JPH11259570A - Management simulation method and management decision support system using the management simulation method - Google Patents

Abstract

(57) [Summary] [Problem] To faithfully simulate a product / service giving / receiving action and a business action generating system by a relatively simple algorithm based on actual human actions and product sales states. An initial setting of a simulation is first performed, and then, a business time section, such as within a store opening preparation time, a business hours, and an end preparation time, corresponding to a realistic store is given to the simulation time. Subsequently, a time step for the business time section is given. Next, the action element is entered into the store, and the customer agent,
Work with employee agents. Then, the action processing of the customer agent in a unit time or the action processing of the employee agent in a unit time is performed according to the type of the action element. Thereafter, a stationary work that does not require the simulation of the behavior of the employee agent in the store is performed, and the time is updated according to the time increment. Finally, the date is updated.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

TECHNICAL FIELD The present invention relates to a retail business, a restaurant business,
A management simulation method that simulates the occurrence of product / service transfer behavior and business behavior in the service industry, etc., in accordance with actual customer, store clerk, and manager behavior in a commercial area or store, and management using this management simulation method It relates to a decision support system.

[0002]

2. Description of the Related Art In a retail business or the like, the layout of a sales floor in a store, the arrangement of merchandise shelves (shelf allocation), the setting of a special sale price, the location conditions in a commercial area, etc., do not greatly affect sales and profits. As is well known, at present, most of these plans, such as layouts, rely on the intuition and experience of the business manager. In addition, a method for making a plan such as a layout as accurate as possible and a method for forecasting sales have been proposed.

[0003] For example, Japanese Patent Application Laid-Open No. 64-21571 discloses a technique relating to shelf arrangement of commodities in a store. The publication states, "If there is a [long] empty space on the shelf and there are no [earnings] and [long] products, and there are [sells] and [long] products,
[Top selling] and [Long] products [1] Face [increase]
If the sales amount is a high level and the gross profit margin is high, it is regarded as a profitable product, and the sales amount is high and the gross profit margin is medium level. The product arrangement is automatically determined based on the characteristics of the product based on the past sales results such that the product becomes a hot selling product if it exists.

[0004] In the actual retail business, however, customers who buy other commodities flow to other stores in a merchandise configuration that is only sold well. In addition, since special-sale (low-sale) products have small profits, there is a problem in that the effect of special sales cannot be exhibited unless a floor layout and shelf layout are considered, in which not only special products but also other related products are bought. is there.
For this reason, simply making a decision by focusing on one task does not lead to the result that the person in charge of the task realizes the final management indexes such as sales, the number of customers, and profit.

Japanese Patent Application Laid-Open No. 6-35896 discloses that
A technology for predicting sales when a store is opened in a commercial area is disclosed. This publication discloses means for obtaining an estimated sales amount of a store based on the number of customers that can be visited in a commercial area set around a new store opening location, and the traffic volume and neighborhood of a road in contact with the new store opening site. By using the means to determine the estimated sales value of the store based on the location conditions of the store, the expected sales value for each factor is directly calculated from the past statistical data,
Further, a knowledge base in which influence determination knowledge indicating an influence degree between each factor is stored in advance, and a means for correcting each of the calculated sales prediction values based on the influence determination knowledge in the knowledge base, The means for obtaining the predicted sales output value by adding the predicted sales values afterward is used to correct the sales.

However, in such a method of directly using the past statistical data, the same tendency as the past data is applied to the present or the future, so that the state inside and outside the store or trade area to be predicted, If the environment changes non-monotonically, of course, past data cannot be used. In particular, the conventional method based on past statistical data of at least one year or so, in accordance with the rapid change in the business conditions of the retail, restaurant and service industries, or the frequency of opening and closing Not only did it fall, but the prediction itself became meaningless.

Further, a method of simulating the behavior of a consumer in a store is disclosed in Japanese Patent Application Laid-Open No. 64-88215. The method includes a consumer trend detecting means for detecting a trend of a consumer in a store in time series, a branch position detecting means for detecting a branch position at which a walking direction changes from the detected consumer trend, and a branch position. A branch rate calculating means for calculating a ratio of each consumer in the branch direction in the branch point, a branch point or a branch of a consumer's flow line at a unique position in the store, which is statistical data from the consumer trend detecting means. The simulation is performed directly based on the rate. Similarly, regarding the simulation of the consumer's product selection behavior, the stoppage rate, the contact rate,
The purchase rate is used directly.

According to this simulation method, it is possible to simulate the flow of a consumer when the arrangement of products in a store is changed. When there is no such data, the user of the simulation inputs an arbitrary value for the data such as the branch rate, and the accuracy of the simulation is extremely low. In addition, since the statistical data of the detected stop rate, contact rate, and purchase rate themselves are not only related to the attribute of the target product but also to the product placement of the product in the store, the product placement was changed. Sometimes the same stop rate, contact rate, and purchase rate values cannot be used for the same product. In addition, the stoppage rate, contact rate, and purchase rate of one product are essentially related to the product placement of all other products in the store, not just the target product. The accuracy of the sales simulation when the product arrangement is changed by using the above is hardly expected.

[0009] On the other hand, there is a method in which statistical data for many types of product arrangements is sufficient, but in an actual store, product arrangements are frequently changed in order to obtain data for many product arrangements. Is costly and difficult because it causes unnecessary confusion for consumers. Even if it is possible, it is data at different points in time series, and even if it is possible to collect data from many stores at the same time, actually the trade area and customer demographics differ for each store. Therefore, it is difficult to use the data when the product arrangement is changed.

[0010]

A management decision support system for management, marketing, merchandising, and operation performed by business managers in the retail business, the food business, the service business, and the like has conventionally required individual store opening plans and shelves. There are several examples of split plans that have limited functions, such as store opening plans, business strategies, floor layouts, assortments, shelf splits, sales promotions, pricing,
There is nothing to support comprehensive business planning and comprehensive evaluation and review of specific business strategies and actions such as inventory setting and product ordering.

However, in the actual management of the retail business, the restaurant business, the service business, and the like, these store opening plans, management strategies, floor layouts, assortment, shelf layout, sales promotion, price setting, stock quantity setting, product ordering, etc. Constitutes a complex system that interacts, so even if one simply focuses on one business and makes a decision, the final management indicators such as sales, number of customers, profits, etc. that the business person in charge aims at It does not lead to the result of realizing.

Further, conventional simulation methods such as product placement and sales forecasting do not simulate the purchasing behavior itself such as individual shopper / customer shop selection, product / service selection, etc. Is based on statistical data such as when, where, who, what (why do you like the product itself, price, atmosphere of the store, clerk's response, etc.), and how to purchase It is not a simulation that expresses what to do. For this reason, it is not possible to accurately grasp the generation process of the customer's purchasing behavior, and it is difficult to provide effective decision support for complex specific strategies and actions.

Furthermore, in these statistical methods, for example, a known regression analysis has been generally used, but the same tendency as past data is applied to the present or the future. If the state or environment inside or outside the target store or trade area changes, past data cannot be used. In addition, even if past data may well apply, for example, if sales decline, a new idea such as strategies and actions that will help maintain and improve sales will be recognized by itself. Is not directly linked to the creation of Especially,
In order to verify a new idea such as opening a store in a new form or changing the layout operation in the store, the statistical processing itself is impossible because there is no past performance data on the idea.

Therefore, the invention according to claims 1 to 7 is a management simulation method capable of faithfully simulating a product / service giving / receiving action and a business action generation system by a relatively simple algorithm to actual human actions and product sales states. I will provide a.

[0015] The invention according to claims 8 and 9 is intended to provide a business strategy for a specific business, a comprehensive planning of actions, and a comprehensive evaluation study to be performed by a business manager in the retail business, the food business, the service business and the like. A management decision support system that verifies ideas for new strategies and actions of the business manager and recommends appropriate strategies and actions necessary for the business manager to reach the target business state at the time of decision making. I will provide a.

[0016]

According to the first aspect of the present invention,
A target space consisting of an array of multiple cells is set for the product / service giving / receiving action and business action generation system that is the target of the simulation, and within this target space, the product / service giving / receiving action and business action including movement are performed. By setting a plurality of target elements that perform interaction related to occurrence and arranging predetermined target elements on cells in the target space, the product / service transfer behavior and business activity generation system are represented as data as target areas by data, and The target element interacts with the recognizable target element within the set scanning range in the target area, performs a product / service transfer action including a movement and a business action including movement according to the set action rule, and outputs the action result. Accordingly, the data expressing the target area is changed, and this is repeated for a plurality of target elements, so that the product / service giving / receiving action and business action are generated. The time-series state change of the system, in management simulation method of simulating the change in the data of the target area.

According to a second aspect of the present invention, in the management simulation method according to the first aspect, the predetermined target element is:
In the target area, interact with recognizable target elements within the set scanning range based on mutual data, perform goods / service giving / receiving operations including movement according to the set behavior rules, and business activities, and The purpose of the present invention is to update the data held by each other according to the result of the action, and thereby change the data expressing the target area.

According to a third aspect of the present invention, in the management simulation method according to the first or second aspect, after one of the predetermined target elements performs a product / service giving / receiving operation and a business operation including movement, the next target element is executed. Before the processing of the action is started, the data expressing the target area is immediately changed according to the action result of the one target element.

According to a fourth aspect of the present invention, in the management simulation method according to any one of the first to third aspects,
The target element is to include a cell located at a position farther from a cell adjacent to the cell at the current position in the scan range.

According to a fifth aspect of the present invention, in the management simulation method according to any one of the first to fourth aspects,
The target element is an action element that performs product / service transfer and business actions including movement, a passive element that operates passively in response to this action element, and an area element that restricts the movement of the action element and the placement of the passive element. First, the area element is arranged on the cell of the target space, and the passive element is arranged on the area element to constitute the target area corresponding to the goods / service transfer behavior and business action generation system. In the specified target area, only the action element interacts with the recognizable target element within the set scanning range based on the data held by each other, and the product / service transfer action and business including movement according to the set action rule. Performs an action, updates the mutual data of the action element and the passive element based on the result of the action of the action element, and repeats the processing in the action of the action element only for the product / service. It is to simulate the sequence of state changes when the bis exchange behavior and operational behavior generation system.

According to such a method, a product / service giving / receiving action and a business action generating system can be described by a target space and a target element, and a target area faithful to reality can be expressed.
In addition, the target element scans the target area for a partner who wants to interact with each other, interacts based on each other's data, and, in accordance with the set behavior rules, goods and goods including movement.
Since the service transfer behavior and the business behavior are performed, the basic part of the actual human behavior can be well expressed, and the behavior of the store, the business area, the customer, the employee, etc. can be expressed by one basic algorithm. Furthermore, since the details of actions and interactions can be described by data, a flexible and varied action pattern is possible. Then, since the data is described according to the result of the action, a change in a time series can be expressed.

According to a sixth aspect of the present invention, in the business simulation method according to any one of the first to fifth aspects,
That is, a newly divided space is made to correspond to each cell constituting the target space as a sub-cell.

According to a seventh aspect of the present invention, in the management simulation method according to the sixth aspect, the object space is expressed as a two-dimensional cell array and the sub-cell space is defined as two-dimensional.
It is expressed as an array of dimensional cells, and one of the dimensions of the space of the subcell expresses the height dimension in the two-dimensional cell array in the target space. With such a method, three-dimensional behavior can be expressed relatively easily.

According to an eighth aspect of the present invention, a product / service giving / receiving action and a business action generation system are represented by data as a target area, and a time-series state change of the product / service giving / receiving action and a business action generating system is set as a target area. Means for simulating by a change in the data of the above, means for securing initial state values and initial estimated values from the actual goods / service giving / receiving action and business action generating system at the start of the simulation by the simulation means, and simulation by the simulation means At the end, means for securing the final state value whose initial state value has finally changed and the final estimated value whose initial estimation value has finally changed, and the initial state from the actual product / service transfer behavior and business behavior generation system The value corresponding to the state value including the state value and the final state value is regarded as the actual state value. The actual state data collection means and the initial state value or the initial estimated value to be adapted so as to match the final state value or the final estimated value to be compared with the set adaptation target. Model adaptation that searches for the value of the adaptation target that should match the target to be compared with the final state value or final estimated value to be compared by changing the adaptation target and performing the simulation by the simulation means. And means for supporting business decision making.

