Information processing device and information processing method

Abstract

To highly accurately analyze an automobile market.SOLUTION: A database including a first data describing a plurality of virtual users and second data describing a plurality of automobile models which is sold in a market is stored in a storage device. Also, a control device acquires sales promotion measures for the market, predicts purchase behaviors of the plurality of virtual users after executing the sales promotion measures on the basis of the first data, the second data and the sales promotion means, and reflects the purchase behaviors predicted on the plurality of virtual users on the first data.SELECTED DRAWING: Figure 7

Description

[Technical Field]

[0001] The present disclosure relates to a technology for simulating a market. [Background Art]

[0002] There exists a technology for simulating market needs based on user preferences. In relation to this, Patent Document 1 discloses a system that performs market analysis based on individual values acquired through questionnaires. [Prior Art Literature] [Patent Literature]

[0003] [Patent Document 1] Japanese Unexamined Patent Publication No. 2009-238182 [Summary of the Invention] [Problem to be Solved by the Invention]

[0004] Along with the development of machine learning, it is thought that opportunities for utilizing simulation data will continue to increase in the future.

[0005] An object of the present disclosure is to accurately analyze an automobile market. [Means for Solving the Problem]

[0006] A first aspect of the present disclosure is an information processing apparatus comprising: a storage device that stores a database including first data describing a plurality of virtual users and second data describing a plurality of automobile models sold in a market; and a control device that executes: acquiring a sales promotion measure for the market; predicting purchasing behaviors of the plurality of virtual users after the sales promotion measure is implemented based on the first data, the second data, and the sales promotion measure; and reflecting the purchasing behaviors predicted for the plurality of virtual users in the first data.

[0007] Furthermore, a second aspect of this disclosure is an information processing method that includes the steps of: acquiring first data describing a plurality of virtual users and second data describing a plurality of automobile models sold in the market; acquiring sales promotion measures for the market; predicting the purchasing behavior of the plurality of virtual users after the implementation of the sales promotion measures based on the first data, the second data, and the sales promotion measures; and reflecting the predicted purchasing behavior for the plurality of virtual users in the first data.

[0008] Another aspect of this disclosure is a computer-readable storage medium that non-temporarily stores a program for performing the above method. [Effects of the Invention]

[0009] This disclosure enables accurate analysis of the automotive market. [Brief explanation of the drawing]

[0010] [Figure 1] A diagram illustrating the processing overview of the analysis system related to this disclosure. [Figure 2] A diagram showing the components of the server device 100 in detail. [Figure 3] A diagram illustrating virtual user data stored in a virtual market database. [Figure 4] A diagram illustrating preferences associated with virtual users. [Figure 5] A diagram illustrating vehicle data stored in a virtual market database. [Figure 6] A diagram illustrating how to generate virtual user data. [Figure 7] A diagram illustrating the data transmitted and received between modules. [Figure 8] A diagram illustrating the overview of the market analysis performed by Analysis Department 1011. [Figure 9] An example of the output results from a market overview analysis. [Figure 10] An example of output results from a competitor vehicle analysis. [Figure 11] An example of an output result in key indicator analysis. [Figure 12] An example of an output result in user analysis. [Figure 13] An example of an output result in potential analysis. [Figure 14] An example of a screen for inputting a measure. [Figure 15] An example of a screen for inputting a measure. [Figure 16] An example of scores calculated for each virtual user. [Figure 17] An example of a result obtained by simulating market reaction after execution of a measure. [Figure 18] An example of a result obtained by simulating market reaction after execution of a measure. [Figure 19] A flowchart of processing executed by the prediction unit 1012. [Figure 20] A diagram illustrating in detail the components of the server device 100 according to the second embodiment. [Figure 21] A diagram explaining the status of virtual users that changes over time. MODE FOR CARRYING OUT THE INVENTION

[0011] Technologies for performing market analysis and future demand forecasting using personal data such as individual preferences are known. For example, when the target is automobiles, matching a plurality of parameters representing individual preferences with a plurality of parameters defined for each vehicle type makes it possible to predict to what extent a new automobile to be introduced to the market will be accepted by users.

[0012] However, while such analysis can predict the immediate user reaction to the introduction of a new model to the market, it cannot predict the future impact on the market. For example, if the introduction of a new model causes a large number of users to switch from existing models, the market background may change. In such cases, it becomes difficult to formulate strategies such as "what model should be introduced next to increase profits." This is because existing demand forecasts are only made for a specific point in time.

