 22: has a function of additionally filling, saving, and exporting mesh attribute data. For example, since the most recent data can always be obtained regarding the population by town-character, the number of single-person households, and the number of households with two or more persons, this item can be written in the mesh statistics DB file from the screen.
Click on the corresponding mesh, launch the input card,
From the screen, you can enter any item name, such as 2003 population, 2003 single households, 2003 2 or more households, and output them in the same mesh statistics DB. To be done. For example, if it is specified in 2003 that the number of households is 2 or more, it can be used in the Huff reality sales convergence simulation.

(Link data such as household consumption, regional statistics D
Module for B14) Module 23: Which attribute household (eg, single household, two or more ordinary households, household head age group, etc.) is used to calculate the mesh consumption estimation value by item category? Has the function of instructing. The number of households by age group is
It is not included in the original census items, and is calculated and obtained within the commercial area analysis system 10. Module 24: Has a function of designating the mesh consumption amount, for example, by category, by item, by age group, or by region. Incidentally, the category can be arbitrarily defined in advance on the user side and input. As the linked data, for example, in addition to the annual household survey report, a national consumption fact-finding report can be used. Module 25: Has a function of specifying a category or an item for which the mesh consumption amount is calculated. It should be noted that various calculation and display functions are common to the module relating to the mesh statistical value described above.

(Analysis Module for Store DB 15) Module 26: Has a function of importing data from the store DB 15. On the display / analysis unit 13, store DB
15 and click the corresponding store DB file name, import of the corresponding data starts on the display / analysis unit 13, and as a result, a store data list (for example, point name and its attribute list) is displayed on the screen. A window opens that lets you choose which files to view and which points in which files should be flagged on the map. You can select multiple file names to import at one time. Module 27: It has a function of selecting a point file in which data for displaying a flag and a name and an individual point name for which a flag is set are selected and referring to an attribute list. This is because when a file name is clicked and selected in the window, a list of point names constituting the file and a table of their attributes is displayed in the point file window. This allows you to see all the attributes of the point. Select the points for which you want to set a flag by checking the check box in the point name list in the window. Multiple selection or full selection is possible. The maximum number of selections is determined by the analyst by looking at the results each time depending on the visibility and purpose.

Module 28: The shape and color of the flag to be displayed on the screen is selected for each DB file.
It has a function to select whether to add a point name or only a flag mark. Module 29: has a function of instructing to display one selected point attribute next to the point name of the flag. Select by checking the check box of the point name and its attribute list in the attribute list window. When executed in Module 1 described above, the selected attribute is displayed next to the flag name. -Module 30: having a function of designating a calculation for totalizing meshes of point data and a mesh share calculation. For each selected attribute, each mesh total value and each mesh share for the total value in the whole view or partial view are written in the mesh.

Module 31: Has a function of designating display of the total, average, and standard deviation of point attribute values of the whole view or partial view in a separate frame. -Module 32: having a function of instructing point search conditions. Regarding the attribute of the point, when the search condition is designated by the size of the numerical value and executed by the module 1 described above, the flag of the point which meets the condition is changed to the designated color. -Module 33: Has a function of displaying and correcting the latitude and longitude of the point flag. By clicking on the desired point, the latitude and longitude of the flag position of the point is displayed. When the flag position correction check is selected and the flag is moved on the screen to correct the position, the correct latitude and longitude are displayed, and the latitude and longitude data in the DB list is also changed to the correct latitude and longitude. Module 34: has a function of instructing to create new point data by setting a flag on the screen. If you select the new flag setting check and set a new flag on the screen using a mouse or the like, the data can be directly written in the corresponding DB list and the latitude and longitude are automatically written.

(Analysis Module for Function between Point Attribute Value and Mesh Statistic Value) Module 35: Arbitrarily defined four arithmetic operation statistics between selected mesh statistic value and selected point attribute value in each mesh- It has a function of instructing to display a point function value (for example, the purchase amount of the customer in the mesh in the mesh item consumption amount). Module 36: Binary scatter plot between selected mesh statistics-point attribute values in whole or partial view (eg number of households-number of customers or number of households-customer acquisition rate or number of leaflets distributed-visit store It has a function to display the number of customers) in a separate frame. Module 37: 2 meshes between selected mesh statistics and point attribute values for meshes in whole or partial view
It has a function to display the color-coding table for each layer in a separate frame based on the term scatter diagram and to instruct to display the characters written in the mesh by color based on it.