According to the ninth aspect of the present invention, a product / service giving / receiving action and a business action generating system are represented by data as a target area, and a time-series state change of the product / service giving / receiving action and a business action generating system is set as a target area. Means for simulating by a change in the data of the above, means for securing initial state values and initial estimated values from the actual goods / service giving / receiving action and business action generating system at the start of the simulation by the simulation means, and simulation by the simulation means At the end, means for securing the final state value whose initial state value has finally changed and the final estimated value whose initial estimation value has finally changed, and the initial state from the actual product / service transfer behavior and business behavior generation system The value corresponding to the state value including the state value and the final state value is regarded as the actual state value. The actual state data collection means and the initial state value or the initial estimated value to be adapted so as to match the final state value or the final estimated value to be compared with the set adaptation target. Model adaptation that searches for the value of the adaptation target that should match the target to be compared with the final state value or final estimated value to be compared by changing the adaptation target and performing the simulation by the simulation means. Means,
Estimated value adapting means for obtaining a fitted estimated value which is an estimated value adapted to the present, and from the present to the future set by the simulation means using the adapted estimated value obtained by the estimated value fitting means, It is provided with time extrapolation simulation means for simulating the product / service transfer behavior and business behavior generation system, and goal attainment simulation means for finding the current initial state value that matches the set fitness target, and the conformity target is a real product.・ The current state value of the service giving / receiving action and business action generation system, the comparison object as the final state value, the adaptation object as the initial estimated value, and the adaptation by the model adaptation means from the past set time to the present. A search for the target value is performed, and a matched estimated value that is an estimated value that has been matched to the present by the estimated value matching means is obtained. The comparison target is set as the final value state value, the adaptation target is set as the initial state value, and from the present time to the future, the value of the adaptation target is searched by the model adaptation means, and the simulation of the target attainment is performed. The present invention is a management decision support system for finding a current initial state value that matches a matching target set by a means.

With such a system, data is collected from a real system, and by adapting the estimated value that compensates for the lack of the data collection to the reality, the real business situation can be accurately grasped. In addition, since the user is verifying his / her own ideas and searching for candidates for achieving the goal, it is possible to support decision-making in a wide range of specific businesses such as the retail business, the restaurant business, and the service business.

[0027]

Embodiments of the present invention will be described with reference to the drawings. (First Embodiment) FIG. 1 shows an apparatus for simulating a product / service giving / receiving action and a business action generating system by a computer. A simulation computer 1 for executing a simulation; an input device 2 such as a keyboard and a mouse for inputting various instructions, simulation conditions, simulation data, and the like to the simulation computer 1;
It comprises an output device 3 such as a graphic display for displaying operation screens and the like and a printer.

The simulation computer 1 has a CPU
(Central processing unit) 4, a main memory 5 for storing an operation program of the CPU 4, a simulation program, simulation data, and the like in a storage area, and a storage device 6 such as a magnetic disk for storing the above programs, data, and simulation results. As shown in FIG. 2, a program storage unit 7a for storing various programs is formed in the main memory 5 for executing a simulation program for setting data and configuring data.
And an element arrangement state table 7 in which various data are set.
b, target element state table 7c, action element to target element reaction coefficient setting table 7d, action element state table 7
e, the target behavior response coefficient setting table 7f for the behavior element,
Business time division status table 7g, time increment setting table 7h, action element attribute status master table 7i, stationary business status table 7j, history file 7k, action element scanning capability setting table 7m, target element scanning result data table 7n, action limit value Setting table 7p, action distance limit value setting table 7q, customer agent status table 7r,
An accounting unit status table 7s, a store accounting operation status table 7t, and the like are formed.

FIG. 3 shows an example of cell division of a space to be simulated (object space) corresponding to a real in-store space. A two-dimensional object space 8 is an array of a plurality of square cells 9. Is represented by Although a hexagonal cell can be used as a cell, a square cell is more suitable for expressing the inside of a building and the like from the viewpoint of the actual layout, and is therefore preferable.

In each of the cells 9, for example, for the cell of interest 9c, if i and j are arbitrary integers, position data (position coordinates) composed of a combination of two numerical values such as {i, j}. Corresponds. From the position data, all cells in the target space 8 can be identified. Therefore, the target space 8
Can be expressed as follows using the eight cells adjacent to the cell of interest 9c.

The position data of the cell 9w on the left is represented by {i−
1, j} The position data of the cell 9e on the right is {i + 1, j} The position data of the cell 9n on the upper side is {i, j-1} The position data of the cell 9s on the lower side is {i, j + 1位置 The position data of the cell 9nw adjacent to the upper left is {i-1, j-
1} The position data of the cell 9ne on the upper right is {i + 1, j−
1} The position data of the cell 9sw adjacent to the lower left is {i-1, j +
1} The position data of the lower right cell 9se is {i + 1, j +
1｝ This target space 8 simply shows a spatial spread as it is, and in order to correspond to a realistic store, a plurality of elements (target elements) to be simulated are: It is necessary to arrange on the target space 8.
This combination of the target element and the target space is called a target area to be simulated.

FIG. 4 shows an example of the target area 11 corresponding to a realistic in-store layout. The target area 8 shown in FIG. Target elements called area elements (areas) are appropriately arranged. That is, a passage area 12 representing a passage or the like where customers and employees can move freely, a boundary area 13 representing walls, windows and the like at boundaries that limit the movement and visibility of customers and employees, and products and sales promotion. A shelf area 14 representing a table on which fixtures, accounting machines, and the like in which in-store advertisements (POPs) are arranged is arranged.

In this area element, a target element called a passive element (unit) is appropriately arranged. That is, as a passive element, there is an accounting unit 15 that represents an accounting machine (POS register) and its related equipment, and the like.
This is arranged in the passage area 12 or the shelf area 14. As a passive element, there is an entrance / exit unit 16 that expresses an entrance / exit, which is arranged in an area element on the boundary between the inside and the outside of the store. With such zone elements and passive elements,
The target area 11 corresponding to a realistic in-store layout can be generated.

FIG. 5 shows an example of the target area 21 corresponding to a realistic sales floor configuration. In the target area 11 corresponding to the realistic in-store layout shown in FIG. Group), an in-store advertising matter (POP), etc., as a passive element, mainly as a shelf area 14.
It is located above. Thereby, it is possible to represent a sales floor in a store where there are no customers or employees.

In FIG. 5, a plurality of sales units 22 are arranged in all the cells on all the shelf areas 14.
2 and the square cell 9 are omitted, and the name of a specific product group represented by each sales unit 22 is described in the shelf area 14 instead. In addition, each product that represents eating and drinking corners, toilets, parking lots, service counters, employee entrances, food processing rooms, warehouses, offices, employee rest rooms, etc. according to realistic retail stores It is also possible to set target elements for services, services and tasks.

FIG. 6 shows a product / product in a realistic store.
An example of the target area 31 corresponding to the situation of service transfer and related business occurrence, in which an action element called an agent is added to the target area 21 corresponding to the actual sales floor configuration shown in FIG. It is. As this behavior element, there is a customer agent 32 representing a customer in the store (one person and a group performing the same behavior), and can move on the passage area 12 and the entrance / exit unit 16. Also,
There is an employee agent 33 representing an employee (one person and a group performing the same action) in the store, and can move on the aisle area 12 and the entrance / exit unit 16.

FIG. 6 shows a simulation of a product / service giving / receiving action and a business action generation system in a store. A customer purchases a product in an actual store and replenishes an employee's product in the store. Expressing (simulating) such behaviors as behaviors of the customer agent 32 and the employee agent 33 in the target area 31 in accordance with a simulation algorithm described later forms a main part of the simulation of the present invention.

The method of setting and arranging the target space and the target element is preferably a method that faithfully represents the actual situation in the store as much as possible. In consideration of the balance with the easiness of simulation, such as performance and data storage capacity, it is desirable not to make very detailed settings at the expense of some degree of fidelity as a real model.

An example of setting a target area will be described below.
In setting a cell of a target space, it is necessary to adopt cells of the same shape and size for one target space in a store in order to easily create a simulation algorithm. The size of the square cell 9 in the target space may be determined based on a distance corresponding to an average step of a person or based on an average size of furniture. The cell size was set from among the average standard dimensions of furniture to express 比較 的 relatively easily. Specifically, for example, the length of one side of the square cell 9 is set to about 150 mm,
0 mm, about 600 mm, etc. Among them, a square cell with a side of about 300 mm (± 100 mm) is almost equal to the minimum floor area occupied by an average adult, which is convenient for creating an algorithm, the fidelity of the model and the ease of simulation. Excellent balance.

In FIG. 6, the square cell 9 is drawn as a square cell having a side of about 600 mm so that the square cell 9 can be easily seen. Note that in a store having a relatively large store area and less congestion, the number of cells can be reduced by making the square cells each having a side of about 600 mm, thereby shortening the calculation time of the simulation. The basic setting of the target element is
It is classified into the following three types based on the difference in functions.

As for the area element, only one area element can be arranged on one cell. The function of this area element is to restrict movement of movement elements and spatial scanning described later, placement of passive elements and limitation of passive directions described later, etc. The direction that can be restricted is set. A plurality of passive elements can be arranged in the area element on one placeable cell. The function of the passive element acts passively on the behavior element, and induces (activates or suppresses) the behavior of the behavior element, and sets the operable direction. The action element can be placed (entered / appear) at a plurality of positions in the area element and the passive element on one cell that can enter and exit. The function of this behavior element is to interact with the behavior element and with the passive element and the area element, and set the direction in which the movement and action are possible, but the actual direction of the movement and action is determined by the simulation algorithm. decide.

If the above basic settings are satisfied, it is possible to appropriately set the target element in accordance with the characteristics of the actual store, and a simulation algorithm described later can be applied as long as the basic settings are followed. Also, if the behavioral element is restricted to only one position (entering / appearing) in the area element and the passive element, the model's fidelity to this restriction is that the square cell is occupied by an average adult It can be achieved relatively easily in terms of being approximately equal to the minimum floor area.

Data regarding the arrangement of these target elements in the target space is recorded in the main memory 5 in an element arrangement state table 7b shown in FIG. It is also stored in the storage device 6. In the element arrangement state table 7b, a zone element arrangement state table 35, a passive element arrangement state table 36, an action element arrangement state table 37, and the like are set.

The area element arrangement state table 35 is set for the position data of all the cells included in the target area 31, and the type of the area element arranged on the cell, the element code of the area element, The element name of the area element is stored. This element code is associated with all target elements on the target area 31 on a one-to-one basis, whereby individual identification of the target elements can be performed.

The element code of the area element is, for example, a point "."
For example, in the element code “11.0001.0001”, the leading number group “11” indicates a boundary area among the types of area elements,
The middle numeral group “0001” indicates “wall” in the boundary area type, and the last numeral group “0001” indicates “wall 1” which is the first wall, corresponding to the element name “wall 1”. doing.

Similarly, element codes are set for other area elements, and the leading numeral group "12" indicates a passage area among the types of area elements, and the intermediate numeral group "0".
“001” indicates the “passage” in the type of the passage area, the last numeral group “0001” indicates the first passage “passage 1”, and corresponds to the element name “passage 1”. Similarly, the leading numeral group “13” indicates a shelf area among the types of area elements, the intermediate numeral group “0001” indicates “shelves” among the types of shelf areas, and the last numeral group “ “0001” indicates the first shelf “shelf 1” and corresponds to the element name “shelf 1”.

The passive element arrangement state table 36 is set for the element codes of all the area elements included in the target area 31 and including the passive element, and all the passive elements arranged in the area element are set. The type of the passive element, the element code of the passive element, and the like are stored.

The element code of the passive element is, for example, a point "."
For example, in the element code "23.0001.0001", the leading numeral group "23" indicates a sales unit among the types of passive elements, and the intermediate numeral group "0001" Indicates the product group “Soy sauce” among the sales unit types, and the last number group “0001”
Indicates an individual product "1L of company soy sauce (liter)" in the product group "Soy sauce", and corresponds to the element name "1L of company a soy sauce". Similarly, the last numeral group “0002” corresponds to the individual product “1L (liter) of soy sauce of Company b” in the product group “Soy sauce”.

Similarly, element codes are set for the other passive elements, and the leading numeral group “21” indicates the entrance / exit unit among the types of passive elements, and the intermediate numeral group “0001” represents the entrance / exit unit. Among the types, “front doorway” is shown, and the last group of numbers “0001” indicates the individual doorway unit “front doorway 1” in “front doorway”, and corresponds to the element name “front doorway 1”. . Similarly, the leading numeral group “22” indicates an accounting unit among the types of passive elements, the intermediate numeral group “0001” indicates “accounting machine” among the accounting unit types, and the last numeral group. "0
“001” indicates an individual accounting unit “accounting machine 1” in the “accounting machine”, and corresponds to the element name “accounting machine 1”.