[0013] To solve this problem, it is necessary not only to predict user purchasing behavior, but also to predict "how the market will change as a result of user purchasing behavior," and to update the underlying market background itself. The information processing device relating to this disclosure solves the aforementioned problems.

[0014] An information processing device according to one aspect of the present disclosure is characterized by comprising: a storage device that stores a database including first data describing a plurality of virtual users and second data describing a plurality of automobile models sold in the market; and a control device that performs the following: acquiring sales promotion measures for the market; predicting the purchasing behavior of the plurality of virtual users after the implementation of the sales promotion measures based on the first data, the second data, and the sales promotion measures; and reflecting the predicted purchasing behavior for the plurality of virtual users in the first data.

[0015] The first data is a collection of data describing multiple virtual users. The first data may include parameters such as age, gender, economic status, and life stage, as well as multiple parameters related to automobile preferences. These parameters can be anything that influences automobile purchasing behavior. Furthermore, the first data The data may include detailed information about the vehicles currently owned and vehicles previously owned. The second set of data is a collection of data describing information about automobiles sold in the market. This second set of data may include multiple parameters, such as vehicle type, class, features, and price.

[0016] The control device obtains sales promotion measures related to automobile sales from the device operator and uses the first and second data to predict the user's purchasing behavior. Sales promotion measures related to automobile sales typically involve changes to the sales lineup, such as the introduction of new models, the discontinuation of older models, and model changes. For example, by reflecting sales promotion measures in a second set of data and matching it with the first set of data, it is possible to predict the purchasing behavior of a hypothetical user.

[0017] However, as mentioned earlier, this alone can only predict user purchasing behavior on an ad-hoc basis. Therefore, the information processing device relating to this disclosure reflects the prediction results in the first data. That is, it assumes that a virtual user has taken purchasing action and updates the data that forms the basis of the prediction. With this configuration, it becomes possible to perform demand forecasting for the next generation using the updated first data. In addition, if the automobile sales lineup changes due to the introduction of sales promotion measures, the second data may be updated along with the first data.

[0018] Furthermore, the control device may predict market changes after multiple virtual users have taken the aforementioned purchasing action and output the results. This makes it possible to provide information such as "how much market share between companies will change as a result of the introduction of sales promotion measures." With such a configuration, it is possible to provide information for deciding whether or not to actually introduce sales promotion measures into the market.

[0019] Furthermore, the control device may perform further market analysis based on the first data reflecting purchasing behavior and output information for determining new sales promotion measures. This configuration makes it possible to analyze the market after sales promotion measures have been implemented. In other words, it can provide information for determining sales promotion measures for subsequent generations.

[0020] Furthermore, the control device may simulate the passage of time and reflect the results of that simulation in the first data. For example, as time passes after purchasing a car, the virtual user's life stage and economic situation may change.

[0021] The following describes specific embodiments of this disclosure with reference to the drawings. Unless otherwise specified, the hardware configurations, module configurations, functional configurations, etc., described in each embodiment are not intended to limit the technical scope of the disclosure to those configurations alone.

[0022] (First embodiment) An overview of the analysis system according to the first embodiment will be described. Figure 1 is a schematic diagram of the server device 100 in this embodiment. The server device according to this embodiment has a database (hereinafter referred to as the virtual market database) that stores information on multiple virtual users and automobile models sold in the market, and uses this database to predict users' purchasing behavior and analyze the market.

[0023] In this embodiment, the server device 100 stores a virtual market database representing the automobile market. The virtual market database is a database of consumers (hereinafter referred to as users) who are potential buyers of automobiles. This is a database that stores data about automobiles, which are the products. In the virtual market database, multiple virtual users are defined based on real consumers. Hereafter, these multiple virtual users will be referred to as virtual users, and the data that represents these virtual users will be referred to as virtual user data. For example, virtual users can be data adjusted so that their age, life stage, preferences, etc., are in proportion to those of real consumers. In this embodiment, the person who operates the server device 100 to perform market analysis is referred to as the administrator, and is distinguished from the user, who is a consumer.

[0024] Furthermore, the virtual market database stores data on multiple automobiles sold in the market. Such data will hereafter be referred to as vehicle data. Vehicle data may define models of automobiles currently on sale, or it may define models of virtual automobiles. For example, if you want to predict the market reaction to implementing some sales promotion measure (e.g., a model change for an automobile currently on sale), you can delete the vehicle data for the pre-model change and add the vehicle data for the post-model change. In the following explanation, sales promotion measures that involve changes to the sales lineup, such as the introduction of new models, model changes, and model discontinuation, will simply be referred to as "measures." Also, in the following explanation, the term "model" will be used to refer to a specific type of automobile sold in the market.