(Analysis Module for Polygon Generation Zone) Module 38: Has a function of polygon generation and zone content file creation. When polygon generation is selected, an arbitrary polygon polygon (zone) can be generated with a mouse or the like on a raster map such as a block map. Polygon features and identification include, for example, mesh statistics, block maps, village identification using field surveys, etc., as well as industry location identification using point flags and numerical maps of town page DBs by industry, city planning maps, etc. You can also refer to it. The zone name, the identification index, and the description (for example, up to 256 characters) can be directly written from the screen to the corresponding line of the data file in the generated zone polygon file window. They are C along with zone position information and zone code.
Can be output to SV file etc. In addition, you can paste the local photos by zone code on the screen in another file and open it again.

The zoning as described above can be executed only when a raster map having a high degree of perfection as a map is used as a background. Zone polygons are exported and saved, but the saved polygon data is
It can be read and reproduced only in the same analysis plane. Examples of this zoning include, for example, agricultural areas and settlements, industrial areas, distribution areas, town centers and shopping areas, rental apartment areas, old low-income residential areas, low-quality old residential areas, and elderly people. Residential area of household, residential area of retired people who live well, new residential area of high-income household, old residential area of high-income household, inexpensive new private residential area, private residential area where skilled workers live, manager or Professional residential area, middle-class private residential area, old condominium, new condominium, condominium with a large number of singles, managerial or professional housing, comfortable condominium, middle-class condominium, condominium Some are not very good, some are the poorest.

Module 39: Has a function of instructing polygon area calculation. When the area calculation is checked and the above-mentioned module 1 is executed, the area is written and displayed in the polygon, and the area data is written in the polygon file. Module 40: Has a function of instructing calculation of polygon statistical values. For the created polygon, it is instructed to calculate the polygon value of various mesh statistical values designated by the area proportional calculation method based on the mesh data. When the module 1 described above is executed, the specified statistical value is written and displayed in the polygon, and the statistical value data is written in the polygon file.

(Analysis Module Concerning Contour Lines by Store Created by Simulation Unit 12) Module 41: Store file storing data for displaying flag and name on screen and store name where flag is set are selected, and attribute list Has the function of referring to. When the file name is clicked and selected in the window, the shop name and its attribute list that form the file are displayed in the shop file window. This allows you to see all the attributes of the store. Select the store where you want to set the flag by checking the check box in the store name list in the window. Multiple selection or full selection is possible. The maximum number of selections is determined by the analyst by looking at the results each time depending on the visibility and purpose.

Further, in the module 41, the shape and color of the flag to be displayed on the screen can be selected for each DB file, and the flag can be selected with a store name or only with the flag mark. Module 42: It has a function of opening a window on the screen in which a store name for displaying probability contour lines is stored and selecting a store name for displaying contour lines. Module 43: having a function of selecting to color-code the probability contour lines according to probability. -Module 44: Has a function of instructing to display three contour lines of 0.05 to 1.00. -Module 45: It has a function of selecting a store name and selecting to display the theoretical sales amount according to store-based consumption category in the form of a table.

The analysis results of the various modules executed in step S78 described above are displayed, for example, in printout or C
Output to an SV file or the like (step S79). Also,
It is also possible to paste an image onto Microsoft Excel (registered trademark) or the like. At this time, all the analysis, data addition, and processing results can be saved in a file, and the entire view, partial view, and any image can be pasted and saved in a table provided by a predetermined application.

Although the functions of the respective embodiments of the commercial area analysis system of the present invention have been mainly described above, the present invention can also be embodied as a commercial area analysis method as described as a method. . Further, like the commercial area analysis system having each of these functions, a form for causing a computer to function as a commercial area analysis system, or a program for causing a computer to execute a commercial area analysis method, or the program is recorded. A form as a computer-readable recording medium is also possible.

An embodiment of a recording medium storing a program and data for realizing the function of the commercial area analysis according to the present invention will be described. As the recording medium, specifically, a CD-R
OM, magneto-optical disk, DVD-ROM, floppy (registered trademark) disk, flash memory, memory card, memory stick, and other various ROMs and RAMs
Etc. can be assumed, and the functions of the system of each of the embodiments of the present invention described above are executed by a computer, and a program for realizing the function of the commercial area analysis is recorded and distributed to realize the function. To facilitate. Then, the recording medium as described above is attached to an information processing device such as a computer and the program is read by the information processing device, or the program is stored in a storage medium included in the information processing device, and as necessary. By reading, the commercial area analysis function according to the present invention can be executed.

[0183]

According to the present invention, when analyzing the commercial area of each store in a business format having a competitive relationship by using GIS, especially when there is intense competition between the shops, the competitive relationship and the commercial area spatial structure of the existing store. In addition to being able to analyze, it is possible to analyze consumer attributes and possible sales by consumption category.

Further, from the above analysis results, the user can formulate appropriate customer policy, competition policy, product policy, sales policy, etc., and, in the case of a new store, the accuracy when the plan scale is variously changed. It is possible to forecast high sales, and as a result (when the cost is taken into consideration), an appropriate investment scale can be determined.