In the passive element arrangement state table 36, the row for the area element in which the accounting unit is arranged and the row for the area element in which the entrance / exit unit is arranged are combined with the area element arrangement state table 35 to obtain the layout in the store. This is called a state table, and is used when generating the target area 11 corresponding to the in-store layout shown in FIG. In the passive element arrangement state table 36,
A table including rows for the area elements in which the sales units are arranged is called a sales floor configuration state table, and is shown in FIG.
Is used when generating the target area 21 corresponding to the sales floor configuration shown in FIG.

The action element arrangement state table 37 is set for the element codes of all the area elements included in the target area 31 and including the action element. , The element code of the action element, and the like.

The element code of the action element is, for example, a dot "."
For example, in the element code "31.0001.0001", the leading numeral group "31" indicates a customer agent among the types of action elements, and the intermediate numeral group "0001" Indicates “male” among the types of customer agents, and the last group of numbers “0001” is an individual customer “A-sama” among the types of customer agents with “male”
, And corresponds to the element name “A-like”. Similarly, the middle numeral group “0002” indicates “female” among the types of customer agents, and the last numeral group “0002” indicates “female”.
Correspond to each customer "B-sama" in the customer agent.

Similarly, element codes are set for other passive elements, and the leading numeral group “32” indicates an employee agent among the types of action elements, and the intermediate numeral group “0011” indicates the employee number. The “male employee” is shown among the agent types, and the last group of numbers “0001” corresponds to the individual employee “employee D” in the “male employee”. Similarly, the middle group of numbers “0022” indicates “female part-time employees” in the type of employee agent, and the last group of numbers “0002” indicates individual women in “female part-time employees”. Corresponds to employee "Part E".

Using such a table, a target area corresponding to a realistic store is constructed as a simulation model. The realistic store at this time does not only mean a real store, but also includes an ideal store and a hypothetical store. Next, an example of an in-store simulation algorithm will be described.

FIG. 8 shows an algorithm of the whole in-store simulation, which is an algorithm for processing two types of action elements in the target area 31 in the store, that is, the actions of the customer agent 32 and the employee agent 33. First, in step S1, simulation initialization is performed. Regarding the setting of the target area, an element arrangement state table preset to the target area 211 level corresponding to the sales floor configuration is called from the storage device 6 and the main memory 5 is set.
To be stored.

Also, the user can input from the input device 2 settings necessary for executing the simulation, such as setting of a simulation period (start date / time, end date / time), or the like.
Called from the storage device 6. In a realistic store,
Since the purchaser is affected by the purchase of the product from the season, the day of the week, etc., the date is indispensable in setting the simulation period.

The next loop L1 is an action processing loop in the simulation period of the action element, in which the action processing from the simulation start date / time to the simulation end date / time is repeatedly performed.

At step S2 in the loop L1, business hours such as within the opening preparation time, during the business hours, during the ending preparation time, outside the business hours, etc., corresponding to the actual store are given to the simulation time. . This refers to the business time division corresponding to the date and time preset in the business time division state table 7g.

In the next step S3, in the action processing loop L1 within the action element simulation period, a time increment, that is, a time step, for the business time section is given. This refers to the increment value preset in the time increment setting table 7h. For example, the simulation can be accelerated by setting one time step as one second during business hours and one hour as one time step outside business hours such as regular holidays. This one time step is the action processing loop L2 within the unit time of the next action element.
In the unit time. This time step is used, for example, to set the number of cells to which the action element can move per time step when setting the movement speed of the action element, and is necessary to make the action of the action element realistic. Therefore, it is referred to as a simulation target time, and the above-described target space, target element and this target time are collectively referred to as a simulation target region.

In the next step S4, a process for entering the action element into the store (in the target area) is performed, and the customer agent and the employee agent are processed. First, for a customer agent, a customer entering the store at each time is determined based on a preset entry rule for the customer agent, for example, statistical data or a commercial area simulation described later. That is, one or more action element codes (customer codes) corresponding to the customer agent entering the store are extracted from the action element attribute state master table 7i in which the customer agent is registered in advance.

The behavior element attribute status master table 7
In i, the customer's age, gender, address, purchase history, etc. are recorded for the action element code. At present, this kind of data can be obtained relatively easily with a customer service card of the store or the company to which the store belongs, which has various benefits such as discounts.

The entry rule for the employee agent is based on a preset employee staffing plan (work plan).
In the same manner as the customer agent, the employee who enters the store at each time is determined based on the above. That is, one or more action element codes (employee codes) corresponding to the entering employee agents are extracted from the action element attribute state master table 7i in which the employee agents are registered in advance.

A process number is assigned to each of the action elements entering the store, for example, using random numbers or the like. Specifically, the extracted action element code is made to correspond to the processing number. As a result, this behavior element customer is subjected to behavior processing until the customer leaves the store. The initial position of the customer agent is in the entrance / exit unit 16 as shown by the customer agent 32a entering the store in FIG. The entrance unit set similarly for the employee agent is set to the initial position. Some of these entry rules do not enter the store, and at this time, no action element code is extracted and the processing proceeds to the next loop L.
Proceed to 2.

The loop L2 is an action processing loop of the action element within a unit time, and is repeated until the action processing is completed for all the action elements in the store.
In this loop, processing is performed in the order of processing numbers assigned to action elements. If there is no action element in the store, the process immediately exits this loop and proceeds to step S8.

In step S5 in the loop L2, the process branches to each action process according to the type of the action element. Then, in step S6, the action processing of the customer agent within the unit time is performed. In step S7, the action processing of the employee agent within a unit time is performed. When the action processing of the loop L2 ends in this way, the next step S8 is performed.

In step S8, a stationary work which is necessary for the business of the retail store but does not require the simulation of the behavior of the employee agent in the store is performed. Here, processing is performed based on the stationary work status table 7j in which the contents of the work are described. The stationary work includes, for example, work related to sales promotion such as flyer distribution to a business area to which the store belongs, payroll, creation of a staffing plan, work related to accounting in the store, work related to store management, and the like. These operations relate to the behavior of the customer agent in the store regarding the purchase of goods and the behavior of the employee agent regarding the business. It is also necessary to calculate the profit and loss of the whole store.

In the next step S9, hh = hh + hh
The processing of u, mm = mm + mmu, and ss = ss + ssu is performed, and the time is updated according to the time increment. In the next step S10, the date is updated. At this time, the main memory 5
The contents of the upper history file 7k are stored in a storage file in the storage device 6. As a result, the main memory 5 is released so that a new data storage area can be secured in the main memory 5. The above is the contents of the behavior processing loop L1 within the behavior element simulation period. After the loop ends, the in-store simulation ends. In addition,
The display of the status of the simulation is output by an output device 3 such as a graphic display or a printer as appropriate in each process.

The algorithm of the entire in-store simulation shown in FIG. 8 expresses that the action processing of the action element is performed according to a time step corresponding to a realistic time. This is because the linear statistical analysis represented by multiple regression analysis, which has been often used in the field of retailing, has been used for a certain period of time, such as sales forecasting, hot selling product analysis, and related purchased product analysis. This is an essential difference from the one that calculates the total value.

FIG. 9 shows an algorithm of the action processing of the customer agent in step S6 of FIG. In this processing, first, in step S11, the next search action is performed in step S12 according to the Mode variable each of which has the action to be taken by the customer agent.
Alternatively, the flow branches to step S13 of the accounting action. The initial value of the customer agent 32a entering the store is “Mode = search”, and the step S12 of the search action is selected.

As described above, the reason why the search variable and the accounting action are selected by adopting the Mode variable is that once the accounting action is started, the processing of the accounting action is not completed in one time step, but several times. This is because a step is required.

Step S12 simulates the search behavior of the customer agent in the store, scanning the store, searching for the target, and moving and acting. The user agent searches for and purchases a realistic customer product (shopping basket). The selection of the buying behavior and the accounting behavior at the end of shopping, the exit from the store, and the like correspond.

Step S13 simulates an accounting action in which the customer agent accounts for the purchased goods by the employee agent. Step S14
Then, a process of storing a data table necessary for the simulation among various data tables in the history file 7k with date and time is performed. This is because the data table contents required for the simulation are stored in the history file 7k before the update processing in step S15 changes the contents of the data table.

In step S15, since the action element has performed any action, the data table belonging to the action element is an update process for changing the action performed. The contents of the data table are various state tables and setting tables. Thus, the action processing for one customer agent in the unit time is completed, and the process returns to the action processing loop L2 in the unit time in FIG. If there is an unprocessed action element, the action processing is performed.

FIG. 10 shows an example of a search action algorithm, which is the most characteristic part of the present invention. Here, step S in FIG.
The search behavior in the behavior processing of the customer agent shown in FIG. 12 will be described as an example. First, in step S21, the customer agent scans the target area and performs spatial scanning for visually recognizing a target passive element based on the scanning condition. And
The passive element visually recognized as the scanning result is stored in the target element scanning result data table 7n.

As the scanning conditions, it is determined that the scanning method is visual scanning, the target element is a passive element, and the obstacle is a visual obstacle, and the scanning range is the action element scanning ability setting table 7m.
The scanning angle of the corresponding customer agent preset from
The scanning distance is extracted and set. This spatial scanning corresponds to a behavior in which a real customer looks inside the store and visually recognizes a product or the like. Specifically, as shown in FIG. 11, first, attention is paid to a customer agent 41 that performs an action process. In the setting of the scanning range, it is assumed that the scanning angle is 180 ° centering on the front direction 42 of the customer agent 41 performing the action processing, and the scanning distance has a value enough to overlook the entire store. At this time, a blind spot occurs in the viewing obstacle 43 such as a shelf area, and the target area 31 is separated into a viewable range 44 and a viewable range 45.

Then, all the passive elements in the visible range 44 are extracted as being visible. At this time, even when one cell has a plurality of passive elements, all the passive elements are extracted. The extracted element code of the passive element and the distance, azimuth, etc. from the customer agent 41 performing the behavior processing to the passive element are stored in the target element scanning result data table 7n.

Next, in step S22, the degree of attraction of the customer agent 41 to the target element is calculated with all the elements stored in the target element scanning result data table 7n as the degree of attraction calculation target. FIG. 13 shows an example of the calculation method.
The above-mentioned target element state table 7c shown in FIG. 14, the above-mentioned action element to target element reaction coefficient setting table 7d shown in FIG. 14, and the above-mentioned action element state table 7 shown in FIG.
e, a method of calculating using the above-described action element response coefficient setting table 7f shown in FIG. 16 will be described.

First, the target element status table 7e indicates that, for example, when the target element is a sales unit expressing a product or the like,
It is a sales unit status table. This table 7e stores, for the element code of the target element (sales unit), the value of the state of the product indicated by the sales unit and the value of the sales state. For example, values such as price strength (%), discount rate (%), POP strength (%), and freshness (%) are stored. The price strength is a value obtained by dividing the price of the cheapest product in the group of similar products to which the product belongs by the price of the corresponding product. The discount rate is a rate of discount from the list price. The POP (sales promotion advertisement in the store) intensity means that the area of the product POP is the largest POP among similar product groups to which the product belongs.
Divided by the area of The freshness is the reciprocal of the number of days elapsed from the production date to the present. These values are used as stimulus values V
i. The target element state table 7c is a value available from a real store.

The action element to target element reaction coefficient setting table 7d is, for example, a sales agent reaction coefficient table for customer agents. This table 7d
Is set for each behavior element code of the customer agent, and stores, for each passive element (sales unit) code, the state of the product indicated by the sales unit and the degree of reaction of the customer agent to the sales state. For example, price intensity response coefficient, discount rate response coefficient, POP intensity response coefficient,
Stores freshness response coefficients and the like. In addition, the price intensity response coefficient is a coefficient indicating the degree of responding to the price intensity, the discount rate response coefficient is a coefficient indicating the degree of responding to the discount rate, and the POP intensity response coefficient is This is a coefficient indicating the degree of reaction, and the freshness reaction coefficient is a coefficient indicating the degree of reaction to freshness. These values are used as the reaction coefficient Ci. The target element response coefficient setting table 7d for the behavior element is a value that is difficult to obtain from a real customer, and an estimated value is used here.

It is a relatively simple method to obtain the degree of attraction m as m = Vi · Ci from the stimulus value Vi and the response coefficient Ci described above. There is also a method using a well-known sigmoid function using two kinds of reaction coefficients in consideration of the fidelity to reality.

In this state, since the search action aimed at the sales unit of the customer agent is not completed, the calculated degree of attraction m is further suppressed here. The behavior element state table 7e is, for example, a customer agent state table when the behavior element currently being processed is a customer agent. The table 7e is set for each behavior element code of the customer agent, and stores various state values of the customer agent at each time. For example, values such as the total number of purchased commodities, the total amount of purchased commodities, and the time elapsed since entering the store are stored. These values are referred to as stimulus values Vsi. This action element state table 7e is a value available from a real store.