[0025] As shown in Figure 1, the server device 100 is configured to perform three main types of functions. The first is a function that performs market analysis based on data stored in the virtual market database and outputs the results. This allows the company to provide information such as what is lacking in its automobile lineup and what kind of measures should be taken to promote sales (for example, what new models should be launched).

[0026] The second function is to simulate the market's reaction when a measure related to automobile sales is implemented (also called launching a measure). For example, using parameters such as preferences assigned to each virtual user, it is possible to predict to what extent a newly introduced measure will be accepted by the virtual users (for example, whether or not the virtual users will purchase the new model of automobile). Furthermore, by performing this process for multiple virtual users, it is possible to simulate how much market share will change.

[0027] The third function is to update the virtual market database, assuming that the measures have actually been implemented in the market. As mentioned earlier, as a result of implementing the measures, some of the virtual users may purchase or switch to a specific model of car. Also, the implementation of the measures will change the lineup of cars available for sale. The server device 100 reflects these market changes in the virtual market database by updating the virtual user data and vehicle data. The server device 100 can perform another market analysis using the updated virtual market database. This makes it possible to simulate over time how the market will change if multiple measures are continuously implemented.

[0028] Next, the server device 100 will be described in detail. Figure 2 is a diagram showing in detail the components of the server device 100 according to this embodiment.

[0029] The server device 100 can be configured as a computer having a processor such as a CPU or GPU, main memory such as RAM or ROM, EPROM, hard disk drive, and auxiliary storage such as removable media. The auxiliary storage includes operating The operating system (OS), various programs, and various tables are stored therein. The programs stored there are loaded into the working area of ​​the main memory and executed. Through the execution of the programs, each component is controlled, thereby enabling the realization of various functions that meet predetermined purposes, as described later. However, some or all of the functions may be implemented by hardware circuits such as ASICs or FPGAs.

[0030] The server device 100 is configured to include a control unit 101, a storage unit 102, and an input / output unit 103. The control unit 101 is a computing device that manages the control performed by the server device 100. The control unit 101 can be implemented by a computing device such as a CPU. The control unit 101 is configured with an analysis unit 1011 and a prediction unit 1012 as functional modules. Each functional module may be implemented by executing a stored program using a CPU.

[0031] The analysis unit 1011 performs market analysis using data stored in a virtual market database and outputs the results. Market analysis involves analyzing the relationship between a specific automobile model and its demanders (users) from a predetermined perspective and outputting the results. Examples of market analysis methods include the following: (1) Market overview analysis This is an analytical method that maps the relationships between arbitrary parameters (for example, the relationship between age group and income) for users who own a particular model of car. (2) Analysis of competing vehicle models This method visualizes other car models that are likely to be compared to a particular model. (3) Important indicator analysis This method analyzes and visualizes the reasons why users choose a particular car model. (4) User analysis This is a statistical method for determining what parameters are present in the demographic that actually purchases a particular car model. (5) Potential analysis This method visualizes what parameters might interest a particular car model in a given group of people. These methods can be based on publicly available information. Specific methods will be described later. Based on the results of market analysis, system administrators can formulate measures that automobile manufacturers should take.

[0032] The forecasting unit 2012 simulates the market response when measures entered by the administrator are introduced to the market. For example, the forecasting unit 2012 performs matching using multiple parameters possessed by virtual users (parameters that influence purchasing behavior towards automobiles; hereinafter referred to as user parameters) and parameters possessed by automobiles to be introduced to the market (hereinafter referred to as vehicle parameters), and calculates a score representing the degree of fit. By performing this for all virtual users, it is possible to predict how many virtual users will take purchasing action (for example, how many virtual users will purchase a newly released model). Furthermore, this allows for simulations such as "what kind of movement (switching) will occur between models" and "how market share between companies will change."

[0033] Furthermore, the prediction unit 1012 reflects the simulation results in the virtual market database based on the administrator's instructions. Specifically, it reflects the status of the virtual user after taking purchasing action (such as the cars they own) in the virtual user data, and the status of cars after the measures are implemented. The lineup is reflected in the vehicle data. This process makes the entered measures appear as if they have been introduced into the virtual market. This allows for continued market analysis in the next step.

[0034] The storage unit 102 comprises a main memory and an auxiliary storage device. The main memory is the memory where programs executed by the control unit 101 and data used by said control programs are stored. The auxiliary storage device is the device where programs executed by the control unit 101 and data used by said control programs are stored.

