JP2008282247A - Device and method for calculating evaluation point of planned commodity, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology allowing calculation of an evaluation point about a factor group corresponding to coordinates of a planned commodity plotted on a perception map. <P>SOLUTION: An arithmetic device 4 generates planned commodity assumption evaluation point data, and calculates the assumed coordinates (assumption coordinates) of the planned commodity in the perception map. The assumption evaluation point about the factor group to the planned commodity corresponding to the assumption coordinates of the planned commodity closest to designation coordinates of the planned commodity designated in the perception map is selected as an estimation evaluation point of the planned commodity, a second change amount parameter corresponding to a second change amount about a component corresponding to an eigenvalue not associated to the perception map among components of a factor eigenvector is calculated about a prescribed factor designated with a target evaluation point, and a planned commodity evaluation point is calculated by use of the second change amount parameter to the component corresponding to the eigenvalue not associated to the perception map among the components of the factor eigenvector to the estimation evaluation point to the planned commodity, and is output. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a perceptual map obtained by performing a perceptual map creation method based on an evaluation point for a factor group for a product group, and an evaluation score for the factor group of the planned product corresponding to the coordinates designated on the perceptual map. This technology relates to technology that supports product planning. The perceptual map creation method here refers to the singular value decomposition processing performed on the evaluation points for the factor group for the product group, and the component corresponding to the eigenvalue representing the product and the feature of the factor represented by the evaluation point, This is a technique that reduces the product group and factor group to a dimension that can be expressed in a perceptual map. Examples of the map creation method include methods such as quantification type III, principal component analysis, biplot analysis, correspondence analysis, and the like.

  The product provider predicts an image of the product that the consumer wants and plans a new product based on the predicted image. However, the image predicted by the product provider may not match the image of the product actually requested by the consumer. For this reason, various investigation techniques and analysis techniques have been developed to know the product image demanded by consumers and support the product planning.

There are survey methods for examining the degree of preference of each product in various answer formats for a product group consisting of a plurality of types of products. For example, for the product field of automobiles, we created a questionnaire that corresponds to “factors” consisting of words representing images such as “fast”, “luxury” and “want to go out,” and asked respondents about each vehicle. There is a survey method to evaluate in stages how much each factor representing the image applies. From the survey results obtained by such a survey, for example, it can be seen what image each of the n types of vehicles (products) from vehicle types 1 to n is recognized by the respondent.
However, it cannot be said that all the information obtained from such a survey is drawn out only from the total results of the obtained answers. For example, in addition to the aggregated results, the relationship between the surveyed products is recognized, and the relationship between the factors that consist of words representing each image is recognized. If we can extract information about the strength of the relationship between them and the degree of their relationship, we can use the survey results more effectively. Analytical methods that enable such analysis have been developed.

In the above survey, items to be surveyed such as “product type” and “factor” are referred to as “items”, and a general term for these items is referred to as “category”. For example, when the evaluation of the factor group is investigated for the product group, the category of the item “product type” and the category of the item “factor” exist. The category and the evaluation given to each item belonging to the category are referred to as “category data”. A method has been developed that uses the statistical relevance of multiple sets of category data to give numerical values to the items that make up the category. For example, methods for creating a perceptual map such as quantification type III, principal component analysis, biplot analysis, correspondence analysis, and the like have been developed, and numerical values can be given to the items themselves.
The numerical value given to the item itself constituting the category represents the “item structure”. The item structure here refers to the relationship and similarity of all items included in the category data. For example, there are two categories, a category “car” composed of n items (product types) and a category “factor” composed of m items, and the evaluation value for a combination of these category items is n × m. When a matrix is configured, according to the perceptual map creation method described above, components representing a plurality of relationships and similarities included in items are calculated. The numerical value given to the item itself represents the degree of the relationship or similarity between items in the feature represented by the component. The numerical value given to the item itself is calculated in consideration of the relationship and similarity of all items.

For example, in the correspondence analysis, it is possible to calculate a component that maximizes the correlation between the two categories “product type” and “factor”. Furthermore, based on the calculated component, a numerical value that makes the correlation as large as possible can be given to the items constituting the categories of “product type” and “factor”. Using a numerical value given to each item constituting the category and the component, a perceptual map representing the relationship between these items can be created. In the coordinate space constituting this perceptual map, items having similar structures of the items described above appear in the same direction.
On the perceptual map obtained by correspondence analysis, both the item of product type and the item of factor can be plotted. A perceptual map (product map) on which product types are plotted and a perceptual map (factor map) on which factors are plotted can be created separately. Alternatively, since the product map and the factor map share the components used as coordinate axes, it is possible to create a product / factor map in which both the product type and the factor are plotted and displayed on a single perceptual map.