The target behavior response coefficient setting table 7f of the behavior element is, for example, a purchase behavior coefficient setting table of the customer agent. This table 7f is set for each purpose action (purchasing action) of the customer agent, and stores the degree of reaction of the customer agent to various states of the customer agent for each action element (customer agent). For example, the degree of reaction to the purchase behavior, such as the total number of purchased commodities, the total amount of purchased commodities, and the time elapsed since entering the store, is set as a response coefficient. These values are used as the reaction coefficient Csi. The target behavior response coefficient setting table 7f of the behavior element is a value that is difficult to obtain from a real customer, and an estimated value is used here.

It is a relatively simple method to suppress the degree of attraction m by the above stimulus value Vsi and the response coefficient Csi, and finally obtain the degree of attraction m as m = m−Vsi · Csi. There is also a method using a well-known sigmoid function using two kinds of reaction coefficients in consideration of the fidelity to reality.

As another suppression method, when the values such as the total number of purchased commodities, the total amount of purchased commodities, and the time elapsed since entering the store exceed the set limits, the value of the action element-to-target element response coefficient setting table 7f is changed to the purchase value. A method of performing a suppression process such as a sharp decrease in inverse proportion to values such as the total number of products, the total amount of purchased products, and the elapsed time since entering the store by updating the data table in step S15 in the algorithm of the customer agent behavior process in FIG. There is also. Similarly, for the accounting unit 15 and the entrance / exit unit 16, the degree of attraction is calculated from the state table and the reaction coefficient setting table for the action element, the passive element, and the target action as in FIGS. .

In this way, the degree of attraction is calculated for all the passive elements within the viewable range, and the one with the maximum value is set as the maximum attraction degree Mmax. Then, the element having the maximum attraction degree Mmax is set as the attraction element. Further, a target action corresponding to the type of the attraction element is selected. That is,
The purpose action in the case of a sales unit is a purchase action, the purpose action in the case of an accounting unit is an accounting action, and the purpose action in the case of an entrance / exit unit is an exit action. In the case shown in FIG. 11, since the attraction element 46 is a sales unit, the target action is a purchase action.

At this time, if the value of the total number of purchased products, the total price of purchased products, the time elapsed since entering the store, and the like increase due to the suppression process, the degree of attraction to the sales unit (purchasing behavior) decreases, and the relative value increases. Then, the degree of attraction to the accounting unit and the entrance / exit unit increases, and the accounting unit is selected with the element having the maximum degree of attraction as the attraction element.
Thus, it can be expressed that the actual customer has finished purchasing the goods and moves to the checkout machine.

Next, in step S23, a moving target is determined. As shown in FIG. 11, since the attracting element 46 is a sales unit and the customer agent 41 cannot enter and exit the attracting element 46, the passive direction of the attracting element (the direction in which the behavior element can purchase the selling unit as the attracting element). ) Is set as the movement target 47. When the inviting element is the accounting unit 15, it is impossible to enter and exit the inviting element, so that the moving target is determined in the same manner. If the attraction element is capable of entering and exiting the action element as in the entrance / exit unit 16, the position where the attraction element is located is set as the movement target.

Next, in a step S24, it is determined whether or not the obtained maximum degree of attraction Mmax is a value sufficient to cause a desired action. The action limit value setting table 7p stores the action limit value LM for the action element code, the attraction element code, and the target action code, and the determination is made based on these values. If the maximum degree of attraction Mmax is equal to or greater than the action limit value LM,
Proceed to the next step S25. Otherwise, the process proceeds to the wandering behavior processing in step S28. The action limit value LM is also a value including the estimated value.

In step S25, it is determined whether or not the distance DM between the action element being processed and the movement target is within a range in which the action element causes the target action. Since the action distance limit value setting table 7q stores the action distance limit value LDM for the action element code, the attraction element code, and the target action code, the determination is made based on this value. The distance DM between the action element being processed and the moving target is the action distance limit value LDM.
If not, the process proceeds to the next step S26. Otherwise, the process proceeds to the movement action processing in step S27. For example, when the purchase behavior is the target behavior, the behavior distance limit value LDM is an estimated value but can be set to a realistic value because it is determined by a reachable range, that is, a reach length.

In the example shown in FIG. 12, the customer agent 41
Is zero, that is, when the customer agent 41 comes over the moving target 47. Similarly, the attraction element is accounting unit 1
5 and the entrance / exit unit 16, the customer agent 41
It is determined that the target action can be performed when the target reaches the moving target.

In step S26, the target action is performed. The purpose action includes a purchase action, an accounting action, and an exit action. In the case of purchasing behavior, the sales unit of the attraction element is used as a purchased product. Specifically, the number of sales units corresponding to the purchased product is reduced from the sales unit status table 7c, and the element code of the sales unit corresponding to the purchased product and the number of purchases are stored in the customer agent status table 7e. is there. In the case of the accounting action, the attraction element is the accounting unit 15 and the position of the customer agent is near the accounting unit.
It may not finish within the time step. In the case of the exit behavior, when the attraction element is the entrance / exit unit 16 and the position of the customer agent is in the entrance / exit unit, the data table of the exiting customer agent is stored in, for example, a storage file of the storage device 6. Then, the customer code corresponding to the processing number of the leaving customer agent is deleted. As a result, this customer agent will not be processed until a new store is entered.

In step S27, if the distance DM between the action element (customer agent) being processed and the movement target is larger than the action distance limit LDM, the action element is determined as the movement target 47.
Move in the direction of. FIG. 12 shows the situation. 1
When one cell is moved in the time step, as shown by the trace 48 of the movement action, the user moves to the movement target in six time steps and performs the purchase action in the seventh time step. However, even during this movement, the spatial scanning is performed at each time step, and the visible range also changes every time. Therefore, there is a good possibility that the movement target is changed to the movement target of the attraction element with a higher degree of attraction. In addition, since the action processing of each action element is asynchronously performed with each other, the current attraction element may be purchased as an action element to be processed later even within one time step of the current time.

In step S28, if the maximum attraction Mmax is less than the action limit value LM, the action element (the customer agent does not have a specific moving target and wanders in the target area 31. At this time, for example, the most visible These rules are set in advance, such as a direction toward a wide range or a direction in which the moving direction is changed at random.
The processing shifts to step S14 of the behavior processing of the customer agent shown in FIG. The search action described here is a basic action pattern that scans the target area (target space), recognizes other target elements that interact with the action element being processed, and identifies The state is evaluated in a complex manner and the target action is performed. Thus, this search behavior is used not only in customer agents, but also in employee agents and other examples.

FIG. 17 shows an example of an accounting action algorithm. First, in step S31, the accounting mode is determined. In other words, when branching by the Mode variable first comes from the search action to the accounting action, step S3
Proceed to 2. Step S when you are already in accounting
Move to 34. In step S32, the total number of products purchased by the customer agent (the total number of purchased products) Np is extracted from the customer agent status table 7e. And step S
At 33, the accounting is substituted for the Mode variable.

Subsequently, in step S34, the number of products Sr that can be paid per unit time is calculated from the total number of purchased products Np.
Subtract (processing speed). That is, the processing speed for the element code of the accounting unit is extracted from the accounting unit status table 7s. The status table 7s is set in association with the accounting operation of the employee agent.

Subsequently, in step S35, it is determined whether the accounting operation has been completed. That is, in order to end the accounting action when the number of unpaid products Np is exhausted, the next step S36
Proceed to. However, if the number of unaccounted products still remains, the process directly proceeds to step S14 of the customer agent behavior process shown in FIG. Thus, in the action processing of the same customer agent in FIG. 9 in the next time step, the processing of the accounting action in step S13 is performed from the beginning.

In step S36, an accounting record process is performed. That is, since the accounting action for this customer agent has been completed, the store accounting operation status table 7t includes
Date / time, accounting unit code, customer code that performed accounting,
Stores sales unit code, sales quantity (purchase quantity), sales price (purchase price), and the like. Therefore, the store accounting operation status table 7t corresponds to the sales data at the actual store.

In step S37, a search mode selection process is performed to change the Mode variable to search. Thereby, in the action processing of the same customer agent in FIG. 9 in the next time step, the search action processing of step S12 is performed from the beginning. As described above, the accounting operation can be continuously performed through a plurality of time steps, so that it is possible to faithfully represent the state of the accounting operation that takes a considerable time in an actual store.

FIG. 18 shows an algorithm of the behavior processing of the employee agent. The flow of this algorithm is basically the same as the algorithm of the behavior processing of the customer agent shown in FIG. 9, and steps S41, S48, S4
9 is the same as steps S11, S14, and S15 in FIG. Therefore, here, six business actions shown in steps S42, S43, S44, S45, S46, and S47 of FIG. 18 will be described.

These business actions are examples of the business performed in the store, and may be appropriately increased or decreased. In a relatively small retail store such as a convenience store, for example,
A human employee may perform all six tasks. On the other hand, in relatively large stores such as supermarkets, the roles are clearly defined. Further, the business behavior is basically based on the search behavior shown in FIG.
Each target task action and fixed task action are added.

In step S42, sales promotion business action is performed in the store. This business action is for sale, discount, PO
P is installed. Among these, the sales promotion business action whose date and time are set in advance by the business plan can be processed in this step, or can be processed in the stationary work in step S8 in FIG. The business performed in step S42 is a sales promotion business that is not determined in advance or whose details are not determined in detail. For example, in a time service for selling out,
This is a discount on a product that is likely to be unsold in a certain time zone, and is performed on a short-lived product such as fresh food or milk.

In the processing of this sales promotion business action,
Employee agents who perform sales promotion activities to correspond to actual stores refer to POS data (point-of-sale information) generated from an accounting machine or the like. In particular, residual sales units (products) in stores It is necessary to recognize the situation, the congestion situation of the customer agent, and the purchase situation of the sales unit (product). For this reason, the sales agent (product) that the employee agent moves while recognizing the inside of the store by spatial scanning by searching behavior, recognizes the situation of the whole store, and discounts,
An algorithm for determining a discount amount (discount rate) and a discount time zone is employed.

In step S43, an accounting operation in the store is performed. Most of this business action is operation of the accounting machine. There is no need to search for an employee agent who always performs accounting operations in all accounting units in a store, but in an actual store, the number of accounting machines that operate according to the degree of customer congestion is changed. Need to deal with this. For this reason, in the accounting business action processing, the employee agent moves while recognizing the inside of the store by spatial scanning by searching behavior, recognizes the situation of customer congestion in the store, and decides which accounting machine to activate when The algorithm is adopted. FIG. 6 shows the situation of the employee agent 33a during the accounting operation.

At step S44, a product ordering operation in the store is performed. In this business operation, an order is placed while confirming the remaining amount of sales units (products) in the store as in the actual store. That is, the employee agent moves while recognizing the interior of the store by spatial scanning through search action, and recognizes the remaining status of the sales units (products) in the store and determines the code, quantity, and time of the sales unit (product) to order. The algorithm to decide is adopted. It should be noted that since this product order has a set period from the time the order is placed until the product is delivered, an algorithm that takes this into consideration may be employed.

In step S45, a product replenishment operation in the store is performed. This business action checks the remaining amount of the sales units (products) in the store and replenishes the small amount as in the actual store. In other words, the employee agent moves while recognizing the inside of the store by spatial scanning by search action, recognizes the remaining state of sales units (products) in the store, and replenishes sales units (products) with a small remaining amount each time. The algorithm is adopted. FIG. 6 shows the situation of the employee agent 33b during the product replenishment operation.

In step S46, the merchandise disposal operation in the store is performed. In this business action, the expiration date of the sales unit (product) in the store is confirmed in the same manner as in the actual store, and those whose expiration date has expired are discarded. That is, the employee agent moves while recognizing the inside of the store by spatial scanning through search action, grasps the expiration date of the sales unit (product) in the store, and sells the sales unit (product) whose expiration date expires.
Is discarded each time.
At this time, in contrast to the behavior processing of the customer agent, the merchandise to be discarded is stored in the store accounting operation status table 7t by the accounting machine with a negative amount. That is, it is recorded as a loss on the store side.

In step S47, a cleaning operation is performed in the store. In this business operation, in addition to the overall cleaning, there is also a business in which a dirty portion in the store is checked and the area is mainly cleaned, as in the actual store. Here, an algorithm is used in which the employee agent moves while recognizing the interior of the store by spatial scanning through search behavior, recognizes the cleanliness of all target elements in the store, and focuses on cleaning the target element and its surroundings doing.