[0035] Furthermore, the storage unit 102 stores virtual user data 102A and vehicle data 102B. Virtual user data 102A is data describing multiple hypothetical consumers who could potentially purchase a car. Figure 3 shows an example of virtual user data. As illustrated, virtual user data includes information such as gender, age, life stage, annual income, place of residence, currently owned car, and cars previously owned. Virtual user data may also include detailed information (vehicle age, mileage, characteristics, etc.) about cars owned (or previously owned) by the virtual user. Virtual user data can also be described as a user model that represents the car life of a virtual user. Furthermore, the virtual user data includes information about the virtual user's preferences regarding automobiles. In this example, the preference information is defined as a preference table.

[0036] Figure 4 shows an example of a preference table. A user's preferences for automobiles are diverse, including preferences for manufacturers, body types, and vehicle classes. In the preference table, a score (for example, out of 10) is assigned to each of these parameters that influence purchasing behavior. In the illustrated example, the hypothetical user is defined as having the highest preference for cars from company A, SUV-type cars, and medium-sized cars.

[0037] The preference table includes definitions for items that a hypothetical user considers important when choosing a car, such as "product image," "expectations," and "vehicle functions" (symbol 401). In the illustrated example, the hypothetical user is assigned a score out of 10, representing how important each of the multiple items is to them when choosing a car. In this example, the hypothetical user is defined as preferring environmentally friendly cars with low running costs. These parameters that influence purchasing behavior regarding automobiles are called user parameters.

[0038] Vehicle data 202B is data that defines multiple car models sold in a virtual market. Figure 5 shows an example of vehicle data 202B. The vehicle data includes items specific to each vehicle model, such as car name, model name, manufacturer, body type, class, price, and engine displacement. The body type and class are determined from several predetermined items. Vehicle data 202B may also include the date sales began (or ended) and the date sales ended (or ended).

[0039] Furthermore, vehicle data 202B includes items that represent the characteristics of the automobile, such as "product image," "expectations," and "vehicle functions." The product image represents a specific image associated with the model (e.g., "cute," "sporty," "urban," etc.). Expectations are the selling points of the model (e.g., "fuel efficiency," "high output," "spacious interior," etc.). Vehicle functions are the functions that the model possesses. These items, as indicated by code 501, include multiple sub-items, each of which is assigned a score. Note that the multiple sub-items (code 501) are assigned to a virtual user. It corresponds one-to-one with the defined preference (code 401).

[0040] The parameters that an automobile possesses, as illustrated in Figure 5, are called vehicle parameters. By utilizing user parameters and vehicle parameters, it is possible to calculate the degree of fit between a virtual user and a specific car model (i.e., a value representing how much the virtual user likes that model).

[0041] Virtual user data 202A may be generated using data from existing users. Figure 6(A) shows an example of a method for generating virtual user data 202A. For example, if an automobile manufacturer has a customer database, virtual user data can be generated by applying anonymization and augmentation processing to the contents of that database.

[0042] Anonymization is the process of removing or altering elements that can identify an individual (for example, name, address, telephone number, date of birth, or combinations thereof). Examples of such alterations include removing information below the city / ward level from a detailed address, or removing the month and day from a date of birth. Another example is replacing these elements with random values.

[0043] Expansion processing is the process of generating a large number of virtual users from a limited set of users. For example, the distribution of parameters may be estimated based on data from real customers, and virtual users may be generated based on the estimation results. For example, if the age distribution, income distribution, and vehicle ownership distribution in a certain region are known, data on virtual users residing in that region may be automatically generated based on that known data. This makes it possible, for example, to generate 100,000 virtual users based on customer data from 1,000 people.

[0044] If customer data consists of multiple databases, it may be possible to perform a process to remove duplicates before anonymizing the data. Figure 6(B) illustrates another method for generating virtual user data 202A. For example, a car manufacturer may have multiple dealerships, each with its own customer database. Additionally, the car manufacturer's website may store online member data. In such cases, there is a possibility of duplicate users across multiple databases. Therefore, a process to eliminate duplicates is performed before anonymization and enhancement processing, as described above. For instance, the personal information before anonymization can be used as a key to link the data before determining duplicates. The processing after eliminating duplicates is similar to the example in Figure 6(A).

[0045] The input / output unit 103 is a unit that receives input operations performed by the administrator of the server device 100 and presents information to the administrator. The input / output unit 103 is configured with, for example, a liquid crystal display, a touch panel display, or a hardware switch.