In the technology of Patent Document 1, based on the above-described method, a correspondence analysis is performed on the evaluation score for each factor for each product to create a product map and a factor map, and information on the demand amount is displayed on these maps as contour lines. To do. According to the technique of Patent Document 1, it is possible to visualize the relationship between a product and a factor that characterizes the product and a demand amount, and to provide information directly related to product planning. By using the technique of Patent Document 1, it is possible to visually display the potential demand and provide information directly related to product planning.
According to the technique of Patent Document 1, when planning a new product, the evaluation point for each factor can be investigated for the planned product as well as the existing product type, and displayed on the perceptual map. In addition, since information on potential demand can be displayed on the perceptual map, the image of the planned product can be examined from the positional relationship between the position of the planned product and the potential demand. For example, identify the position where the potential demand is high on the map, and consider the product image and concept so that the planned product is closer to that position (ie, closer to the image that the consumer wants) can do.
Japanese Patent Application No. 2006-117984

In the technique of Patent Document 1, by creating a perceptual map, the amount of information that the evaluation score for each factor in each product type originally has in multiple dimensions can be expressed in a perceptual map (in many cases) Then, it is reduced to 2 dimensions or 3 dimensions). For example, when there are a large number of product types and factors, and trying to create a two-dimensional perception map for them, each product type has an evaluation score for those many factors, Information such as the relationship between types and the relationship between items is reduced to a scale represented by two axes.
Therefore, even if you try to calculate evaluation points for each factor that the planned product originally has from the position of the planned product in the reduced dimension (that is, perceptual map), a series of evaluation points There is a problem that it is difficult to calculate uniquely. For example, when considering the image and concept of a product so that the position of the planned product is a specific position on the perception map, how much score should be given for each of a number of factors for the product? I can't know if it's good. In the technique of Patent Document 1, even though the position of the planned product can be examined from the positional relationship with the potential demand amount shown on the perceptual map, the plan for each factor is determined from the position of the planned product. It is impossible to know information on how much product evaluation points are required.

  In the present invention, in the perceptual map obtained by performing the perceptual map creation method, the evaluation points for the factor group corresponding to the coordinates of the planned product plotted on the perceptual map are calculated. Provide technologies that can provide information directly related to product planning.

  In the perceptual map obtained as a result of performing the perceptual map creation method based on the evaluation score for the factor group for the product group, the apparatus of the present invention relates to the factor group of the planned product corresponding to the coordinates designated on the perceptual map. Calculate the evaluation score. The apparatus includes original evaluation point data describing evaluation points for the factor group for a product group, eigenvalue data describing eigenvalues obtained by performing a perceptual map creation method based on the original evaluation point data, Product eigenvector data describing eigenvectors for product groups corresponding to eigenvalues, factor eigenvector data describing eigenvectors for factor groups corresponding to eigenvalues, and product coordinate data describing coordinates in the perceptual map for each product group; And means for storing factor coordinate data describing coordinates in the perceptual map for each factor group.

The above-described original evaluation point data can be obtained by conducting a survey with respect to the consumer for each of the image items represented by the factor group for the product group. When a perceptual map creation method such as correspondence analysis is applied to the evaluation points described in the original evaluation point data, eigenvalues that characterize the relationship between the product group and the factor group are calculated. The eigenvector and the eigenvector of the factor group are calculated. Intuitively grasp the relationship between products, the relationship between factors, and the relationship between products and factors by expressing the eigenvectors of the product group and the eigenvectors of the factor group on the same perceptual map Can do. The coordinates of the eigenvector of the product group and the eigenvector of the factor group in the perceptual map are determined from the components corresponding to the eigenvalue associated with the perceptual map among the components of the eigenvector.
Note that the above-described storage unit only needs to store data describing values calculated by the perceptual map creation method, and the perceptual map creation method to be executed is not limited to the above-described correspondence analysis. For example, data describing values obtained as a result of performing a perceptual map creation method such as quantification type III, principal component analysis, and biplot analysis may be stored.

  The above-described apparatus includes means for acquiring planned product designation coordinate data describing coordinates (designated coordinates) designated in the perception map of the planned product. For example, if the apparatus of the present invention is mounted on an information processing apparatus or the like capable of displaying a perceptual map, the above-described means can perform the coordinates of the position specified by the user on the perceptual map displayed on the display apparatus. May be obtained. Alternatively, numerical values input by an operation of a keyboard or the like may be acquired as designated coordinates.

  The above-described apparatus selects a reference product (reference product) from the product group, extracts evaluation points for the factor group for the reference product from the original evaluation point data, and outputs the evaluation points as reference product evaluation point data. Means. The reference product may be selected from a product group having coordinate values on the perception map, and is not particularly limited. For example, the product plotted at the closest position from the designated coordinates of the planned product designated on the perception map may be selected as the reference product.

The above-described apparatus is configured to estimate estimated points (factors) for a factor group for a planned product based on planned product designation coordinate data, reference product evaluation point data, original evaluation point data, eigenvalue data, and factor eigenvector data. (Evaluation point) and the eigenvalue associated with the perceptual map among the components of the factor eigenvector as the change amount (first change amount) from the evaluation point for the factor group for the reference product to the estimated evaluation point for the factor group for the planned product Is provided with means for calculating a value (first variation parameter) expressing the component corresponding to 1 as a scale and outputting it as planned product estimated evaluation point data and first variation parameter data, respectively.
The first change amount represents the difference between the evaluation score of the reference product and the estimated evaluation score of the planned product, using the evaluation score for each factor group as a scale. On the perception map, the coordinate positions of the reference product and the planned product differ based on this difference.
Further, the first change amount is represented by a movement amount from the coordinates of the reference product to the designated coordinates of the planned product on the perception map. Since the coordinates of products (including planned products) in the perceptual map are calculated from the evaluation points for each factor in the factor group, the first variation corresponds to the eigenvalue associated with the perceptual map among the components of the factor eigenvector It can be calculated as a value (first variation parameter) expressing the component to be measured as a scale. For example, if the coordinates of the planned product are specified in a two-dimensional perceptual map, the first change amount specific to each dimension is measured using the component corresponding to each of two eigenvalues associated with the perceptual map as a scale. Parameters are calculated. The estimated evaluation point of the planned product is obtained by adding a numerical value obtained by multiplying the eigenvalue vector corresponding to each eigenvalue associated with the perceptual map and the first variation parameter (that is, the first variation amount) to the evaluation score of the reference product. Can be calculated with When the first change amount parameter is changed, a combination of different estimated evaluation points can be calculated.
By using the estimated evaluation point, the original evaluation point of the product group, the eigenvalue, and the factor eigenvector, the coordinates of the planned product in the perceptual map when the estimated evaluation point is taken can be calculated. A combination in which the coordinates on the perceptual map calculated for the estimated evaluation point become the designated coordinates of the planned product is output as the planned product estimated evaluation point data, and the first change parameter when the estimated evaluation point is taken is the first change parameter. It can be output as quantity parameter data. Further, the means for outputting the planned product estimated evaluation point data and the first change amount parameter data described above calculates the estimated evaluation point by changing the first change amount, and the coordinates on the perceptual map calculated for the estimated evaluation point The first change amount parameter corresponding to the first change amount that becomes the designated coordinates of the planned product may be calculated.