With the above algorithm and the data based on the setting of the target area, a program for simulating the behavior of the customer agent and the employee agent in the store can be constructed, and this program is executed by the simulation computer 1 shown in FIG. In addition, the behavior simulation of the customer agent and the employee agent in the store can be realized. The simulation model consisting of the algorithm and the target area represents the spatial and temporal situation in the actual store and the complex interaction between the elements as faithfully as possible. Therefore, it is a realistic model,
In addition to being able to reproduce the inside of a real retail store in the simulation computer 1, it also performs spatial and temporal expansion that does not exist in a real store, and adds complex interactions between elements. However, a realistic result can be obtained as a result of the simulation. For example, it is possible to accurately simulate whether the number of purchased (sales) products increases if the layout in a store is changed, or whether the total price of purchased (sales) products increases if the price of a product is changed. Also, if the target areas 31 shown in FIG. 6 are connected in a plurality of layers, this simulation can be applied to a large store such as a department store having a plurality of floors.

As described above, the product / service giving / receiving action and business action generating system can be simulated faithfully with the actual behavior of a person and the sales state of the product by using a relatively simple algorithm.

(Second Embodiment) In this embodiment, a case where a newly divided space is made to correspond to a sub-cell for each cell constituting a target space will be described.
Simulation in a two-dimensional target area in a store is possible in the formats shown in FIGS. 6, 11, and 12. However, the actual in-store space is three-dimensional, and it is possible to configure the target region formed therein in three dimensions by, for example, dividing the target space by cubic (hexahedral) cells.
The target element is also configured in three dimensions, a more realistic target area is generated, and the simulation can be basically performed by the same algorithm. In this method, the fidelity to reality is improved, but there are problems such as an increase in calculation time due to the three-dimensional space and a shortage of the area of the main memory 5.

Therefore, here, a simulation is performed in which the first embodiment is extended so that the sales units can be arranged also in the height direction only in the shelf area, for example.
That is, FIG. 19 shows a shelf space which is a target of the simulation when the shelf area is viewed from the front, and is referred to as a target shelf space. Similar to the store space, the target shelf space 51 is divided by square cells 52. The divided cells are called sub cells. The position data and the like of each subcell can be set exactly in the same manner as in the in-store space. However, there is a height direction. In addition, as a subcell, in addition to dividing a space in the height direction, a flat shelf space arranged on a plane may be further divided into a plurality of spaces.

In FIG. 19, the sales unit 53 has already been arranged and is configured as the target shelf area 54. The target shelf area 54 belongs to each shelf area in the store. The setting of the target element in the shelf space is performed by using the boundary area 55 as the area element.
And a sub-shelf area 56. The boundary area 55 is a boundary that restricts movement and visibility of customers and employees, and expresses a wall (partition plate), a window (transparent partition plate), and the like. Also,
The sub-shelf area 56 expresses a space of furniture in which goods and in-store advertising materials (POP) for sales promotion are arranged. There is a sales unit 53 as a passive element. The sales unit 53 expresses a product, a POP, and the like, and is arranged on the sub shelf area 56. Further, there are a customer agent and an employee agent as action elements. The customer agent represents the shopper in the store. Employee agents represent employees in the store.

For these target elements, as in the case of the in-store space, the element code and the position data are stored in the shelf space element arrangement table together with the element codes of the shelf area in the store to which they belong. The cell size in the shelf space is smaller than that in the store, and one side of 50 to 150 mm is appropriate,
If the space is set to 300 mm, which is the same as the space in the store, the effect of simplifying the algorithm is great.

Next, a search operation for a shelf having a sub-shelf area formed in the height direction will be described. First, considering that the customer agent performs the purchase behavior as shown in FIG. 12 in the in-store simulation, in step S26 in the processing of the search behavior shown in FIG. 10, the target behavior is not the purchase behavior but the shelf search behavior. select. This shelf search action is basically the same as the search action algorithm in the store shown in FIG. 10 except that the search target space is within the target shelf space. It is done within.

In the first step, a spatial scanning process is performed. In this spatial scanning process, the only scanning target element is the sales unit. However, the scanning range is significantly different, and FIG.
In the case of the inside of the store, the scanning range was almost fan-shaped, but here, as shown in the viewable range 58 of FIG. 20, the scanning range is almost elliptical, and corresponds to the range of the visual field of human eyes. I have. In FIG. 20, since there is no visual obstacle, the scanning range and the viewable range match. The space outside the visible range is the non-visible range 59. If there is a viewing obstacle, a blind spot occurs and the viewable range is narrowed.

A plurality of target shelf areas belonging to a shelf area in a store are connected to form a target shelf area to be scanned so as to satisfy the size of the scanning range. FIG.
The target shelf space shown in FIG. 20 connects the target shelf areas belonging to the three shelf areas of the shelf area to which the sales unit 46 of the attracting element in front of the customer agent 41 shown in FIG. It is composed. In a shelf search action, the center of the scan range corresponds to the position of the viewpoint center 57 of the action element, just as the top of the fan-shaped scan range is the position of the action element (of the customer agent) when in the store.

In the next step, the process of calculating the degree of attraction is performed. In this process as well, the scanning target element is only the sales unit, and therefore, the target behavior is only the purchasing behavior. The next step is to determine the move target, but the attraction here is visible and can be grasped by hand,
The position where the attraction element exists is set as a movement target 59.

Then, similarly to the case shown in FIG. 10, the action limit determination and the action distance limit determination are performed, and any of the execution of the target action, the movement action, and the wandering action is performed. Then, if the target action is performed, only the purchase action is set.
In the case of a moving action, the viewpoint center of the action element moves toward the moving target. At this time, since the shelf search action is only movement at the center of the viewpoint, the user can move very quickly. Also, since the cell is small, one time step in the shelf search action is considerably shorter than one time step in the action processing in the store. For example, if one time step in the store is one second, one time step in the shelf search action is about 0.1 second. In the case of a wandering behavior, the wandering behavior is almost the same as when the viewpoint center of the behavior element is in the store.

Although the processing of the shelf search action is completed as described above, it is not possible to shift to the search action of the in-store simulation being processed as it is. For this reason, for example, when the maximum degree of attraction is lower than a certain limit, the process shifts to the search behavior of the original in-store simulation. Also,
It is also effective to set a leaving behavior from the shelf space that corresponds to the leaving behavior.

In this way, the in-store simulation to which the shelf search action has been added can faithfully represent the actual product arrangement situation. Therefore, it is possible to accurately simulate whether the number of purchased (sales) products increases when the product arrangement on the shelf is changed.

(Third Embodiment) In each of the two embodiments described above, the in-store simulation has been described. Here, the commercial area simulation will be described.
Real retail stores belong to a real area called a trade area, and a plurality of situations in the trade area affect the operations and business conditions of the store. Therefore, to comprehensively simulate the management of a retail store, it is necessary to simulate the behavior of a customer (purchaser) and an employee in a business area to which the target store (target store) belongs.

This simulation does not represent the movement of people in the entire trade area, the status of goods and services, but the status of business and sales at the target store. Since there is no actual boundary of the trade area, the situation in the area outside the trade area has little effect on the business (sales) of the target store, or it is quantitative even if the degree of the influence does not depend on the simulation The basic definition of a trade area is that it can be set in a specific way.

Here, a commercial area simulation will be described. The method of the commercial area simulation is basically the same as the in-store simulation, and includes the setting of the target area and the simulation algorithm. The same method as the in-store simulation can be used to set the target area representing the business area.

A method for setting a target area will be described with reference to FIG. This figure shows a two-dimensional space, which is a target space 61, which is a target area of a simulation that is a target of a simulation in which a bird's-eye view of the trade area is taken. Similar to the store space, the target space 61 is divided by square cells 62. Cell position data and the like can be set exactly in the same manner as in the in-store space. Here, the target elements are already arranged and configured as a target area 63 in the commercial area.

Next, the setting of the target element in the commercial area space will be described. As the area elements, there are a traffic area 64 and a restricted area 65. The traffic area 64 expresses a traffic network such as a road or a railway line where customers and employees can move freely. An area other than the traffic area 64, a river, a park, or the like is expressed as an area in which the movement of the employee is restricted. Then, one area element is arranged on the square cell 62.

The passive elements include a living unit 66, a store unit 67, and an entry / exit unit 68. The living unit 66 corresponds to a residence where a person in the business area lives, and an address to which the person belongs. The store unit 67 represents a place other than a residential unit that exchanges goods and services, such as a retail store, a restaurant, a bank, a hospital, a school, an office, a factory, and a station. The entry / exit unit 68 represents a boundary where entry / exit between the trade area and the outside is performed. These are usually
Located in the traffic area. And a plurality of passive elements can be arranged in one area element.

Although there is a situation of being a living unit and a store unit, here, both the living unit and the store unit are arranged and expressed in one area element. A space such as a parking lot basically belongs to a living unit or a store unit to which it belongs. Also, passive elements are not located in all restricted areas. This part represents a place where goods and services are not exchanged and cannot be freely passed. For example, a restricted area, an unoccupied house, an unoccupied land where the owner is located, and the like are expressed. It also describes places that are uninvestigated or uninvestigable and for which you do not know what is.

As the action elements, the customer agent 69,
There is an employee agent 70 and a customer agent 69
Represents a customer at the target store and a customer candidate (one person and a group performing the same action), such as a person living in the commercial area and a person moving into the commercial area. Also, the employee agent 70
Expresses the employees of the target store (one person and a group performing the same action). Then, a plurality of passive elements can be arranged in one area element and one passive element. Also, when moving using a means of transportation such as an automobile or a train, the movement speed is responded by paying attention only to the movement of a person.

[0129] In addition, since a large number of customer agents may be handled in a trade area ratio, customer agents are set based on the definition of "groups performing the same action". For example, it is used as a group of students returning home from school to a certain area. These target elements are stored in the element arrangement state table as element types, element codes, and the like, as in the case of the in-store simulation. The position data of the element arrangement according to the real commercial area can be easily constructed from a commercially available residential map or an electronic data thereof.
The setting of the customer's address can also be easily obtained from the aforementioned customer service card.

The size of the cell in the target space in the trade area varies depending on the purpose of the simulation, and is from 1 m to 1 m.
It may be up to km or other. For example, in order to grasp the visibility of the target store in the commercial area, about 1 to 3 m is appropriate, and the limit that can maintain the proportional relation between the area of the store and the road is about 5 to 10 m. Although the proportional relation disappears, the distance relation appears almost exactly when 20 to 6
It is about 0 m.

Further, for example, for a large-scale commercial area having a radius of 5 km or more using a means of transportation, 100 m to 1 m
About km is used. In this case, the distance relationship between elements requires a route to be calculated from data such as a road map. Also, for such a large trade area,
Since linearity and right-angledness of buildings and roads hardly matter, it is better to use isotropic regular hexagonal cells in which all adjacent cells are under the same conditions. In the case of FIG. 21, the target store is mainly open to customers who come to the store on foot, so a square cell of about 60 m on one side sets a 1620 m square commercial area.

The setting of the time step, which is the target time,
There are various factors related to the size of the cell and the moving speed of the action element. Basically, it is desirable to set such that one or more cells can be moved in one time step, and if the cell is 100 m or less on one side, the setting may be one time step of 1 second to 5 minutes. However, to understand the visibility from the car,
There is a case where the setting for one second or less is performed. Even in the case of a cell exceeding 100 m on one side, it takes less than 30 minutes except for a special case. Other settings can be made in the same manner as in-store simulation.

Next, the algorithm of the commercial area simulation will be described. Similar to the in-store simulation, the commercial area simulation is intended to simulate the transfer of goods and services in the target area of the action element and the occurrence of business related thereto. The algorithm of the in-store simulation can be used by slightly changing the setting.

First, the overall algorithm of the commercial area simulation is almost the same as the overall algorithm of the in-store simulation shown in FIG. In the commercial area appearance behavior, an action element that enters the target commercial area from outside the commercial area in the entry / exit unit 68 and an action element that starts an action in the target commercial area from the resident unit 66 are set. Appearance in the trade area is determined with reference to the date and time. For example, the time to go to school or work and the time to work can be determined relatively realistically, and the time to shop from a house to a store can be set by statistical data or personal data from a customer service card.

The algorithm of the action processing of the customer agent in the next commercial area is almost the same as that of the in-store simulation shown in FIG. 9, and the accounting action may be the in-store action. The algorithm of behavior processing of the customer agent is also applied to the algorithm of behavior processing of the employee agent in the commercial area.