[0046] Note that the configuration shown in Figure 2 is just one example, and all or part of the illustrated functions may be performed using specially designed circuits. Furthermore, program storage and execution may be performed using combinations of main memory and auxiliary memory other than those shown.

[0047] Next, we will explain the details of the processing performed by the control unit 101. Figure 7 is a diagram showing the flow of data input and output between the multiple functional modules of the control unit 101 and the database. As a prerequisite for processing, the virtual market database is assumed to contain virtual user data and vehicle data that replicate the current automotive market. For example, the virtual user data is: This data is generated based on actual customer data, and the vehicle data represents the models of cars currently being sold.

[0048] The administrator uses the server device 100 to conduct market analysis and determine what measures should be introduced into the market (for example, whether to redesign an existing model, launch a new model, or discontinue sales of an existing model). As mentioned above, the analysis unit 1011 can provide five types of analysis methods. Based on instructions from the administrator, the analysis unit 1011 conducts market analysis in a predetermined manner and outputs the results. Figure 8 shows the data output by each analysis method.

[0049] (1) Market overview analysis Market overview analysis outputs the results of mapping the relationships between arbitrary user parameters for users who own a particular model of automobile. Figure 9 shows an example of the data output in market overview analysis. In this example, the user's life stage, gender, and age are set as the analysis axes, and a table (distribution map) mapping the distribution of the number of users who own a specific model is output. Furthermore, it is possible to set user parameters other than those exemplified, as long as they are defined in the virtual user data. This analysis method allows us to understand the distribution of the number of users for each attribute.

[0050] (2) Analysis of competing vehicle models Competitive vehicle analysis is a method for visualizing other models that are likely to be compared to a particular car model. For example, if virtual user data includes information about models that have been compared or models that have been previously owned, a ranking of competing models (car ranking) can be generated by taking statistics from multiple virtual users. Figure 10 shows an example of a competitive vehicle ranking generated for a specific model. According to this analysis method, it is possible to extract other models that are similar in preference to users who own a particular car model.

[0051] (3) Important indicator analysis Key performance indicator analysis is a method for analyzing and visualizing the reasons why a particular model is supported by users. In this embodiment, multiple virtual users have preference parameters, as shown in Figure 4. Therefore, by analyzing the preferences of users who own model A and users who own model B, it is possible to analyze "what makes model A attractive compared to model B."

[0052] Specifically, the system calculates significant differences and effect sizes regarding preferences among multiple virtual users who own the target vehicle and multiple virtual users who own the comparison vehicle, and outputs a ranking (ranking of key indicators). Significant differences and effect sizes can be calculated using, for example, p-values ​​or chi-squared tests. For example, if a virtual user who owns model A tends to place more importance on the image of "cuteness" than a virtual user who owns model B, then it can be determined that the "cuteness" characteristic of model A is well-received by users.

[0053] Figure 11 shows an example of the data output when comparing two models, A and B. In this example, the highest statistical significance and effect size were observed for the image of "cute." In such cases, it can be said that for model A, a strategy to change the image of "cute" should not be adopted.

[0054] (4) User analysis User analysis is a method of visualizing what parameters the target demographic has when actually purchasing a particular car model. User analysis can be performed by extracting virtual users who own a specific model from virtual user data and then compiling statistics on their user parameters. Figure 12 shows an example of statistical analysis of the attributes of virtual users who own a specific model, compiled into an analysis report.

[0055] (5) Potential analysis Potential analysis is a method for visualizing which segments of consumers, based on their parameters, might be interested in a particular automobile model. For example, even someone who has purchased a station wagon might have a strong preference for SUVs. Therefore, by analyzing the preferences defined in the virtual user data, it is possible to identify the segment of the population that might be interested in a particular model of car.

[0056] Potential analysis involves setting arbitrary user parameters as analysis axes for hypothetical users who own a particular model of car, and outputting a distribution map that maps the strength of their preferences. For example, in the example in Figure 13, it is shown that among hypothetical users who own a particular model of car, the largest number prefer 2-box compact cars, followed by the largest number of hypothetical users who prefer compact SUVs. In such cases, it can be concluded that compact SUVs have the potential to meet the demand.

[0057] Returning to Figure 7, let's continue the explanation. The results of the analysis by the analysis unit 1011 are provided to the administrator via the input / output unit 103. Based on the market analysis results, the administrator plans the measures to be introduced into the market. Next, the administrator inputs the details of the planned measures into the server device 100 (prediction unit 1012). Figure 14 shows an example of a screen for inputting measures. In this embodiment, as shown, "Maintain current status," "Abolish," and "Change" can be selected for each model. It is also possible to add new models. Figure 15 shows an example of a screen for inputting vehicle parameters for a new model. In this example, the screen used when adding a new model is shown as an example, but a similar screen is used when performing a model change.