Further, the above-described apparatus includes means for acquiring planned product target evaluation point data describing a target evaluation point (target evaluation point) for the planned product for a predetermined factor in the factor group.
In general, when examining the image concept of a planned product, the relative relationship with the image of existing products is often considered. For example, when planning the next model of an existing product (current model), it may be desired to add a new image concept while following the main image of the current model. Alternatively, an image that is “urban” can receive the same evaluation as a specific existing product, and a higher evaluation can be obtained for an image that is “gorgeous” than that existing product. There are cases where you want to build an image concept for a planned product.
The means described above can set a target evaluation point intended by the product planner. For example, the product planner says that among the evaluation points regarding the factor group of the planned product, the image of being “urban” has the same evaluation score “55” as a specific existing product, and is “gorgeous” For the image, the target evaluation score can be set so that the evaluation score “70” is higher than the evaluation score “60” of the existing product. At this time, the target evaluation score “55” for the factor “urban” and the target evaluation score “70” for the factor “gorgeous” are acquired by the above-described means. In the planned product target evaluation point data, a target evaluation point (fixed value) for one factor may be set, or each target evaluation point may be set for a plurality of factors.

The above-described apparatus is configured to perform a predetermined factor for which a target evaluation point is set based on planned product estimated evaluation point data, planned product target evaluation point data, and factor eigenvector data, from the estimated evaluation point for the planned product to the planned product. Calculating a value (second variation parameter) that represents the amount of change to the target evaluation point (second variation) as a measure of the component corresponding to the eigenvalue that is not associated with the perceptual map among the components of the factor eigenvector And means for outputting the second change amount parameter data.
For example, even when a target evaluation point “55” for the factor “urban” and a target evaluation point “70” for the factor “gorgeous” are set, the estimated evaluation of the planned product described in the planned product estimated evaluation point data A point is not necessarily its numerical value. In many cases, there is a discrepancy between the estimated evaluation score and the target evaluation score for the factor “urban” and the factor “gorgeous”. The second change amount can be calculated from the difference between the estimated evaluation point and the target evaluation point in the predetermined factor.
The second change amount described above represents the difference between the estimated evaluation point and the target evaluation point of the planned product, using the evaluation point for each factor group as a scale. In the above-mentioned means, the component corresponding to the eigenvalue not associated with the perceptual map among the components of the factor eigenvector, that is, the component that is contracted when generating the perceptual map and is not reflected on the perceptual map. As a scale, a value (second variation parameter) expressing the second variation is calculated.
The first variation parameter with the component corresponding to the eigenvalue associated with the perceptual map among the components of the factor eigenvector as a scale is known at the time of calculating the second variation parameter. If these first change amount parameters change, the coordinates of the planned product assumed on the perceptual map change from the position closest to the designated coordinates, which is not preferable. Accordingly, by adjusting the second change amount parameter that does not affect the coordinates on the perceptual map, the estimated evaluation point related to the predetermined factor of the planned product can be brought closer to the target evaluation point.
According to a perceptual map creation method such as correspondence analysis, a plurality of eigenvalues are calculated from evaluation points for factor groups for a product group. In the means for calculating the second variation parameter and outputting it as the second variation parameter data, the second variation parameter for each of the components corresponding to the eigenvalues not associated with the perceptual map is used as a scale. Or the second variation parameter may be calculated using a component corresponding to the specified eigenvalue as a scale.

The above-described apparatus multiplies each component corresponding to the eigenvalue not associated with the perceptual map among the components of the factor eigenvector by the estimated evaluation point for the planned product for each factor of the factor group. A means is provided for calculating an evaluation score for a factor group for the planned product by adding a value, and outputting the result as planned product evaluation score data.
The estimated evaluation point already includes a change in the evaluation point based on the difference in position on the perceptual map, which is represented by the first change amount parameter. By adding a value obtained by multiplying the component corresponding to the eigenvalue not associated with the perceptual map among the components of the factor eigenvector to the estimated evaluation point, the position on the perceptual map is set to the designated coordinate. While maintaining, an evaluation point closer to the target evaluation point can be obtained as planned product evaluation point data.
According to the present invention, it is possible to calculate an evaluation point corresponding to the designated coordinates of the planned product plotted on the perceptual map and close to the target evaluation point for the predetermined factor.