Since the consumer attribute can be grasped from the estimated or predicted commercial area space structure, and the possible consumption suction amount (sales) by category of each consumption layer can be estimated from it.
The user can formulate floor plans, product policies, sales policies, and the like.

Further, a commercial area analysis system, method, program, which can improve the accuracy of the sales forecast of a new store and the forecast of the commercial space structure, and realize a high-level marketing analysis regardless of the new store or the existing store, And a recording medium can be provided.

[Brief description of drawings]

FIG. 1 is a diagram illustrating a configuration example of a commercial area analysis system according to an embodiment of the present invention.

FIG. 2 is a diagram showing an example of a theoretical total sales list according to the present invention.

FIG. 3 is a diagram showing an example of a commercial area analysis-regional map list according to the present invention.

FIG. 4 is a diagram showing an example of a commercial area analysis-all stores and a characteristic table according to the present invention.

FIG. 5 is a diagram showing an example of a commercial area analysis-contour line list according to the present invention.

FIG. 6 is a diagram showing an example of a commercial area analysis-item list according to the present invention.

FIG. 7 is a diagram showing another example of a commercial area analysis-item list according to the present invention.

FIG. 8 is a diagram showing an example of a screen for setting the area between trade area analysis and contour lines according to the present invention.

9 is a diagram showing an example of map data selected by the regional map list shown in FIG.

FIG. 10 is a diagram showing an example of a case where probability contour lines of respective stores selected in the store selection fields shown in FIGS. 4 and 5 are displayed on map data.

11 is a diagram showing an example of a case where numerical data corresponding to an item selected in the item list shown in FIG. 6 is displayed on map data.

12 is a diagram showing an example of a case where numerical data corresponding to an item selected in the item list shown in FIG. 7 is displayed on map data.

13 is a diagram showing an example in which the map data shown in FIG. 12 is enlarged and displayed as a partial view.

14 is a diagram showing an example of a calculation result of a trade area analysis-category consumption amount calculated based on the area setting between contour lines shown in FIG.

FIG. 15 is a flow chart for explaining an example of a processing procedure in analysis area designation, map data and area statistical data reading, processing and editing, and analysis plane data creation to which the present invention is applied.

FIG. 16 is a flowchart for explaining an example of a processing procedure in various settings for simulation to which the present invention is applied.

17 is a flowchart for explaining another processing procedure in various settings for simulation to which the present invention is applied, and is a continuation of the flowchart shown in FIG.

FIG. 18 is a flowchart for explaining a processing procedure of an existing store simulation to which the present invention is applied.

FIG. 19 is a flowchart for explaining a processing procedure of a new store sales prediction simulation to which the present invention is applied.

FIG. 20 is a flowchart for explaining an example of a processing procedure of analysis plane drawing and analysis operation to which the present invention is applied.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... PC, 10 ... Commercial area analysis system, 11 ... DB editing part, 11a ... Analysis area definition module, 11b ... Analysis plane file creation module, 12 ... Simulation part, 12a ... Input I / F module, 12b ... Simulation calculation execution module , 12c ... New store sales forecast module, 13 ... Display / analysis unit, 13a ... Start-up module, 13b ... Display / analysis module, 13c ... Output module, 14 ... Regional statistics DB, 14a ... Regional statistics information,
14b ... Link data, 15 ... Store DB, 15a ... Store information, 16 ... User polygon DB, 16a ... Attribute information,
17 ... Map DB, 17a ... Map data, 20 ... Theoretical total sales list, 21 ... Store name, 22 ... Business category, 23 ... Theoretical monthly sales,
24 ... All department monthly sales, 25 ... Monthly sales relative error, 26 ... Direct sales surface, 27 ... Customer attraction coefficient, 28 ... Customer attraction, 29 ... Limit distance, 30 ... Store number, 40 ... Regional map list, 41
… Numeric display screen selection, 42… Overall view screen display method selection,
43 ... Number of grids for displaying partial map, 44 ... List of whole area map, 5
0 ... All shops and characteristics table, 51 ... Group selection column, 52
... Store display column, 53 ... Characteristic table, 60 ... Contour line list, 6
1 ... Probability contour line selection column, 62 ... Store selection column, 63 ... Screen designation, 70 ... Item list, 71 ... Linear combination designation, 72 ...
Index preparation, 73 ... Item list, 80 ... Contour line area setting screen, 81 ... Contour line area display selection column, 82 ... Contour line value designation column, 83 ... Store name list, 90 ... Map data, 91 ...
Longitude direction grid number, 92 ... Latitude direction grid number, 93
... Probability contour lines, 94, 95 ... Numerical data, 96 ... Detailed data list regarding numerical data, 97 ... Partial view, 100 ...
Calculation result of consumption amount by category.