[0058] The forecasting unit 1012 simulates the market's response when the input measures are implemented, based on virtual user data and vehicle data included in the virtual market database. Specifically, the forecasting unit 1012 performs the following processes:

[0059] (1) Use a virtual market database to generate a temporary database for use in the simulation. To simulate how the market will react to the implemented measures, a copy of the existing virtual market database is made, and a temporary virtual market database (hereinafter referred to as the second database) is created for the simulation. In this process, the prediction unit 1012 reflects the input measures in the second database. For example, when testing the measure "add a new model," the data for the new model is added to the vehicle data held in the second database. The second database is temporarily stored in the storage unit 102 until the simulation is completed.

[0060] (2) After implementing the measures, vehicle data is matched with virtual users, and a score is calculated. Next, the prediction unit 1012 performs matching using multiple vehicle parameters from the vehicle data and multiple user parameters from the virtual user, and calculates a score (hereinafter referred to as the matching score) for each virtual user that represents the degree of fit with each of the multiple models. The matching score may be calculated using a mathematical formula or other method based on the degree of agreement between parameters, or it may be calculated using a machine learning model. This process makes it possible to generate a model-specific fitness ranking for each of multiple virtual users, as shown in Figure 16.

[0061] (3) Predict car purchases based on the matching score. Next, the prediction unit 1012 predicts the virtual user's purchasing behavior (whether or not to purchase a suitable model of automobile). The prediction can be made based on the virtual user's income, life stage, age of the automobile currently owned, mileage, timing of vehicle inspections, presence or absence of sales activities, presence or absence of sales promotion activities (such as the timing of commercial broadcasts), etc. The prediction process may be performed using, for example, a machine learning model.

[0062] For example, by using a machine learning model that has learned the relationship between a matching score and the multiple factors mentioned above for users with a history of purchasing cars in the past, it is possible to predict whether a virtual user will actually purchase a car under certain conditions. The prediction unit 2012 updates the virtual user data in the second database based on the prediction results. For example, if the prediction result is that a certain virtual user will purchase a car, the cars owned by that virtual user will be updated.

[0063] The processes described in (2) and (3) above are performed for all virtual users included in the second database. This makes it possible to simulate what kind of market reaction can be expected when a certain measure is introduced to the market. For example, it becomes possible to analyze how the market share of multiple automobile manufacturers changes as a result of replacement purchases, or how the user distribution for each model within the same manufacturer changes.

[0064] Figure 17 shows an example of simulated data showing how market share, sales volume, and revenue change before and after the implementation of measures for multiple automobile manufacturers. Figure 18 shows an example of simulated data showing how sales volume for each model changes before and after the implementation of measures for a specific automobile manufacturer.

[0065] Based on the analysis results, the administrator decides whether or not to update the virtual market database with the entered measures. If the virtual market database is to be updated with the entered measures, the forecasting unit 1012 overwrites and updates the virtual market database with the second database. As a result, the virtual market database reflects "the state of the market after the measures entered by the administrator have been implemented."

[0066] The server administrator can then perform market analysis on the updated virtual market database and determine the next course of action. They can also run simulations based on these actions and reflect the results back into the virtual market database. This enables simulations across multiple generations, allowing for more accurate, time-series simulations.

[0067] Figure 19 is a flowchart of the processes executed by the prediction unit 1012. The processes shown are initiated based on a request from the administrator. First, in step S11, the system retrieves the measures entered by the administrator. As illustrated in Figure 14, there are four types of measures that can be entered: maintain current status, abolish, change, and introduce a new vehicle (new model). If change or introduce a new vehicle is selected, the system accepts input of specific vehicle parameters. In step S12, a second database is generated, which is a copy of the virtual market database. The policies entered by the administrator are also reflected in the second database. Depending on the measures taken, the vehicle data held in the second database will be updated.

[0068] In step S13, a matching score is calculated for each of the multiple virtual users based on the vehicle data and virtual user data held in the second database. This results in a ranking of matching scores for each virtual user, as shown in Figure 16.