In the above-described apparatus, the means for calculating the second variation parameter is a characteristic value that is not associated with the perceptual map among the components of the factor eigenvectors at the estimated evaluation point for the planned product for the predetermined factor. It is preferable to calculate the second variation parameter so that a difference between an evaluation score obtained by adding a value obtained by multiplying each corresponding component and a target evaluation score for the planned product is minimized.
Since the target evaluation score is arbitrarily given, in some cases, it may not be possible to calculate the evaluation score for the factor group for the planned product that matches the target evaluation score for all the predetermined factors. Even in such a case, by adopting the above configuration, it is possible to obtain an evaluation score for a factor group for a planned product closest to the target evaluation score for all the predetermined factors.

  Further, in the above-described apparatus, the means for calculating the estimated evaluation point of the planned product and the first change parameter is an assumed first change parameter describing the assumed first change parameter (assumed first change parameter). Based on the means for generating data, the reference product evaluation point data, the assumed first variation parameter data, and the factor eigenvector data, the evaluation point assumed for the factor group for the planned product (assumed evaluation point) is calculated. Assuming the factor group for the planned product from the evaluation points for the product group and the factor group for the planned product based on the means for outputting the planned product hypothetical evaluation point data and the original evaluation point data and the planned product hypothetical evaluation point data A means for calculating an assumed standardized evaluation score obtained by standardizing the evaluation score and outputting it as planned product assumed standardized evaluation score data, eigenvalue data, A means for calculating the assumed coordinates (assumed coordinates) of the planned product in the perceptual map on the basis of the factor-specific vector data and the planned product assumed standardized evaluation point data, and outputting them as planned product assumed coordinate data; Based on the quantity parameter data, planned product assumed coordinate data, planned product assumed coordinate data, and planned product designated coordinate data, the assumption corresponding to the assumed coordinate of the planned product closest to the designated coordinate of the planned product in the perceptual map It is preferable to provide means for selecting the first evaluation parameter for the first variation parameter and the factor group for the planned product as the estimated evaluation score for the first variation parameter and the factor group for the planned product, respectively.

As described above, the first change amount represents the difference between the evaluation score of the reference product and the estimated evaluation score of the planned product using the evaluation score for each factor group as a scale, and the first change amount represents the factor eigenvector. Of the components, the component corresponding to the eigenvalue associated with the perceptual map can be calculated as a value (first variation parameter) expressed as a scale. When the first variation parameter is changed, different estimated evaluation points Combinations can be calculated. Therefore, the means for generating the assumed first change parameter data generates the assumed first change parameter by assuming various values. If different assumed first change amount parameters are used, the first change amount can be changed to various values.
In the means for calculating the assumed evaluation score and outputting it as the planned product assumed evaluation point data, the planned product is obtained by adding the first change amount calculated using the above-described assumed first change parameter to the evaluation point of the reference product. Can be calculated. A combination of a plurality of assumed evaluation points can be calculated by changing the assumed first variation parameter.

When the hypothetical evaluation score is calculated, the hypothetical standardized evaluation score that standardizes the hypothetical evaluation score for the factor group for the planned product is calculated from the hypothetical evaluation score of the planned product and the original evaluation score of the product group, and standardization of the planned product assumption Can be output as evaluation point data.
Planned product hypothesis standardized evaluation score data is based on the assumption of the planned product and the product using, for example, a method of calculating a chi-square statistic against the original score of the product group and the assumed score of the planned product. It is possible to calculate and output the divergence of the expected value of the evaluation points for the group and the factor group for the planned product. By standardizing the assumed evaluation score, it is possible to remove the effects specific to the product (row element) and the factor (column element) from the assumed evaluation score, and to convert it into a numerical value representing the relationship between the product and the factor.

  When the assumed standardized evaluation score of the planned product is calculated, the assumed coordinates (assumed coordinates) of the planned product in the perceptual map can be calculated corresponding to the assumed standardized evaluation score of the planned product. The assumed coordinates of the planned product can be calculated, for example, by dividing the eigenvector assumed for the planned product by the square root of the sum of the evaluation score of the product group and the evaluation score of the planned product. At this time, the eigenvector assumed for the planned product includes the eigenvalue and factor eigenvector stored in the storage means (that is, the eigenvalue and factor eigenvector calculated without adding the evaluation score of the planned product) and the above-described assumption standardized evaluation. Calculated using points. This is based on the premise that the relationship between the product group and the factor group does not change even if the planned product is newly added to the product group. This makes it possible to calculate assumed coordinates on the perceptual map corresponding to the evaluation points assumed for the planned product.

  A combination of a plurality of assumed evaluation points is calculated by changing the assumed first change amount parameter, and assumed coordinates are calculated corresponding to each combination of the plurality of assumed evaluation points. Since the designated coordinates in the perceptual map of the planned product are known, an assumed first variation parameter corresponding to the assumed coordinates of the planned product closest to the designated coordinates of the planned product in the perceptual map from the calculated assumed coordinates The assumed evaluation point for the factor group for the planned product can be selected as the first change amount parameter and the estimated evaluation point for the factor group for the planned product, respectively. The selected first variation parameter and estimated evaluation point are output as first variation parameter data and planned product estimated evaluation point data, respectively.