The algorithm of the search action in the commercial area is almost the same as the search action of the in-store simulation shown in FIG. 10 except for the following changed parts. The first change is
The purpose action for the type of the attraction element is an in-store action for a store unit, a return action for a resident unit, and an exit action for an entry / exit unit. The exit behavior performs exactly the same processing as that of the in-store simulation in the entry / exit unit. The returning behavior performs exactly the same processing as the above-described leaving behavior in the resident unit. In-store behavior is such that once a customer agent enters a store unit, it stays in the store unit for a predetermined time step and performs a predetermined action. Then, after a predetermined time step has elapsed, the user goes out of the store unit. This algorithm has the same algorithm and method as the accounting behavior of the in-store simulation.

The next change is in the movement behavior, which is basically the same as in the case of the in-store simulation, except that the behavior element can be selected as appropriate for the means of transportation, and the traveling speed is set accordingly. The behavior of the behavior element in the commercial area can be faithfully represented by the algorithm of the commercial area simulation configured in this manner. For example, not only the behavior of going out of the house and going to the store, but also the behavior of going to the store on the way home from work or school, and even the behavior of going to a relatively nearby store in rainy weather, set the response coefficient to the above-mentioned weather etc. It becomes possible by doing. In addition, understanding which regions will increase the number of customers entering the target store when distributing leaflets, and what effects on competition among peers by seeing changes in customer store selection Can be grasped. In addition, if a store is configured at an arbitrary point on the simulation and the number of customers expected can be estimated by the simulation, this becomes an effective judgment material when opening a new store.

[0138] Another purpose of this commercial area simulation is to combine with the store entry behavior of the in-store simulation. In the first combination method, first, a commercial area simulation is performed, and a store entry pattern with respect to the time of the customer agent in the target store is recorded. Next, the customer agent enters the store based on the entry pattern recorded in the entry behavior of the in-store simulation.

The second method is a method in which the in-store behavior of the in-store simulation and the in-store behavior and the exit behavior of the in-store simulation are linked to form an algorithm as a whole. This method can be the same method as the connection between the search action in the shop space and the search action in the shelf space. That is,
What is necessary is just to start the store entrance behavior of the in-store simulation with the element code of the customer agent who wants to start the in-store behavior in the commercial area simulation. Similarly, the in-store behavior may be terminated by the elemental code of the customer agent who intends to perform the exit behavior in the in-store simulation by the commercial area simulation, and the corresponding customer agent may be taken out of the store unit. In this method, it is necessary to ensure consistency since one time step of the simulation is different between the commercial area and the store.

As described above, a comprehensive simulation in which the business area and the store are connected can be performed. This allows
The fidelity of the simulation to the reality can be further increased. In addition, if a simple layout can be obtained for other stores in the same industry, a plurality of in-store simulations can be combined with the simulation of a commercial area, which is effective for more accurately grasping the competitive relationship.

(Fourth Embodiment) Next, a simulation of a product / service giving / receiving operation and a business operation generation system related thereto, as exemplified in the in-store simulation and the commercial area simulation of the retail store in each of the above-described embodiments, will be described. A system for supporting management decision-making for an actual retail store will be described.

Here, the retail store that is the target of the management decision-making support is called the target store, and the target store and the environment surrounding it are called the target management system, and include the inside of the actual retail store and the commercial area. ing. Therefore, a simulation of a product / service giving / receiving action and a business action generating system is simply called a management simulation.

FIG. 22 is a block diagram showing the configuration of the management decision support system. The simulation computer 1, input device 2, and output device 3 are the same as those in FIG.

In the figure, reference numeral 71 denotes a target management system, which indicates the inside of a real retail store or the commercial area. In the target management system 71, various real data are collected by a real data collection unit. Specifically, a time at the time of accounting, a customer identification code, a product name, a product code, a product quantity, and a sales price are used. Point-of-sale data collection device 72 such as a POS system that collects information, etc .; product order data collection device 73 such as an electronic ordering system that collects the time, product name, product code, product quantity, purchase price, etc. at the time of product order; An image data collection device 74 such as a video camera for collecting data on attributes and movements of customers / employees, arrangement of products, and the like as images is provided. The image data collection device 74 is connected to an image data processing unit 75 composed of, for example, a computer and an image processing program in order to extract data such as attributes of the customer, such as age and gender, from the image.

In addition, other data such as weather and temperature,
In-store data such as TV program data, map data of commercial areas, data on traffic volume, data on competitor stores, data on the business status of the store (target store), data on product attributes, data on customer attributes, etc. Or the data necessary for the simulation of the business area and the data necessary for the implementation of the decision support system are collected directly from the target management system 71, or the data of wide-area weather and temperature, the data of the attributes of mass-produced products, etc. , Collect existing data. Here, all means required for data collection are assumed to be actual data collection means.

The simulation computer 1,
Point-of-sale data collection device 72, product order data collection device 73, image data processing means 75, and data storage means 76
Is connected to the network 77.

The actual data collected by the target management system 71 is stored in a data storage means 76 such as a magnetic disk via a network 77. The simulation computer 1, input device 2, output device 3
Is provided with the recommendation information generating means, which is a computer program, to provide the management decision support system user (hereinafter, simply referred to as a user) with recommendation information for management decision support. ing. The program constituting the recommended information generating means is stored in the storage device 6 and the main memory 5.

Next, an example of a functional configuration of the recommended information generating means will be described with reference to FIG. The configuration of the function of the computer program of the recommended information generating means is divided into model adapting means 81, management simulation means 82, and estimated value setting means 83, and various data parts and input / output parts related thereto.

The management simulation means 82 is the program itself for the simulation of the transfer of goods / services and the business generation system related thereto. Regarding the algorithm part of this simulation program,
It is determined by the method disclosed in each of the embodiments described above. Also, regarding the data portion of the simulation program, the status values stored in a status table such as a sales unit status table, which can be obtained from the real target management system 71, and the status values obtained from the real target management system 71, for example. Can be classified into estimated values stored in a setting table such as a customer agent response coefficient setting table for sales agents, for example.

Further, even if the information can be obtained from the actual target management system 71, it is difficult to obtain all the information temporally or economically. These difficult-to-obtain data are also classified into estimated values.

First, the actual data directly collected from the point-of-sale data collecting device 72, the merchandise order data collecting device 73, and the image data collecting device 74 and the real data extracted from the data storage means 76 are stored in the storage device 6 in the form of a real data. It is stored in the state value data table 84. In the actual state value data table 84, data representing the state of the actual management system is stored in chronological order from the past to the present.

The values of the actual state value data table 84 directly correspond to the state values of the management simulation means 82, and are used to start the management simulation.
Used as 5. The estimation value setting means 83 generates an initial estimation value 86 of the management simulation used when starting the management simulation from the values of the actual state value data table 84 using, for example, a well-known statistical method.

The intermediate forced fluctuation state value 87 and the intermediate forced fluctuation value estimated value 88, which are the intermediate forced fluctuations in the management simulation.
Is equivalent to a disturbance to the management system, and the user inputs a value using the input device 2 or selects from a preset data. In the execution of the management simulation, the state value and the estimated value are updated from the respective initial values based on the algorithm at each time step, but the intermediate forced fluctuation state value 87 and the intermediate forced fluctuation value estimated value 8
Numeral 8 is a compulsory fluctuation value not based on the algorithm, and can change the state value or the estimated value to an arbitrary value at an arbitrary time after the initial stage.

The final state value 89 and the final estimated value 90 of the management simulation are obtained by
This is the last updated value of the state value and the estimated value at the simulation end date and time, and expresses the status of the target management system at the simulation end date and time.

The model adapting means 81 calculates the final state value 89 of the management simulation to be compared between the state value 91 as the adaptation target input from the input means and the adaptation target.
Are changed to all or any combination of the initial state value 85, the initial estimated value 86, the intermediate forced fluctuation state value 87, and the intermediate forced fluctuation value estimated value 88 of the management simulation to be adapted so as to match. To go. (Implementation of adaptation) Search for the value of the adaptation target that enables the conformance target and the comparison target to be the same or extremely close by repeating the process of changing the adaptation target and executing the management simulation. I do. For the search and adaptation of the value to be adapted, a method such as a well-known genetic algorithm (GA) is used. The adaptation is similarly performed on the estimated value 92 as the adaptation target input from the input means.
Either one of the adaptation targets, or the adaptation may be performed by simultaneously inputting the state value 91 and the estimated value 92. Further, the status of the execution of the simulation by the management simulation means 82 and the execution of the adaptation by the model adaptation means 81 are displayed or printed by the output device 3 to the user.

Next, an example of a management decision making procedure using the recommended information generating means will be described with reference to a flowchart of FIG. Here, recommendation information is generated using real data, various data related to the management of the target store is derived, and the data is output as data used when making a management decision.

First, in step S 51, the process of adapting the estimated value of the simulation, which is an incomplete value in performing the management simulation, by the model adapting means 81 is started.

In this adaptation processing, first, in step S5
In step 2, the estimated value is converted to the model adapting means 8 according to the processing conditions.
Adapted by 1. The processing conditions include a simulation period, an initial value state value, an initial value estimated value, an intermediate forced fluctuation, a matching target, a comparison target, and a matching target. Here, the intermediate forced fluctuation is not included.

Then, a certain point in the past of the real management system is set as the starting point of the simulation, and from there to the current time is set as the simulation period of the management simulation, and the state value of the real management system at the certain point in the past is used as the simulation. The initial value is the state value. The initial estimated value is a value set in advance. The relevance target is the current state value of the real management. The comparison target is the final state value of the simulation, and the adaptation target changes the initial estimated value of the simulation while comparing with the final state value.

In the next step S53, the model adapting means 81 obtains an initial estimated value that can be generated in a simulation from the past state values based on the current state values, thereby obtaining the initial estimated values in the simulation from the past to the present. Obtain a final estimate and use this as the current estimate of the real business. This estimated value is used as an adapted estimated value, and is used as an initial estimated value in subsequent simulations for the target management system.

Each of the data collection devices 72 to 74 of the target management system 71 shown in FIG.
If the data that becomes the state value is stored in the real state value data table 84 at an interval of once a day, it is possible to automatically perform the adaptation processing, whereby the adapted estimated value is calculated every day. It will be updated automatically.

In the next step S54, condition settings for performing the simulation from the present to the future are set according to the simulation conditions. The simulation conditions include a simulation period, an initial value state value, an initial value estimated value, an intermediate forced fluctuation, a matching target, a comparison target, and a matching target. Here, the simulation period of the management simulation is from the present to a certain point in the future inputted.

In the next step S55, the user selects a time extrapolation simulation or a target reaching simulation according to the purpose. When the time extrapolation simulation is selected, the process of the time extrapolation simulation is started in step S56. That is, initialization is performed, and a change in the state of the management system based on the passage of time thereafter is expressed on a computer.

In step S57, a management simulation is performed in accordance with the condition setting in step S54. In the condition setting, if the initial state value is the current state of the system and there is no intermediate forced fluctuation, a situation in which the current target management system elapses as it is is simulated. For example, in a retail store, a simulation can be performed in a case where there is no large change from one day to one to three weeks. Therefore, in a retail store, for example, it is possible to predict the number of customers entering the store, predict the sales of individual products, and so on, and prevent the occurrence of expired products due to out-of-stocks or excessive ordering, thereby enabling proper product ordering.

In the condition setting, when the initial state value is an input value obtained by changing the current system state and there is no intermediate forced fluctuation, if a new idea is applied to the current target management system, Simulate the situation of For example, in the case of a retail store, if a sales promotion activity such as a special sales day is planned at a non-regular time, the effect of the plan is based on specific management indicators such as the size of the sales amount and the number of employees required to sell the product. And can be used to evaluate ideas.

Further, when there is an intermediate forced fluctuation in the condition setting, a relatively long-term simulation of one month or more is targeted. For example, when the production of vegetables and fruits changes significantly after a few months due to the unpredictable warm winter or cool summer, and there is a large price fluctuation, the simulation of the target management system is performed before the warm winter or cool summer actually occurs. Or, before such prediction is made, the situation can be grasped as crisis management.

In step S58, the specific situation of the simulation is output to the output device 3, for example, a display, and the display is confirmed by the user. Step S59
Then, the user himself / herself determines whether or not the displayed situation is satisfactory. If the state value can be satisfied, the process proceeds to step S60, and if not, the process proceeds to step S61. For example, in the above condition setting,
If the initial state is the current system state and there is no intermediate forced fluctuation, and if the sales situation of each product at a certain point in the future is satisfactory, the process proceeds to step S60, and if not, the process proceeds to step S61. In this step, an initial state for changing the current state of the target management system is set, and the process returns to step S54 to set the simulation conditions, and then performs the time extrapolation simulation again.