[0069] In step S14, the purchasing behavior of each virtual user is predicted based on the matching score generated for each virtual user. The prediction results are reflected in the virtual user data held in the second database. In step S15, market statistics are compiled and output based on the forecast results. These statistics can be performed from multiple perspectives, such as model-specific inflow and outflow forecasts, inter-company inflow and outflow forecasts, and portfolio forecasts. In step S16, based on the administrator's instructions after reviewing the simulation results, a decision is made as to whether or not to reflect the currently entered measures in the virtual market database. If the result in this step is positive, i.e., if there is an instruction to reflect the measures in the virtual market database, the process proceeds to step S17, and the second database is overwritten with the virtual market database. If the result in step S16 is negative, i.e., if there is an instruction to discard the measures, the process ends.

[0070] As described above, the server device according to this embodiment maintains a database representing a virtual market, enabling market analysis and simulation of market reactions when any measure is introduced to the market. This allows for the determination of more appropriate measures. Furthermore, by reflecting the results of the implemented measures in the virtual market database, market analysis in the next step becomes possible. In other words, it becomes possible not only to decide on a single measure, but also to derive answers to questions such as "what measures should be introduced continuously?"

[0071] In this embodiment, an example of inputting measures related to the company's own products was given, but the input of measures is not limited to those related to the company's own products. For example, it is possible to assume that a competitor has implemented some measures and input the details of those measures. This makes it possible to plan appropriate measures that the company should take in response to the actions of other companies.

[0072] (Second Embodiment) The circumstances of car owners (e.g., age, life stage, income, etc.) can change over time. The second embodiment is an embodiment that reflects the lifestyle of a virtual user, which changes over time, in the database.

[0073] The server device 100 according to the second embodiment is further configured to have a function that simulates the passage of a specified period and reflects the results in virtual user data included in the virtual market database. Figure 20 is a diagram showing in detail the components of the server device 100A according to the second embodiment. The server device 100A according to the second embodiment further comprises an update unit 1013. The update unit 1013 simulates the passage of a specified period and updates the user parameters of virtual users included in the virtual market database.

[0074] For example, consider a case where a further measure is introduced three years after an initial measure has been launched in the market. While the server device 100 according to the first embodiment cannot account for the passage of three years, in reality, various user parameters such as the user's age, income, life stage, or family environment change over time. For example, the number of children may increase over time, making a car too small. Therefore, in order to conduct accurate market analysis, These changes must be reflected in the database. Figure 21 illustrates the status of virtual users as it changes over time.

[0075] The update unit 1013, when a certain period (for example, 3 years) is specified, estimates the status of virtual users as if that period had elapsed and updates the corresponding user parameters. How the status of virtual users changes can be estimated using statistical data, etc. Furthermore, the update unit 1013 may estimate the number of virtual users who retired from driving during the period and delete those virtual users. It may also estimate the number of virtual users who newly obtained a driver's license during the period and add those virtual users.

[0076] In the second embodiment, market analysis can be performed by the analysis unit 1011 after the virtual user data has been updated. This allows for market analysis that reflects the passage of time. For example, it becomes possible to perform simulations such as, "How will market users react if another measure is introduced three years after the first measure was introduced?"

[0077] (Other variations) The embodiments described above are merely examples, and this disclosure may be modified as appropriate without departing from its essence. For example, the processes and means described in this disclosure can be freely combined and implemented, as long as no technical inconsistencies arise.

[0078] Furthermore, although the description of the embodiment mentions an example in which the control unit 101 and the storage unit 102 are located in the same device, the device that stores the virtual market database may be a separate device from the server device 100.

[0079] Furthermore, while the description of the embodiment showed an example of changing the lineup of automobiles sold as a measure to introduce a product to the market, other measures related to automobile sales may also be used in combination. For example, conducting campaigns such as television commercials and online advertisements for a particular model of automobile may increase media exposure and lead to user purchasing behavior. In this way, advertising can also be included as part of the "measure."

[0080] In this case, a virtual advertisement (virtual ad) may be placed in the virtual market, and the purchasing behavior of virtual users may be predicted (for example, in step S14). The details of the virtual advertisement may be specified by the administrator. For example, the administrator may be asked to input the target of the advertisement (car model), the medium (television, newspaper, internet, etc.), the scale and duration, and after estimating its effect (for example, the number of users exposed to the advertisement), the purchasing behavior of virtual users can be predicted. Data related to the virtual advertisement may also be included in the virtual market database. With this configuration, it is possible to simulate what kind of advertisement would be more effective in appealing to users when the lineup of cars is changed.

[0081] Furthermore, a process described as being performed by a single device may be divided and executed by multiple devices. Conversely, a process described as being performed by different devices may be executed by a single device. In a computer system, the hardware configuration (server configuration) by which each function is implemented can be flexibly changed.