  The present invention can also be embodied as a method. The method of the present invention relates to the factor group of the planned product corresponding to the coordinates designated on the perceptual map in the perceptual map obtained as a result of performing the perceptual map creation method based on the evaluation point of the factor group for the product group. This is a method for calculating the evaluation score. The method includes original evaluation point data describing evaluation points for the factor group for the product group, eigenvalue data describing eigenvalues obtained by performing a perceptual map creation method based on the original evaluation point data, Product eigenvector data describing eigenvectors for the product group corresponding to the eigenvalues, factor eigenvector data describing eigenvectors for the factor groups corresponding to the eigenvalues, and coordinates in the perceptual map for each of the product groups Product coordinate data, a step of storing factor coordinate data describing coordinates in the perceptual map for each of the factor groups, and planned product designation coordinate data describing coordinates (designated coordinates) designated in the perception map of the planned product And the standard product from the product group (standard Product), extracting the evaluation points for the factor group for the reference product from the original evaluation point data, and outputting as the reference product evaluation point data, the planned product designation coordinate data, the reference product evaluation point data, Based on the original evaluation score data, eigenvalue data, and factor eigenvector data, the estimated score (estimated evaluation score) for the factor group for the planned product and the evaluation score for the factor group for the reference product are used for the planned product. A value (first variation parameter) representing a change amount (first variation amount) to the estimated evaluation point for the factor group as a scale of a component corresponding to the eigenvalue associated with the perceptual map among the components of the factor eigenvector. Calculate and output each of the planned product estimated evaluation point data, the first change parameter data, and a predetermined factor in the factor group. Process for obtaining planned product target evaluation point data describing the target evaluation point (target evaluation point) for the planned product, planned product estimated evaluation point data, planned product target evaluation point data, and factor-specific vector Based on the data, for the predetermined factor, the change amount (second change amount) from the estimated evaluation point for the planned product to the target evaluation point for the planned product is set to the eigenvalue that is not associated with the perceptual map among the components of the factor eigenvector. A step of calculating a value (second variation parameter) expressing the corresponding component as a scale and outputting it as second variation parameter data, and for each factor of the factor group, an estimated evaluation point for the planned product is 2 Add the value obtained by multiplying the change parameter by each component corresponding to the eigenvalue that is not associated with the perceptual map among the components of the factor eigenvector. In addition, the method includes a step of calculating an evaluation score for a group of factors for the planned product, and outputting the result as planned product evaluation score data.

  According to the above-described method, it is possible to calculate an evaluation point corresponding to the designated coordinates of the planned product plotted on the perceptual map and close to the target evaluation point for the predetermined factor.

  Furthermore, the present invention can be embodied as a program. The program according to the present invention relates to the factor group of the planned product corresponding to the coordinates designated on the perceptual map in the perceptual map obtained as a result of performing the perceptual map creation method based on the evaluation point of the factor group for the product group. Is a program for calculating the evaluation score. The program includes original evaluation point data describing evaluation points for the factor group for the product group, eigenvalue data describing eigenvalues obtained by performing a perceptual map creation method based on the original evaluation point data, Product eigenvector data describing eigenvectors for the product group corresponding to the eigenvalues, factor eigenvector data describing eigenvectors for the factor groups corresponding to the eigenvalues, and coordinates in the perceptual map for each of the product groups Product coordinate data, processing for storing factor coordinate data describing coordinates in the perceptual map for each of the factor groups, and planned product designation coordinate data describing coordinates (designated coordinates) designated in the perception map of the planned product And the standard product from the product group Select the reference product), extract the evaluation score for the factor group for the reference product from the original evaluation point data, and output it as the reference product evaluation point data, the planned product designated coordinate data, and the reference product evaluation point data Based on the original evaluation score data, the eigenvalue data, and the factor eigenvector data, the planned product is estimated from the evaluation score (estimated evaluation score) for the factor group for the planned product and the evaluation score for the factor group for the reference product. A value representing the change amount (first change amount) to the estimated evaluation point for the factor group with respect to a factor group as a scale corresponding to the eigenvalue associated with the perceptual map among the components of the factor eigenvector (first change parameter) And a process for outputting the estimated product estimated evaluation point data, the first variation parameter data, and a predetermined factor in the factor group. Process for obtaining planned product target evaluation point data describing the target evaluation point (target evaluation point) for the planned product, planned product estimated evaluation point data, planned product target evaluation point data, and factor-specific vector Based on the data, for the predetermined factor, the change amount (second change amount) from the estimated evaluation point for the planned product to the target evaluation point for the planned product is set to the eigenvalue that is not associated with the perceptual map among the components of the factor eigenvector. Calculate the value (second variation parameter) that expresses the corresponding component as a scale, and output it as second variation parameter data, and for each factor in the factor group, The two-variation parameter is multiplied by each component corresponding to the eigenvalue that is not associated with the perceptual map among the components of the factor eigenvector. A value is added to calculate a score for a factor group for the planned product, and the computer is caused to execute a process of outputting it as planned product evaluation score data.

According to the present invention, it is possible to calculate an evaluation score including information represented by various components (eigenvalues) for a group of factors for a planned product, using the coordinates of the planned product designated on the perception map. Since the target evaluation point for a given factor is specified, when the assumed coordinates of the planned product are closest to the specified coordinates of the planned product on the perceptual map, the estimated evaluation points that the planned product can take for the factor group are unique. The problem of not being determined is also resolved.
According to the present invention, it is possible to know how many evaluation points the planned product for which the position (designated coordinates) on the perceptual map is designated corresponds to each factor. The product planner can examine an image concept that can obtain the evaluation score based on the evaluation score of the factor group for the obtained planned product. Information directly related to product planning can be obtained.

  Embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing a schematic configuration of a product planning support apparatus 2 according to an embodiment of the present invention. The product planning support device 2 according to the present embodiment uses the evaluation points for each factor of each product obtained as a result of investigating the product image with respect to the consumer, and perceptual map in which the factor representing the product and the image is plotted And the evaluation score for each factor of the planned product is calculated from the coordinate value of the planned product plotted by the product planner at an arbitrary position on the perceptual map.