If a satisfactory state is obtained by a number of changes, a change in the current state of the target management system is displayed on the output device 3 in step S60, and the state is changed for the user. Is recommended. At this time, the simulation computer 1 directly transfers data such as a product order and a change in sales price to these devices via the network 77 and the point-of-sale data collection device 72 or the product order data collection device 73, and It is also possible to change the selling price.

When the target simulation is selected in step S55, the process of the target simulation is started in step S62. That is, in the time extrapolation simulation, step S59,
As shown in S61, the user himself had to set an initial state that would change the current state of the target management system. Search for an initial state candidate that changes the state of.

First, in step S63, adaptation is performed by the model adaptation means 81 in accordance with the processing conditions set in step S54. The adaptation target at this time is a value input by the input device 2 or a future state value of the target management system as a user's target preset by selection. The target to be compared with this matching target is the final state value of the management simulation indicating the future state value of the target management system in the simulation.

Then, the initial state value of the management simulation, which indicates the current state value of the target management system in the simulation and is to be matched, is changed so that the matching target and the comparison target match. For example, when a total value of sales of individual commodities such as a target amount of total sales is set as the tailoring target, a plurality of initial state values for matching the target with the comparison target are derived.

Then, in step S64, the model adaptation means 81 searches for one or more initial state values that will result in the future state value of the target management system, and outputs it to the output device 3, for example, a display. And display it to the user. In step S65, the user himself / herself determines whether or not the displayed situation is satisfactory. For example, the initial state value is a change from the current management system, but if it is difficult to change, the conformance target may not be possible. Otherwise, if too many initial state values are displayed, a limit is added to the initial conditions to be searched, and the process returns to step S54 to set the simulation conditions and then perform the target attainment simulation.

If a satisfactory state is obtained by several changes, the process moves to step S66, and in this step,
Changes to the current state of the target management system are displayed on the output device 3, and the user is encouraged to change the state. This concludes the description of an example of the procedure of the management decision support using the recommended information generating means. However, the procedure of the management decision support using the recommended information generating means is based on the estimated value which is a basic configuration. Fitting processing, time extrapolation simulation,
It is only necessary to include a goal attainment simulation.
However, the present invention is not limited to this procedure.

With this management decision-making support system, the user himself / herself can perform specific operations such as store opening plan, management strategy, floor layout, assortment, shelf layout, sales promotion, price setting, inventory setting, and product ordering. Verification of ideas can be evaluated using concrete management indicators as a scale. In addition, we can present multiple specific recommended information for one management goal,
The range of choices of the user can be freely changed. It can also be used for strategic issues such as long-term forecasting and crisis management. In this way, it is possible to supply the user with recommended information that supports effective decision-making during comprehensive planning and comprehensive evaluation.

Next, a specific embodiment will be described in which such a management decision support procedure is applied to an actual retail business. a. Estimated value adaptation processing First, the estimated value adaptation processing shown in steps S51 to S53 of FIG. 24 is performed.

For example, it is assumed that various state values of the target store from one month ago to the present time are stored in the actual state value data table 84 in FIG. Here, among the various status values, as an example, the area element allocation status table 35, the passive element allocation status table 36, the action element allocation status table 37 shown in FIG. 7, the sales unit status table 7c shown in FIG. Indicating customer agent status table 7e
Take up.

[0177] Therefore, at any point in time from one month ago to the present time, the status of the in-store layout is the area element arrangement state table 35 and the passive element arrangement state table 36.
And a target area 11 corresponding to the in-store layout shown in FIG. 4 can be formed. Similarly,
The state of the sales floor configuration at any time from one month ago to the present time is expressed using the passive element arrangement state table 36, and the target area 21 corresponding to the sales floor configuration as shown in FIG. 5 can be formed. Further, the status of the product / service giving / receiving action and the occurrence of business action in the store at any time from one month ago to the present time are described in the action element arrangement state table 3.
7, the target area 31 as shown in FIG.
Is formed. The sales method status (including the sales promotion status) of each product from one month ago to the present time is expressed in the sales unit status table 7c, and the customer (purchase customer) product purchase status is , The customer agent state table 7e.

As described above, the state value at an arbitrary point in time from one month ago to the present time is stored in the actual state value data table 84 and expresses the situation in the store at an arbitrary point in time from one month ago to the present time. it can. However, FIG.
The estimated values in the sales agent response coefficient setting table 7d of the customer agent shown in FIG. 7 and the purchase behavior response coefficient setting table 7f of the customer agent shown in FIG. 16 are not accurately determined. Therefore, the purpose of the estimate adaptation process is to accurately adapt the estimate to the current state value.

This adaptation method will be described with reference to FIG. For example, the state value one month ago is set as the initial state value 85 of the management simulation, and the estimated value setting means 83 is used to set a temporary estimated value one month ago from the state value, and this is used as the initial estimation of the management simulation. Value 86. Thereby, the management simulation means 82 is operated,
A management simulation from one month ago to the present time is performed to calculate an estimated value 92 which is a matching target. This is compared with a state value 91 which is a matching target provided from the actual data collecting means in the model matching means 81. The model adapting means 81 compares the final state value 89 of the management simulation with respect to the present time with the actual state value at the present time, and calculates the temporary estimated value one month ago so that the two agree with the set accuracy. Make corrections and run the management simulation again.

By repeating this step, the final state value 89 of the management simulation with respect to the present time and the actual present state value can be matched with the set accuracy, and an estimated value one month ago can be obtained. At the same time, it is possible to obtain a final estimate 90 of the current management simulation that exactly matches the actual current state value. Thereafter, a business simulation from the current time to an arbitrary future time set as the initial estimated value obtained by adapting the final estimated value 90 of the current business simulation to the current state value is performed. Then, based on the management simulation from the present time to an arbitrary time in the future, some recommendation information for the current management decision is supplied to the user as management decision support information.

The recommended information generating method includes, for example, two methods, a time extrapolation simulation and a target arrival simulation, and the conditions and selection are performed in steps S54 and S55 in FIG. b. Time Extrapolation Simulation An example in which the time extrapolation simulation shown in steps S56 to S61 in FIG. 24 is applied to an actual store will be described.

In this simulation, first,
The current state values are left as they are, and, for example, the current value is adopted for the area element arrangement state table 35 and the passive element arrangement state table 36 to make the current sales floor configuration, and further, for the sales unit state table 7c. Also adopts the current value as the current product sales method. This and
The management simulation means 82 is operated using the initial estimated value adapted to the current state value, and a management simulation is performed from the current time, for example, one week later.

As a result of the management simulation,
Various state values after one week are calculated. For example, the customer agent state table 7e is calculated, the total number of purchased products and the total price of purchased products for each customer agent are calculated, and this is displayed and confirmed to the user via the output device 3 as recommended information. Become.

If the user is not satisfied with the recommended information, for example, a change in the price of the product or a change in the discount rate is planned, and the sales unit status table 7c is set to correspond to this. Change price strength and discount rate,
This is set as the initial state value 85 of the new management simulation, and the management simulation from the present time to one week later is performed again.

If the user's satisfaction is still not satisfied, the store layout and the sales floor configuration are further planned to be changed, and the area element arrangement state table 35 and the passive element arrangement state table 36 are changed to correspond to this. , And this as the initial state value 85 of the new management simulation,
Further, a management simulation is performed again from the present time to one week later.

In this way, the above steps are repeated until the user is satisfied, and various state values one week later are finally calculated. The sales unit state table 7c, the area element arrangement state table 35, the passive element arrangement state table 36, and the like at that time become recommended information for the manager of the target store as the user, that is, management decision support information.

C. An example of a case where the target arrival simulation shown in steps S62 to S67 of FIG. 24 is applied to an actual store will be described. In this simulation, the steps of creating a new business plan such as a change in the price of a product in the time extrapolation simulation and repeating the management simulation are performed relatively easily using the model adaptation means 81 described above. .

First, a state value one week after the target management system, which is a target of the user, is input via the input device 2 as a matching target. Next, while keeping the current state value as it is, for example, the current value is adopted for the area element arrangement state table 35 and the passive element arrangement state table 36, and the current value is also applied to the sales unit state table 7c. The value is adopted as the current product sales method. The management simulation means 82 is operated by using this and the initial estimated value adapted to the current state value, and the management simulation is performed from the present time to one week later.

As a result of the management simulation,
One week later, the customer agent status table 7e is calculated, and the total number of purchased products and the total price of purchased products are calculated. The model matching means 81 compares this with a state value 91 which is a matching target indicating a state value one week after the user is satisfied. As described above, the model adapting means 81 uses both (the result of the simulation and the set value of the user).
For example, the area element arrangement state table 35 and the passive element arrangement state table 3
6. Correct the current status values in the sales unit status table 7c and the like, and perform the management simulation again.

By repeating this process, it is possible to obtain a current state value at which the final state value 89 of the management simulation for one week later matches the actual state value that the user is satisfied with the set accuracy. . Finally,
For example, the total number of purchased products one week after the user is satisfied and the current product sales method or sales floor configuration that satisfies the total purchased product price are calculated, and the corresponding information is the sales unit status table 7c and the area element arrangement status table. 35, the passive element arrangement state table 36 and the like. This is the recommended information for the manager of the target store that is the user, that is, the management decision-making support information.

In this goal attainment simulation, if a state value 91 which is an adaptation target far from the current state is set, recommended information which cannot be executed in the actual retail business, such as an extreme discount rate, is calculated. In this case, in this case, it is necessary to change the state value one week after the user who is satisfied with the state value 91 which is the adaptation target is satisfied, and perform the goal attainment simulation again.

Next, a case will be described in which a service giving / receiving behavior in a store is simulated by using a search behavior algorithm of the in-store simulation shown in FIG. 10 by using a simulation method of a product / service giving / receiving action and a business action generating system. State.

For example, in the case of simulating the packaging service of a product at the packaging service counter in a retail store, it is possible to simulate the packaging service receiving behavior of the customer agent by using the algorithm of the search behavior. In this case, first, a packaging service counter unit is set as a passive element, and is arranged in the target area 11 corresponding to the in-store layout shown in FIG. Therefore, the passive element arrangement state table 36 of the element arrangement state table 7b
In, the element code of the packaging service counter unit is set corresponding to the element code of the area element where the packaging service counter unit is placed.

Further, the target element status table 7c shown in FIG.
Then, the stimulus value of the packaging service counter unit is set. In response to this, a response coefficient corresponding to the stimulus value of the packaging service counter unit of the customer agent is set in the action element-to-target element response coefficient setting table 7d. In addition, a reaction coefficient related to the packaging service receiving behavior of the customer agent is set in the response factor setting table 7f for the behavior element.

The algorithm for simulating the packaging service enjoying behavior of the customer agent is as follows:
The algorithm of the search behavior of the in-store simulation shown in FIG. 10 can be used. In this case, the degree of attraction of the packaging service counter unit is calculated based on the above setting,
If the inviting element is a packaging service counter unit,
If the packaging service receiving action is set as the corresponding target action, the customer agent moves closer to the packaging service counter unit as in the case of the first embodiment, and performs the packaging service receiving action. Can be implemented.

The employee agent behavior processing corresponding to this can be achieved by adding the product packaging operation behavior of the employee agent to the employee agent behavior processing algorithm shown in FIG. Naturally, other in-store services such as a food and drink providing service and a food preparation service can be similarly simulated.

Next, a description will be given of a case of simulating a service giving / receiving action in the service industry using a method of simulating a product / service giving / receiving action and a business action generating system. For example, when simulating a game providing service in a game center, the first
The in-store simulation algorithm described in the embodiment can be used to simulate the game providing service receiving behavior of the customer agent. The simulation method in this case is basically the same as the in-store simulation. By expressing the game machine as a sales unit, the same target area setting and simulation algorithm as in the above-described in-store simulation can be applied.

That is, the customer agent searches the target area in the game center corresponding to FIG. 6 for the game machine to which the customer agent is most attracted, and provides the service (game play itself) provided by the game machine. And simulate the action of paying the price before or after the service is provided. In addition, game machine maintenance work, charge collection work, and the like of the employee agent corresponding to this can be added to the algorithm of the behavior processing of the employee agent. Similarly, other services such as a video rental business, or a product / service transfer behavior in a store such as a restaurant business can be similarly simulated.

Further, with regard to the expansion of the adaptation range of the commercial area simulation, products and services such as retail stores, restaurants, banks, hospitals, schools, offices, factories, and stations, which are shown as store units 67 in FIG. Since various types of business types to be exchanged can be expressed, it is possible to simulate the selection behavior of a customer agent for a store or a building of such various types of business.