Continued front page    (72) Inventor Hirohide Nagatsuka             2-3-1105 Kugenumahigashi, Fujisawa City, Kanagawa Prefecture F term (reference) 2C032 HB03 HB05 HC24 HC26 HC27                 5B050 AA03 BA08 BA17 CA07 EA19                       FA02 FA12 FA14

Claims (55)

[Claims] 1. A trade area analysis system for setting a trade area on map data and analyzing the set trade area, wherein statistical information about consumers is obtained for each area obtained by dividing map data prepared in advance under specified conditions. At least one of a stored regional statistical database, target area designating means for designating a target area for commercial area analysis on the map data, and a plurality of store names, sales floor areas, latitude and longitude indicating a location, and actual sales values for each business category By performing a simulation calculation for estimating the commercial area spatial structure and theoretical sales for each of the stores included in the target area designated by the target area designating means and the store database storing the store information, a predetermined probability model Simulation means for determining the parameter so as to approach the actual sales value stored in the store database, A trade area analysis system having: 2. The trade area analysis system according to claim 1, further comprising a store designating unit that designates a simulation target store from the stores included in the target area,
The simulation means determines a parameter of the probabilistic model so as to approach a sales record value stored in the shop database by performing a simulation for each shop specified by the shop specification means. Trade area analysis system. 3. The commercial area analysis system according to claim 1, wherein the simulation means stores the estimated sales value calculated based on the result of simulation using the probabilistic model and the store database. A trade area analysis system, wherein the parameter of the probabilistic model is determined by repeatedly performing simulation until the difference from the actual sales value falls within a preset allowable range for the actual sales value. 4. The trade area analysis system according to claim 1, wherein the probabilistic model is the exponentiation β of the sales floor area of a store specified by business category and the distance from the customer residence point to each store. Power α, a limit distance representing the distance from the simulation target store included in the target area to the customer's residential area located furthest away, and the size of the attraction force (attractiveness) that attracts the customer at the store. The customer attraction expressed, the maximum number of stores that can be selected by the customer within the limit distance that can be selected by business type, and the total number of households per household in all consumption categories in the household survey that the stores of each business type deal with For the target category consumption,
It is defined as a parameter with a consumer business category share that represents the ratio that a collective consumer can actually consume in a store of each business category as an average of one household, and the customer attraction force is the customer attraction force with respect to the sales floor area of the store. A trade area analysis system, which is calculated by multiplying by a customer attraction coefficient for adjusting the size of a shop and a bottom-up coefficient for raising the sales floor area in advance according to business types. 5. The trade area analysis system according to claim 1, wherein the simulation means exponentiates the sales floor area β, exponentiates the distance α, the marginal distance, the consumer business category share, and the bottom up. The coefficient and the maximum number of stores are designated in advance according to the business category of the store, and the power β of the designated sales floor area, the power of the distance α, the limit distance, the consumer business category share, the bottom-up coefficient, and the maximum number of the stores are the stochastic model. Is set to, and the customer attraction coefficient is determined by repeating the simulation while correcting the customer attraction coefficient according to the immediately preceding simulation result for each store included in the target area. Trade area analysis system. 6. The commercial area analysis system according to claim 1, wherein the simulation means specifies a simulation count for a store included in the target area, and performs the simulation of the specified count. After the execution, if the difference between the estimated sales value calculated based on the simulation result and the actual sales value stored in the store database is equal to or greater than a predetermined threshold value,
The store showing the difference is extracted, the customer attraction coefficient corresponding to the extracted store is determined at the time when the simulation is completed, and further repeated simulation is performed while changing the limit distance within a specified range. Accordingly, the commercial area analysis system characterized in that the limit distance is determined. 7. The commercial area analysis system according to claim 1, wherein the simulation means specifies a simulation count for a store included in the target area, and performs the simulation of the specified count. After the execution, if the difference between the estimated sales value calculated based on the simulation result and the actual sales value stored in the store database is equal to or greater than a predetermined threshold value,
A store showing the difference is extracted, the customer attraction coefficient corresponding to the extracted store is determined at the time when the simulation is finished, and further repeated simulation is performed while changing the power α of the distance within a specified range. The trade area analysis system is characterized in that the power α of the distance is determined by performing 8. The commercial area analysis system according to claim 1, wherein the simulation means specifies a simulation count for a store included in the target area, and performs the simulation of the specified count. After the execution, if the difference between the estimated sales value calculated based on the simulation result and the actual sales value stored in the store database is equal to or greater than a predetermined threshold value,