As shown in FIG. 1, the product planning support device 2 includes an arithmetic device 4, a storage device 12, a display device 10, a mouse 8, and a keyboard 6. The display device 10 is connected to the arithmetic device 4 and displays an image according to a signal input from the arithmetic device 4.
The mouse 8 is an input device including an input switch and an input roller. The mouse 8 is connected to the arithmetic device 4. By operating the input switch or moving the input roller by moving the mouse 8 on the desk, the command value is input from the mouse 8 to the computing device 4. The keyboard 6 is an input device having a large number of input switches. The keyboard 6 is connected to the arithmetic device 4, and a command value is input to the arithmetic device 4 by operating each input switch.

The storage device 12 is configured by a hard disk or the like, and is connected to the arithmetic device 4. As shown in FIG. 1, the storage device 12 includes product group evaluation point data 14, product group standardized evaluation point data 16, product group dimension reduction data 18, perceptual map data 20, and planned product designation coordinate data 28. The planned product estimated evaluation point data 30, the planned product target evaluation point data 22, the planned product evaluation point data 32, the correspondence analysis program 24, and the evaluation point calculation program 26 are stored. The contents of various data will be described in detail later.
The storage device 12 stores data input from the arithmetic device 4. Further, the storage device 12 inputs the stored data to the arithmetic device 4 according to the command value input from the arithmetic device 4.

  The arithmetic device 4 is constituted by a CPU, a ROM, a RAM, and the like. The arithmetic device 4 is connected to the storage device 12, the display device 10, the mouse 8, and the keyboard 6. The arithmetic device 4 executes various arithmetic operations according to command values input from the mouse 8 and the keyboard 6 and causes the display device 10 to display the arithmetic results. In addition, the arithmetic device 4 reads data stored in the storage device 12 and performs an arithmetic operation using the read data. Further, the arithmetic device 4 stores the data in the storage device 12. Further, the arithmetic device 4 reads out and executes the correspondence analysis program 24 stored in the storage device 12, thereby creating a perceptual map in which the product and factors representing the image of the product are plotted, and an evaluation score calculation program 26 is read and executed to calculate an evaluation point for each factor of the planned product from the coordinate value of the planned product plotted by the product planner at an arbitrary position on the perceptual map.

Next, the product group evaluation point data 14 stored in the storage device 12 will be described. FIG. 4 shows a diagram representing information described in the product group evaluation score data 14 (original evaluation score data). The evaluation score for the factor group (evaluation item) for the product group (product type) is obtained by examining the image of the product with respect to the consumer. This survey may be conducted by scoring each product for factors that represent the image, such as “youthful” and “luxurious”, or by evaluating the image of each product in a scale format. May be. In this embodiment, j factors are set for i types of products A to i. For example, the evaluation score of the factor “urban” for the product A is “13”, and the evaluation score of the factor “unique” for the product B is “83”. Hereinafter, the evaluation point of the factor j for the product i is expressed as “n _ij ”. This evaluation score may be a score based on the above-described survey method, a percentage of respondents, or the like. Moreover, the merchandise group evaluation score data 14 may be stored for each consumer grouped according to age, sex, preference, and the like.

Next, the commodity group standardized evaluation point data 16 stored in the storage device 12 will be described. The product group standardized evaluation point data 16 is obtained by standardizing the product group evaluation point data 14 using a chi-square test method or the like. The product group evaluation point data 14 includes not only an element representing the relationship between the product type and the factor type, but also an element that specifically affects the product type and an element that specifically affects the type of factor. There is. For example, depending on the consumer, the evaluation may be based on different value criteria for each type of product, or the value criteria may be different for each type of factor representing the image. May contain such differences in value standards. Therefore, from the evaluation score for each product and each factor combination, excluding the unique effects of row elements (product types) and column elements (factor types), the relationship between each product type and each factor element is determined. It is necessary to standardize the value a _{ij to be} represented (product group standardized evaluation point data 16). By standardizing the evaluation point n _ij for each factor of each product, it is possible to eliminate the variation of each evaluation point n _ij . The processing executed by the arithmetic device 4 at the time of standardization will be described using a flowchart described later.

Next, the product group dimension reduction data 18 will be described. FIGS. 5A and 5B are diagrams showing information described by the product group dimension reduction data 18. The product group dimension reduction data 18 includes eigenvalue data describing eigenvalues calculated from the product group standardized evaluation point data 16 using correspondence analysis or the like, and products and factors describing eigenvectors for each product and each eigenvalue of each factor. Of eigenvalue vector data. For example, the standardized evaluation points a _ij for the factor group for the commodity group can be regarded as an i × j matrix A having these values as elements. By decomposing this matrix A into t eigenvalues λ _t of the matrix A, eigenvectors u _{it of} row elements, and eigenvectors w _{jt of} column elements, the dimension of the standardized evaluation point a _ij can be reduced. . The product group dimension reduction data 18 includes data obtained by reducing the standardized evaluation points a _ij for each factor for a product in units of dimensions.
The product group evaluation score data 14 and the product group standardized evaluation score data 16 include judgments based on various evaluation levels for various types of factors, whereas in the product group dimension reduction data 18, each of the t dimensions is characteristic. Represents a typical evaluation level. The magnitude of the eigenvalue indicates the strength of the relationship between the characteristic evaluation level of each dimension with respect to the evaluation point. The larger the value, the more the judgment based on the characteristic evaluation level in that dimension becomes. It has a strong influence on For example, in the case of FIG. 5, the relationship between the product group and the factor group characterized by the one-dimensional eigenvalue having the largest eigenvalue has the strongest influence on the evaluation level of the evaluation score. Therefore, when the product planner examines the analysis result, it is necessary to examine what evaluation level is represented on the scale of the component corresponding to each eigenvalue.