In each of the above-described embodiments, the management simulation method, that is, the simulation method of the product / service giving / receiving action and business action generation system, and the management decision support system using the simulation method are used for the retail business. Although the case where the present invention is applied to in-store / trade area simulation has been described, the present invention is not necessarily limited to this, and can be widely applied to the transfer of goods / services and the business generation system related thereto. For example, since the situation of a store can be expressed, it can be easily applied to other businesses having a store, such as a restaurant business and a service business. In addition, it can be applied to stores of various sizes from convenience stores to supermarkets and department stores.

Further, the present invention can also be applied to an area where a plurality of business types are combined, such as a shopping center, a shopping district, and an underground shopping mall. The simulation and the simulation of each store having a different business type can be easily combined. In addition, since it is possible to simulate the relationship between services and their users in a specific area, layouts for amusement centers such as game centers, large-scale leisure centers such as amusement parks, exhibition halls, hospitals, libraries, government offices, stations, etc. And so on.

Furthermore, users of the management decision support system are not only managers of the target stores but also wholesalers who distribute products to several stores in a certain area.
Alternatively, in a maker that manufactures products, or a financial institution such as a bank, it is also possible to accurately determine the location of the store itself and the configuration within the store, and to apply for financing when opening or renovating a store.

[0203]

According to the first to seventh aspects of the present invention, it is possible to faithfully simulate a product / service giving / receiving action and a business action generating system with a real person's action and a sale state of a product by a relatively simple algorithm. . Therefore, the present invention is effective in giving and receiving of goods and services and in grasping the current state of work related thereto, improving services, and improving work efficiency. Further, since the simulation can be configured by the same algorithm from the shelf for each stage to the inside of the store, and even to a wide range of trade areas, it is possible to cope with a store having a greatly different scale and a plurality of business types and a combination thereof.

According to the first aspect of the present invention, a plurality of target elements are set in the target space divided into cells, and the action of the target element is expressed by the interaction of the target elements. Simulations that can faithfully represent the behavior of customers and employees in stores and commercial areas.

According to the second aspect of the present invention, further, an interaction based on the data of the target element is performed, the data of the target element is updated based on the action result of the interaction, and the data is updated again based on the data of the target element. By repeating the interaction, the action result at the previous time is reflected in the action at a later time, such as the action of a customer or an employee in a real store or a commercial area. Accordingly, it is possible to realize a simulation that faithfully expresses the elapse of real time, which is difficult in a simulation that performs statistical data processing.

According to the third aspect of the present invention, further, 1
The data is updated immediately after one action element has taken action, so the time at which each action takes place autonomously is determined, such as the behavior of customers or employees in a real store or commercial area. Can express the situation. As a result, a simulation that faithfully represents reality can be easily realized.

According to the fourth aspect of the present invention, the scanning range of the target element further includes a cell located at a position farther from the cell adjacent to the cell at the present position of the target element. It scans over a wide range, such as the behavior of customers and employees in a commercial area, and can recognize the state of target elements within that range. As a result, a simulation that faithfully represents reality can be easily realized.

According to the fifth aspect of the present invention, the target elements are further classified into behavioral elements, passive elements, and area elements. The situation and the situation of the trade area can be easily set as data. In addition, since an algorithm that processes only the action element among the target elements is employed, a simulation that faithfully represents reality can be easily realized.

[0209] According to the invention of claim 6, since the space divided into sub-cells is associated with each cell constituting the target space, the product in the product display shelf is simulated in the store. Simulation including placement can be realized. In addition, a simulation including a layout in a store can be realized in a commercial area simulation.

According to the seventh aspect of the present invention, the space divided into sub-cells is represented as an array of two-dimensional cells, and one of the dimensions represents the dimension in the height direction. It is possible to easily realize a three-dimensional simulation including a shelf on which goods are arranged in the height direction as described above.

According to the eighth and ninth aspects of the invention, the actual management situation can be accurately grasped by collecting the data from the actual stores and adapting the estimated value that compensates for the lack of the data collection to the reality. . In addition, since users are verifying their own ideas and searching for multiple candidates for achieving their goals, they are planning to open stores, manage strategies, floor layouts, and assortment of products in a wide range of businesses, such as retail, restaurants, and services. , Shelf layout,
With regard to specific tasks such as sales promotion, price setting, stock quantity setting, and product ordering, it is possible to provide users with effective recommended information that supports decision-making during comprehensive planning and comprehensive evaluation and review.

According to the invention of claim 8, furthermore, accurate simulation data can be set by the actual data collecting means for collecting data from the actual product / service giving / receiving action and business action generation system. For incorrect data, simulation data that is well suited to the real system can be set by the model matching means.

According to the ninth aspect of the present invention, the management simulation from the present time to the future time can be accurately performed by the time extrapolation simulation means. It is possible to supply management decision support information that allows the user to easily grasp the state and the effects of various plans on the management of the user.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a first embodiment of the present invention.

FIG. 2 is a diagram showing a memory configuration of a main memory according to the embodiment;

FIG. 3 is a view showing an example of cell division of a target space in the embodiment.

FIG. 4 is an exemplary view showing an example of a target area corresponding to an in-store layout according to the embodiment;

FIG. 5 is a view showing an example of a target area corresponding to a sales floor configuration according to the embodiment;

FIG. 6 is an exemplary view showing an example of a target area corresponding to an in-store situation in the embodiment.

FIG. 7 is a diagram showing a configuration of an element arrangement state table in the embodiment.

FIG. 8 is a flowchart showing an algorithm of the entire in-store simulation according to the embodiment.

FIG. 9 is a flowchart showing an algorithm of a behavior process of a customer agent in FIG. 8;

FIG. 10 is a flowchart showing an algorithm of a search action in FIG. 9;

FIG. 11 is a schematic diagram for explaining spatial scanning and moving target determination in the embodiment.

FIG. 12 is a schematic diagram for explaining a movement behavior and a purchase behavior in the embodiment.

FIG. 13 is a diagram showing a configuration of a target element state table used for calculating the degree of attraction in the embodiment.

FIG. 14 is a diagram showing a configuration of a behavior element-to-target element reaction coefficient setting table used for calculating the degree of attraction in the embodiment.

FIG. 15 is a diagram showing a configuration of an action element state table used for calculating the degree of attraction in the embodiment.

FIG. 16 is a diagram showing a configuration of a target behavior reaction coefficient setting table of a behavior element used for calculating the degree of attraction in the embodiment.

FIG. 17 is a flowchart showing an algorithm of the accounting action in FIG. 9;

FIG. 18 is a flowchart showing an algorithm of behavior processing of the employee agent in FIG. 8;

FIG. 19 is a diagram for explaining a target shelf space and a target shelf area according to the second embodiment of the present invention.

FIG. 20 is an exemplary view for explaining a scanning range in a target shelf area according to the embodiment;

FIG. 21 is a diagram illustrating a target space and a target region corresponding to a commercial area space according to the third embodiment of the present invention.

FIG. 22 is a block diagram showing a configuration of a management decision support system according to a fourth embodiment of the present invention.

FIG. 23 is a block diagram showing a functional configuration of a recommendation information generating unit according to the embodiment.

FIG. 24 is a flowchart showing a procedure of management decision-making support by the recommended information generating means in the embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Simulation computer 2 ... Input device 3 ... Output device 4 ... CPU (Central processing unit) 5 ... Main memory 6 ... Storage device

Claims (9)

[Claims] 1. A target space composed of an array of a plurality of cells is set for a product / service giving / receiving action and business action generating system to be simulated, and goods / services including movements are set in the target space. By setting a plurality of target elements that perform an interaction related to the occurrence of an action and a business action, and arranging a predetermined target element on a cell in the target space, data of a product / service transfer action and a business action generation system is set as a target area. In the target area, the predetermined target element interacts with the recognizable target element within the set scanning range, and performs the goods / service giving / receiving operation and business operation including movement in accordance with the set action rule. The data expressing the target area is changed in accordance with the action result, and this is repeated for a plurality of target elements, so that goods and services can be exchanged. The time-series state change of the dynamic and operational behavior generation system, management simulation wherein simulating the changes in the data of the target area. 2. A method according to claim 1, wherein the predetermined target element is in a target area,
Interacts with the recognizable target element within the set scanning range based on the data held by each other, and performs goods / service transfer and business actions including movement according to the set action rules, and according to the result of the action. 2. The management simulation method according to claim 1, wherein the data held by each other is updated, and thereby the data expressing the target area is changed. 3. After one of the predetermined target elements has performed a product / service giving / receiving action and a business action including movement, and before the processing of the action of the next target element is started, the action of the one target element is started. 3. The management simulation method according to claim 1, wherein the data representing the target area is immediately changed according to the result. 4. The management simulation method according to claim 1, wherein the target element includes a cell located at a position farther from a cell adjacent to the cell in which the target element is located in the scanning range. 5. The target element is a behavior element that performs a product / service giving / receiving operation and a business operation including movement, a passive element that operates passively with respect to the behavior element, and movement of the behavior element and arrangement of the passive element. The area element is first placed on the cell of the target space, and the passive element is placed on this area element to define the target area corresponding to the goods / service transfer behavior and business action generation system. In the configured target area, only the behavior element interacts with the recognizable target element within the set scanning range based on the data held by each other, and includes a product including movement according to the set behavior rule. Perform service transfer and business activities,
Based on the result of the action element, the data held by the action element and the passive element are updated, and the time series state change of the product / service transfer action and business action generation system is simulated by simply repeating the processing of the action element action. 5. The management simulation method according to claim 1, wherein the management simulation method is performed. 6. The management simulation method according to claim 1, wherein a newly divided space is made to correspond to each cell constituting the target space as a sub-cell. 7. The object space is expressed as a two-dimensional cell array, and the subcell space is expressed as a two-dimensional cell array. One of the subcell space dimensions is changed to a two-dimensional cell array in the object space. 7. The management simulation method according to claim 6, wherein a dimension in a height direction is expressed. 8. A product / service giving / receiving action and a business action generating system are represented by data as a target area.
A simulation means for simulating a time-series state change of a service giving / receiving action and a business action generating system by a change in data of a target area, and a real product / service giving / receiving action and a business action generating system at the start of the simulation by the simulation means. Means for securing an initial state value and an initial estimated value; securing a final state value at which the initial state value has finally changed and a final estimated value at which the initial estimated value has finally changed at the end of the simulation by the simulation means. Means for collecting, as real state values, values corresponding to the state values including the initial state values and the final state values from the actual product / service giving / receiving action and business action generation system; The final state value or final estimated value to be compared with the The initial state value or the initial estimated value to be adapted is changed as described above, and the process of performing the change of the adaptive object and performing the simulation by the simulation means is repeated to obtain the final state value or the final estimated value to be compared with the adaptive target. A management decision-making support system comprising: a model adaptation unit that searches for an adaptation target value to be matched with a value. 9. A product / service giving / receiving action and a business action generating system are represented by data as a target area.
A simulation means for simulating a time-series state change of a service giving / receiving action and a business action generating system by a change in data of a target area, and a real product / service giving / receiving action and a business action generating system at the start of the simulation by the simulation means. Means for securing an initial state value and an initial estimated value; securing a final state value at which the initial state value has finally changed and a final estimated value at which the initial estimated value has finally changed at the end of the simulation by the simulation means. Means for collecting, as real state values, values corresponding to the state values including the initial state values and the final state values from the actual product / service giving / receiving action and business action generation system; The final state value or final estimated value to be compared with the The initial state value or the initial estimated value to be adapted is changed as described above, and the process of performing the change of the adaptive object and performing the simulation by the simulation means is repeated to obtain the final state value or the final estimated value to be compared with the adaptive target. A model adaptation means for searching for a value to be adapted to be matched with a value, an estimated value adaptation means for obtaining a fitted estimated value which is an estimated value adapted to the present; Time extrapolation simulation means for simulating a product / service giving / receiving action and business action generation system from the present to the future set by the simulation means using the obtained fitted estimated value, and Achievement simulation means for finding the current initial state value that has been adapted is provided. The current state value of the service giving / receiving action and business action generating system, the comparison object is the final state value, the adaptation object is the initial estimated value, and the model adaptation means is used by the model adaptation means from the past set time to the present. The target value is searched for, and the estimated value adapting means obtains an adapted estimated value which is an estimated value adapted to the present value. The target is set as an initial state value, and from the present to the future, a search for a value of an adaptation target is performed by the model adapting means from the present time to an initial value at the present time which matches the adaptation target set by the goal attainment simulation means. A management decision support system characterized by obtaining a state value.