The store showing the difference is extracted, the customer attraction coefficient corresponding to the extracted store is determined at the time when the simulation is finished, and the consumer business category share is changed within a specified range, and further repeated simulation is performed. A trade area analysis system, characterized in that the consumer business category share is determined by carrying out. 9. The trade area analysis system according to claim 1, wherein the simulation unit includes a power β of the sales floor area and a power α of a distance for stores included in the target area. The customer attraction, the consumer business category share, and the maximum number of stores are designated in advance, and the exponential β of the specified sales floor area, the power of the distance α, the customer attraction, the consumer category share, and the maximum number of stores are set to the probability model. A commercial area analysis system, wherein the limit distance is determined by repeatedly performing simulation while setting the limit distance and changing the limit distance within a specified range according to the immediately preceding simulation result for each store. 10. The trade area analysis system according to claim 1, wherein the simulation unit is a power of the sales floor area β, a limit distance, and the customer for a store included in the target area. Suction power, consumer business category share, the maximum number of stores is specified in advance, and the specified sales floor area exponent β, marginal distance, customer suction force, consumer business category share, and maximum number of stores are set in the stochastic model, By repeatedly performing simulation while changing the exponentiation α of the distance within a specified range according to the immediately preceding simulation result for each of the stores,
A trade area analysis system characterized by determining a power α of the distance. 11. The commercial area analysis system according to claim 1, wherein the simulation means includes a power β of the sales floor area and a power α of the distance for stores included in the target area. Customer attraction,
Limit distance, the maximum number of stores is specified in advance, and the specified power of sales floor area β, power of distance α, customer attraction, limit distance, and maximum number of stores are set in the stochastic model,
A business area analysis system, wherein the business category share is determined by repeatedly performing simulations while changing the consumer business category share within a specified range according to the immediately preceding simulation result for each of the stores. 12. The trade area analysis system according to claim 1, wherein the simulation means calculates based on a result of simulation using the probabilistic model for all stores included in the target area. The stochastic model is obtained by repeatedly performing simulation until the average of the difference between the estimated sales value and the actual sales value stored in the store database falls within an allowable range set in advance for the actual sales value. A trade area analysis system characterized by determining the parameters of. 13. The commercial area analysis system according to claim 1, wherein the simulation unit is based on store information stored in the store database for each business type, based on store information, sales floor efficiency, sales floor efficiency, and sales form. , A store is extracted in advance from the viewpoint of including at least one or more of the location area concentration degrees, the extracted stores are classified as different business categories into different business category files, and the parameters of the probability model are classified for each of the classified business category files. A trade area analysis system characterized by making decisions. 14. The trade area analysis system according to claim 1, wherein the simulation means calculates a sales estimate value calculated based on a first simulation result for a store included in the target area. If the difference from the actual sales value stored in the store database is equal to or more than a predetermined threshold value, the store showing the difference is extracted, the extracted store is classified as a different business type into different business type files, and the classification is performed. A trade area analysis system, wherein the parameters of the probabilistic model are determined for each business type file. 15. The business area analysis system according to claim 13 or 14, wherein the simulation means stores the business category files classified as the different business categories before executing the simulation for the stores included in the target area. Among them, it is characterized in that a store format file to which a store having a small sales floor area and a high sales record belongs is selected, and the sales floor area of the store belonging to the selected store file is multiplied by the above-mentioned bottom-up coefficient in advance. Trade area analysis system. 16. The commercial area analysis system according to claim 1, wherein the simulation unit performs the target estimation when the sales forecast of a new store in the target region designated by the target region designation unit is performed. By performing a simulation on each of the existing stores included in the area, the parameters of the probabilistic model are determined so as to approach the sales actual values stored in the store database, and then the new store for which the sales forecast is performed The store name, sales floor area, and latitude and longitude data indicating the location are set, and the determined parameters are set in the probabilistic model, and a simulation is performed using the probabilistic model in which the parameters are set to predict sales. A commercial area analysis system characterized by performing. 17. The trade area analysis system according to claim 1, wherein the simulation unit is based on a simulation result executed for an existing store or a new store included in the target area,
A list of analysis results including at least one of a sales estimate or a forecast value, a sales actual value, a difference between the estimated sales value and the actual sales value, a customer attraction coefficient, a customer attraction force, and a limit distance at each of the existing store and the new store. A trade area analysis system characterized by generating. 18. The trade area analysis system according to claim 17, wherein the simulation means stores a simulation result executed for an existing store or a new store included in the target region and the regional statistical database. A commercial area analysis system, which generates a sales estimate or a predicted value for each consumption category in a household budget survey for each of the existing store and the new store based on the statistical information. 19. The commercial area analysis system according to claim 1, wherein the simulation means simultaneously executes a simulation for each store included in the target area. 20. The commercial area analysis system according to claim 1, wherein the simulation means uses a stochastic model that formally matches the Hough model. 