  Next, the perceptual map data 20 will be described. FIG. 5C shows a diagram representing information described in the perceptual map data 20. The perceptual map data 20 is composed of product coordinate data describing the coordinates of each product and factor coordinate data describing the coordinates of each factor in a coordinate space having a dimension corresponding to the eigenvalue. The coordinates of each product and the coordinates of each factor are calculated from the eigenvector of each product corresponding to the eigenvalue and the eigenvector of each factor. From the perceptual map data 20, for example, for a product or a factor, a coordinate component corresponding to the largest eigenvalue (one dimension in FIG. 5) and a coordinate component corresponding to the next largest eigenvalue (two dimension in FIG. 5) are used. A two-dimensional perceptual map can be created. Alternatively, when creating a three-dimensional perceptual map, products and factors can be plotted on the perceptual map using coordinate components corresponding to eigenvalues from one to three dimensions. FIG. 6 shows a perceptual map created from the perceptual map data 20 of FIG.

The above-mentioned product group standardized evaluation score data 16, the product group dimensional reduction data 18, and the perceptual map data 20 are calculated by executing the correspondence analysis program 24 on the product group evaluation point data 14 by the arithmetic unit 4. Is done. The processing operation of the correspondence analysis program 24 will be described with reference to a flowchart described later.
The storage device 12 may store a perceptual map created by the correspondence analysis program 24.

  Next, the planned product designation coordinate data 28 will be described. FIG. 7 shows an example of designated coordinates of the planned product described by the planned product designated coordinate data 28. Designated coordinates on the perception map of the planned product can be determined by the product planner. When the designated coordinates of the planned product are input to the arithmetic device 4 by operating the keyboard 6 or the mouse 8, the designated coordinates of the planned product are stored in the storage device 12 as planned product designated coordinate data 28.

  Next, the planned product estimated evaluation point data 30 will be described. The planned product estimated evaluation point data 30 describes evaluation points (estimated evaluation points) estimated for each factor of the planned product when the planned product is plotted at the specified coordinates described by the above-mentioned planned product specified coordinate data 28. It is data to be. The estimated evaluation points are calculated based on the product group evaluation point data 14, the product group dimension reduction data 18, and the planned product designation coordinate data 28. The processing executed by the calculation device 4 when calculating the estimated evaluation score will be described with reference to a flowchart described later.

  Next, the planned product target evaluation point data 22 will be described. The planned product target evaluation point data 22 describes evaluation points (target evaluation points) targeted by at least one predetermined factor with respect to evaluation points that the planned product can take for each factor. The target evaluation point is set to a fixed value for a predetermined factor. For example, if the next model of the product A of the current model is to be planned, the factor is to maintain the main image of the product A (for example, the image of “luxury” in the case of FIG. 4). It is possible to set a target evaluation score for “high class”. Alternatively, if a product with an image contrasting with the product A is to be planned, the evaluation score for the factor “urban” can be set to be high as opposed to the product A. When the target evaluation point of the planned product is input to the arithmetic device 4 by operating the keyboard 6 or the mouse 8, it is stored in the storage device 12 as the planned product target evaluation point data 22.

  Next, the planned product evaluation point data 32 will be described. The planned product evaluation point data 32 describes evaluation points calculated for the factor group of the planned product when the planned product is plotted at the position specified by the planned product specifying coordinate data 28 in the perceptual map. The evaluation point of the planned product is calculated based on the planned product estimated evaluation point data 30, the planned product target evaluation point data 22, and the product group dimension reduction data 18. The processing executed by the calculation device 4 when calculating the evaluation score of the planned product will be described with reference to a flowchart described later.

  The storage device 12 further stores an evaluation point calculation program 26. When the arithmetic unit 4 executes the evaluation point calculation program 26, a process for calculating evaluation points for each factor necessary for the planned product plotted on the perceptual map to be plotted at the designated coordinates designated is executed. To do. The processing operation of the evaluation score calculation program 26 will be described with reference to a flowchart described later.

Next, processing executed by the arithmetic device 4 will be described with reference to the flowcharts of FIGS.
When the operator (product planner) executes a predetermined operation with the mouse 8 and the keyboard 6, the arithmetic device 4 reads the correspondence analysis program 24 from the storage device 12 and executes it. Thereby, the arithmetic unit 4 starts the process shown in the flowchart of FIG. Note that when the arithmetic device 4 starts the following processing, the product group evaluation score data 14 is already stored in the storage device 12.

The arithmetic device 4 first reads the product group evaluation point data 14 from the storage device 12 (step S2). Next, in step S4, the evaluation score n _ij for each factor of each product is converted into a standardized evaluation score a _ij . In this step, the evaluation point n _ij is standardized by executing a process using a chi-square statistical formula or the like. For example, the standardized evaluation point a _ij is calculated by processing the following expression. In the following formula, _ni. Is the sum of evaluation points (row elements) for each product, n _{. j} is the sum of evaluation points (column elements) for each factor, n _{. .} Is the sum of evaluation points of products and factors.