[0082] This disclosure can also be realized by supplying a computer program implementing the functions described in the above embodiments to a computer, and having one or more processors in the computer read and execute the program. Such a computer program is stored on a non-temporary computer-readable storage medium that can be connected to the computer's system bus. It may be provided to a computer or provided to a computer via a network. Non-temporary computer-readable storage media include, for example, any type of disk such as magnetic disks (floppy disks, hard disk drives (HDDs), etc.), optical disks (CD-ROMs, DVDs, Blu-ray discs, etc.), read-only memory (ROM), random access memory (RAM), EPROM, EEPROM, magnetic cards, flash memory, optical cards, and any type of medium suitable for storing electronic instructions. [Explanation of symbols]

[0083] 10.. Vehicles 100... Server device 101... Control Unit 102...Storage section 103...Input / output section

Claims

1. A storage device that stores a database including first data describing multiple virtual users, second data describing multiple car models sold in the market, and third data describing virtual advertisements placed in the market, To acquire a sales promotion measure for the aforementioned market, which includes at least one of the following, as a change in the sales lineup of the automobiles: the introduction of a new model to the market, the modification of an existing model in the market, and the discontinuation of an existing model in the market; and the placement of the virtual advertisement for the at least one of the above in the market. Based on the first data, the second data, the third data, and the sales promotion measures, predict the purchasing behavior of multiple virtual users after the sales promotion measures have been implemented. The predicted purchasing behavior for the aforementioned multiple virtual users is reflected in the first data and the second data, A control device that performs the following: An information processing device having

2. The first data includes a plurality of user parameters that influence the automobile purchasing behavior, defined for each of the virtual users. The information processing apparatus according to claim 1.

3. The user parameters include at least one of the virtual user's preferences, life stage, and economic situation. The information processing apparatus according to claim 2.

4. The user parameters include data about the car owned by the virtual user. The information processing apparatus according to claim 2.

5. The second data includes a plurality of vehicle parameters defined for each of the plurality of automobile models, The information processing apparatus according to claim 2.

6. The control device calculates the degree of fit between the virtual user and each of the multiple automobile models based on the result of matching the user parameters and the vehicle parameters, and predicts the purchasing behavior based on the degree of fit. The information processing apparatus according to claim 5.

7. The control device predicts market changes after the multiple virtual users have taken the purchasing action and outputs the results. The information processing apparatus according to any one of claims 1 to 6.

8. The control device further performs market analysis based on the first data reflecting the purchasing behavior and the second data, and outputs information for determining new sales promotion measures. The information processing apparatus according to any one of claims 1 to 6.

9. The control device simulates the passage of a specified date and time, and updates the first data based on the results of the simulation. The information processing apparatus according to any one of claims 1 to 6.

10. The steps include obtaining first data describing multiple virtual users, second data describing multiple car models sold in the market, and third data describing virtual advertisements placed in the market, A step of acquiring a sales promotion measure for the market, which includes at least one of the following, as a change in the sales lineup of the automobile, the introduction of a new model to the market, the modification of an existing model in the market, and the discontinuation of an existing model in the market, and the placement of the virtual advertisement for the at least one of the above in the market. A step of predicting the purchasing behavior of a plurality of virtual users after the implementation of the sales promotion measures, based on the first data, the second data, the third data, and the sales promotion measures. A step of reflecting the predicted purchasing behavior for the aforementioned multiple virtual users in the first data and the second data, Information processing methods, including those mentioned above.

11. The first data includes a plurality of user parameters that influence the automobile purchasing behavior, defined for each of the virtual users. The information processing method according to claim 10.

12. The user parameters include at least one of the virtual user's preferences, life stage, and economic situation. The information processing method according to claim 11.

13. The user parameters include data about the car owned by the virtual user. The information processing method according to claim 11.

14. The second data includes a plurality of vehicle parameters defined for each of the plurality of automobile models, The information processing method according to claim 11.

15. Based on the matching of the user parameters and the vehicle parameters, the degree of fit between the virtual user and each of the multiple automobile models is calculated, and the purchasing behavior is predicted based on the degree of fit. The information processing method according to claim 14.

16. The process further includes the step of predicting market changes after the multiple virtual users have taken the purchasing action and outputting the results. The information processing method according to any one of claims 10 to 15.

17. The further step includes conducting a market analysis based on the first data reflecting the aforementioned purchasing behavior and the second data, and outputting information for determining new sales promotion measures. The information processing method according to any one of claims 10 to 15.

18. The process further includes the step of simulating the passage of a specified date and time, and updating the first data based on the results of the simulation, The information processing method according to any one of claims 10 to 15.

Images