21. The commercial area analysis system according to claim 1, wherein statistical information and store information corresponding to the target area designated by the target area designating unit are provided to the regional statistical database and the store database, respectively. From the statistical information and the shop information that has been read, and the simulation means repeatedly executes a simulation to select one of the read statistical information, the shop information, and the analysis result of the executed simulation. Alternatively, a commercial area analysis system having an analysis plane creating means for creating an analysis plane as a data group processed and edited so that a plurality of layers can be directly layer-displayed on map data. 22. The commercial area analysis system according to claim 1, wherein the regional statistical database stores statistical information regarding consumers for each mesh and / or for each town-kanji, and households nationwide. Storing data on household consumption in which the actual conditions of consumption and purchase are investigated, and when the analysis plane creating means reads statistical information from the regional statistical database, it reads household consumption data according to the statistical information to be read, A commercial area analysis system, characterized in that the read household consumption data is combined with the statistical information and edited, and an analysis plane is created based on the edited statistical information. 23. The trade area analysis system according to claim 1, further comprising: analysis of statistical information stored in the regional statistical database on the analysis plane created by the analysis plane creating means; An analysis unit configured to perform analysis on store information stored in the store database, and any one or more of the statistical information, the store information, and the result analyzed by the analysis unit,
A commercial area analysis system, wherein layers are displayed on map data having a target area designated by the target area designation means. 24. The commercial area analysis system according to claim 1, wherein the simulation unit is an existing store included in the target area on the analysis plane created by the analysis plane creation unit. A consumption probability contour line for each store is generated based on a simulation result executed for a new store, and the analysis unit displays the consumption probability contour line generated by the simulation unit as a layer on the map data. A commercial area analysis system, which analyzes the commercial area structure of the existing store or the new store based on the displayed consumption probability contour lines. 25. The commercial area analysis system according to claim 24, wherein the analysis means can display the consumption probability contour lines by color coding according to the consumption probability. 26. The commercial area analysis system according to claim 1, wherein an arbitrary polygon area defined by a user is generated on map data having a target area designated by said target area designating means. And a polygon database storing statistical information according to the generated polygon area and attribute information including store information, wherein the analysis means is an analysis plane created by the analysis plane creation means. A commercial area analysis system, characterized in that the attribute information stored in the polygon database is analyzed above, and the analysis result is displayed as a layer on the map data. 27. The trade area analysis system according to claim 1, wherein the map data is in a raster format or a bitmap format. 28. The commercial area analysis system according to claim 1, wherein the analysis result of the statistical information and / or the store information according to the area divided on the map data is applied to the entire area of the map data. When a desired portion of the map data that is displayed across the analysis result is displayed in an enlarged manner, it is possible to enlarge and display the analysis result in an arbitrary size as a partial view in a frame separate from the map data. A characteristic trading area analysis system. 29. A trade area analysis system for setting a trade area on map data and analyzing the set trade area,
A map database that stores map data, a regional statistical database that stores statistical information about consumers for each area obtained by dividing the map data under specified conditions, and a desired map from the map data stored in the map database. At least at least one of a target area designating unit for selecting data and designating a target area for commercial area analysis on the selected map data, and a store name, a sales floor area, a latitude and longitude indicating a location, and a sales result value for each business category By performing a simulation calculation for estimating the commercial area spatial structure and theoretical sales for each of the stores included in the target area designated by the target area designating means and the store database storing the store information, a predetermined probability model A simulation that determines the parameter so that it approaches the actual sales value stored in the store database. Trade area analysis system; and a Deployment means. 30. A trade area analysis method for setting a trade area on map data and analyzing the set trade area, wherein statistical information about consumers is obtained for each area obtained by dividing map data prepared in advance under specified conditions. Prepared beforehand a stored regional statistical database and a store database storing store information including at least one of a store name, sales floor area, latitude and longitude indicating a location, and actual sales value for each business type, and a trade area on the map data. A target area designating step for designating a target area for analysis, and a simulation calculation for estimating a commercial area spatial structure and theoretical sales for each store included in the designated target area, whereby parameters of a predetermined probability model are obtained. And a simulation step for determining so as to approach the actual sales value stored in the store database. A trade area analysis method characterized by the following. 31. The trade area analysis method according to claim 30, wherein a store to be simulated is designated from the stores included in the target area, and the simulation is performed for each of the designated stores. A method for analyzing a commercial area, characterized in that the parameters of the probabilistic model are determined so as to approach the actual sales value stored in the store database. 32. The sales area analysis method according to claim 30 or 31, wherein an estimated sales value calculated based on a result of simulation using the probabilistic model and an actual sales value stored in the store database. The difference is
A trade area analysis method, characterized in that the parameters of the probabilistic model are determined by repeatedly performing simulation on the actual sales value until it falls within a preset allowable range. 33. The trade area analysis method according to claim 30, wherein the probabilistic model is the exponentiation β of a sales floor area of a store designated by business category, and a distance from a customer residence point to each store. Power α, a limit distance representing the distance from the simulation target store included in the target area to the customer's residential area located furthest away, and the size of the attraction force (attractiveness) that attracts the customer at the store. The customer attraction expressed, the maximum number of stores that can be selected by the customer within the limit distance that can be selected by business type, and the total number of households per household in all consumption categories in the household survey that the stores of each business type deal with For the target category consumption,
It is defined as a parameter with a consumer business category share that represents the ratio that a collective consumer can actually consume in a store of each business category as an average of one household, and the customer attraction force is the customer attraction force with respect to the sales floor area of the store. Is calculated by multiplying by a customer attraction coefficient for adjusting the size of the product and a bottom-up coefficient for raising the sales floor area in advance according to business type. 34. The trade area analysis method according to claim 30, wherein the exponent β of the sales floor area, the exponentiation α of the distance, the marginal distance, the consumer business category share, the bottom-up coefficient, and the maximum number of stores. And are designated in advance according to the business format of the store, and the power β of the specified sales floor area, the power of the distance α, the limit distance, the share of the consumer business category, the bottom-up coefficient,
While setting the maximum number of stores in the probabilistic model, while repeating the customer attraction coefficient according to the immediately preceding simulation result for each of the stores included in the target area, by repeatedly performing simulation, the customer attraction coefficient is calculated. A trade area analysis method characterized by making a decision. 35. The commercial area analysis method according to claim 30, wherein a simulation count is designated for a store included in the target area, and after the simulation is performed for the designated count, If the difference between the estimated sales value calculated based on the simulation result and the actual sales value stored in the store database is equal to or greater than a predetermined threshold value, the store showing the difference is extracted, and the extracted store is set as the extracted store. A trade area characterized by determining the customer attraction coefficient according to the simulation at the time when the simulation is completed, and further repeating the simulation while changing the limit distance within a specified range. Analysis method. 36. The commercial area analysis method according to claim 30, wherein a simulation count is specified for a store included in the target area, and after the simulation is performed for the specified count, the simulation is performed. If the difference between the estimated sales value calculated based on the simulation result and the actual sales value stored in the store database is equal to or greater than a predetermined threshold value, the store showing the difference is extracted, and the extracted store is set as the extracted store. The customer attraction coefficient according to the determination is determined at the time when the simulation is completed, and the power α of the distance is determined by further repeating the simulation while changing the power α of the distance within a specified range. A method of analyzing commercial areas. 37. The commercial area analysis method according to claim 30, wherein the number of simulations is specified for the stores included in the target area, and after the simulations of the specified number of times are executed, If the difference between the estimated sales value calculated based on the simulation result and the actual sales value stored in the store database is equal to or greater than a predetermined threshold value, the store showing the difference is extracted, and the extracted store is set as the extracted store. It is possible to determine the consumer business condition share by determining the customer attraction coefficient according to the simulation at the end of the simulation, and further repeating the simulation while changing the consumer business condition share within the specified range. A method of analyzing commercial areas. 38. The commercial area analysis method according to claim 30, wherein the sales area is a power β and a distance is a power α for a store included in the target area.
And the customer attraction force, consumer business category share, and maximum number of stores are specified in advance, and the power of the designated sales floor area β, power of distance α, customer attraction force, consumer business category share, maximum number of stores are the probability. A trade area analysis characterized by setting the model and determining the limit distance by repeatedly performing simulation while changing the limit distance within a specified range according to the immediately preceding simulation result for each of the stores. Method. 39. The commercial area analysis method according to claim 30, wherein, for a store included in the target area, the exponent β of the sales floor area, a limit distance, the customer attraction, and the consumer. Specify the business category share and maximum number of stores in advance, set the specified sales floor area exponentiation β, marginal distance, customer attraction, consumer business category share, maximum number of stores in the probability model, and for each of the stores A method for analyzing a trade area, characterized in that the exponent α of the distance is determined by repeatedly performing simulation while changing the exponent α of the distance within a specified range in accordance with the immediately preceding simulation result. 40. The commercial area analysis method according to any one of claims 30 to 39, wherein for the stores included in the target area, the exponent β of the sales floor area and the exponentiation α of the distance.
And the customer attraction force, the limit distance, and the maximum number of stores are designated in advance, and the power β of the sales floor area, the power α of the distance, the customer attraction force, the limit distance, and the maximum number of stores are set in the probability model. At the same time, the market share analysis is characterized in that the consumer business category share is determined by repeatedly performing simulations while changing the consumer business category share within a specified range according to the immediately preceding simulation result for each of the stores. Method. 41. The sales area analysis method according to any one of claims 30 to 40, wherein the estimated sales value is calculated based on the result of simulation using the stochastic model for all stores included in the target area. By repeating the simulation until the average difference between the actual sales value stored in the store database and the actual sales value falls within an allowable range set in advance,
A method for analyzing a trade area, which comprises determining a parameter of the probabilistic model. 42. The commercial area analysis method according to any one of claims 30 to 41, wherein the sales floor area, sales floor efficiency, sales form, and location area concentration level are based on the shop information stored in the shop database according to business category. From the viewpoint including at least one or more of the above, stores are extracted in advance, and the extracted stores are classified as different business types into different business type files,
A trade area analysis method, characterized in that the parameters of the probabilistic model are determined for each of the classified business category files. 43. The commercial area analysis method according to claim 30, wherein the estimated sales value calculated based on a first simulation result for a store included in the target area and the store database are stored in the store database. When the difference from the stored sales actual value is equal to or more than a predetermined threshold value, the store showing the difference is extracted, the extracted store is classified as a different business category into different business category files, and each of the classified business category files A method for analyzing a trade area, which comprises determining a parameter of the probabilistic model. 44. The sales area analysis method according to claim 42 or 43, wherein the sales floor area is included in the business category file classified as the separate business category before the simulation is performed on the stores included in the target area. Is selected, and a business category file to which a store with a high sales record belongs is selected, and the sales floor area of the store belonging to the selected business category file is multiplied by the bottom-up coefficient in advance. 45. In the commercial area analysis method according to any one of claims 30 to 44, when forecasting sales of a new store in the target area, a simulation is performed for each existing store included in the target area. Thus, after determining the parameters of the probabilistic model so as to approach the actual sales value stored in the store database, the store name of the new store for which the sales forecast is performed,
The sales floor area and latitude / longitude data indicating the location are set, the determined parameters are set in the probability model, and a simulation is performed using the probability model in which the parameters are set to perform sales prediction. The trade area analysis method. 46. The commercial area analysis method according to claim 30, wherein the existing store or the new store is based on a simulation result executed for the existing store or the new store included in the target area. It is possible to generate an analysis result list that includes at least one or more of sales estimate or forecast value of each store, sales actual value, difference between estimated sales value and actual sales value, customer attraction coefficient, customer attraction force, and limit distance. A method of analyzing commercial areas. 47. The commercial area analysis method according to claim 46, based on a simulation result executed for an existing store or a new store included in the target region, and statistical information stored in the regional statistical database. Then, a sales area analysis method or a forecast value for each consumption category in a household budget survey is generated for each of the existing store and the new store. 48. The commercial area analysis method according to claim 30, wherein a simulation is simultaneously executed for each store included in the target area. 49. The trade area analysis method according to claim 30, wherein a stochastic model that formally matches the Hough model is used. 50. The commercial area analysis method according to claim 30, wherein statistical information and store information corresponding to the target area are read from the regional statistical database and the store database, respectively, and the read statistical information is read. Further, a simulation is repeatedly executed based on the shop information, and any one or more of the read statistical information, shop information, and analysis result of the executed simulation can be directly layer-displayed on the map data. A trade area analysis method characterized by creating an analysis plane as a data group that has been processed and edited. 51. The commercial area analysis method according to claim 30, wherein, on the analysis plane, an analysis relating to statistical information stored in the regional statistical database and a store stored in the store database. A method for analyzing a commercial area, characterized in that an analysis related to information is executed and any one or a plurality of the statistical information, the shop information, and the result of the analysis and execution are layer-displayed on the map data. 52. The trade area analysis method according to claim 30, based on a simulation result executed on an existing store or a new store included in the target area on the analysis plane. Generate a consumption probability contour line for each store, display the generated consumption probability contour line in a layer on the map data, and analyze the commercial area structure of the existing store or the new store based on the layer-displayed consumption probability contour line. A method for analyzing commercial areas characterized by. 53. The trade area analysis method according to any one of claims 30 to 52, wherein statistical information and / or store information analysis results corresponding to areas divided on the map data are distributed over the entire map data. When a desired portion of the map data that is displayed across the analysis result is displayed in an enlarged manner, it is possible to enlarge and display the analysis result in an arbitrary size as a partial view in a frame separate from the map data. A method of analyzing commercial areas. 54. A commercial area analyzing method according to any one of claims 30 to 53 for causing a computer to execute the function as the commercial area analyzing system according to any one of claims 1 to 29. Program to run on. 55. A computer-readable recording medium in which the program according to claim 54 is recorded.