By the above-described processing, for example, the evaluation points from evaluation point n ₁₁ (corresponding to evaluation point “13” for item “urban”) of product A to n _ij are converted into standardized evaluation points a _{11 to} a _ij. Can be converted to The values a _{11 to} a _ij represented in this way can be regarded as an i × j matrix A having these values as elements. The calculated value is stored in the storage unit 12 as the product group standardized evaluation score data 16 (step S6).
In step S8, the matrix A described above is singularly decomposed into t eigenvalues of the matrix A, eigenvectors u of row elements, and eigenvectors w of column elements. By this processing, the dimension of each standardized value a _ij is reduced. For example, the singular value decomposition process can be executed by processing the following expression.

U is a normal eigenvector whose element is the eigenvector u _it for the product type corresponding to the t-dimensional eigenvalue, and Λ is a diagonal matrix whose diagonal element is the square root of the t-dimensional eigenvalue λ _t , W are normal eigenvectors orthogonal to the normal eigenvector U and having eigenvectors w _jt corresponding to the column elements, that is, the t-dimensional eigenvalues of the factors.
By the above processing, the element a _ij of the matrix A is subjected to singular value decomposition, the t-dimensional eigenvalue λ _t (see FIG. 5A), the product eigenvector u _it and the factor for the product corresponding to the t-dimensional eigenvalue. A factor eigenvector w _jt (both see FIG. 5B) is calculated. In this step, the arithmetic device 4 processes the above-described expression, whereby t eigenvalues and eigenvectors corresponding to the respective eigenvalues are calculated. Each eigenvalue represents a different characteristic of the relationship between the product and the image (factor). The calculated eigenvalues, product eigenvectors, and factor eigenvectors are stored in the storage device 12 as product group dimension reduction data 18 (step S10).
FIG. 5A illustrates eigenvalues from one dimension to four dimensions in descending order of eigenvalues. The number of eigenvalues calculated in step S4 varies depending on the number of elements of the matrix A. For example, when the matrix A is a matrix composed of 4 × 4 elements, four eigenvalues are calculated. Depending on the number of product types i and the number of factors j, the number of eigenvalues calculated differs, and the number of dimensions that can constitute the perceptual map also differs.

In step S11, an eigenvalue corresponding to a scale for creating a perceptual map is specified. In the present embodiment, the perceptual map is specified in order from the largest eigenvalue. In the case of creating a two-dimensional perceptual map as shown in FIG. 6, the largest eigenvalue (one-dimensional eigenvalue in FIG. 5A) is specified as the first component in this step, and the next largest eigenvalue. (A two-dimensional eigenvalue in FIG. 5A) is specified as the second component.
In step S12, the identified eigenvalues t, from product eigenvectors u _it and factor eigenvector w _jt corresponding to each eigenvalue, coordinate y of each product to measure the component corresponding to the eigenvalues of t dimension _it (the matrix A The value corresponding to the row item) and the coordinates x _{jt of} each factor (value corresponding to the column item of the matrix A) are calculated. The coordinates of the product and the factor are calculated by processing the following formula, for example.

The eigenvalue specified in step S11 and the coordinates calculated in step S12 are stored in the storage device 12 as perceptual map data 20 (step S14).
The eigenvectors calculated for each product and factor may be used as coordinates, but the coordinates of each product and factor are standardized so that the variance of each eigenvector element is 1 and the average is 0. Is preferably calculated.

  In step S18, the component corresponding to the eigenvalue specified in step S11 is used as a scale, and products and factors are plotted at positions corresponding to the coordinates in each component to create a perceptual map. For example, products and factors can be plotted with the first component (one dimension) corresponding to the largest eigenvalue as the horizontal axis and the second component (two dimensions) corresponding to the next largest eigenvalue as the vertical axis. . The created perceptual map is output to the display device 10 and displayed (step S18). The product and factor plot positions in the perceptual map of FIG. 6 are based on the interval scale. The position and distance between the product group and the factor group represent the relationship between each product and each factor and their strength as a positional relationship in space.

  In the perceptual map created by the series of processes described above, the positions of products with respect to the axis with the component (dimension) as a scale represent the relationship between the products, and the distance between the products represents the strength of the relationship. ing. For example, the product A and the product D in FIG. 6 have received the same evaluation on the one-dimensional and two-dimensional features. For example, when planning a product and investigating its image, if it is evaluated that the planned product has the same image as the products A and D, the planned product is close to the products A and D on the product map. If the planned product is evaluated to have an image contrasting with the products A and D, the planned product is plotted in any region of the first to third events in the graph of FIG.

The perceptual map described above can also plot factors. Also in this case, as in the case of the product, the factor coordinates are represented by an interval scale, and the position and distance between factors represent the relationship between the factors and the strength of the relationship.
For example, in the case of factors having similar relationships (for example, “simple” and “individual” in FIG. 6), it is evaluated that the closer the distance, the stronger the similarity. However, in the case of factors having contrasting relationships (for example, “urban” and “cute” in FIG. 6), the farther the distance, the stronger the relationship. It means that it is evaluated.

Using the perceptual map, the relationship between the product and the factors can be seen from the above-mentioned products or the same relationship as the factors. Products and factors whose coordinates are close to each other are evaluated similarly in the dimension represented by the axis. On the other hand, it is understood that the product and the factor plotted at the contrast positions are subjected to the contrast value evaluation. In this case, it is possible to know not only the position and distance between products or factors but also the value evaluation for products and factors plotted in the same direction (or opposite direction) with respect to the axis. Interpretation considering both products and factors becomes possible.
For example, from FIG. 6, it can be seen that the product A is highly related to the image of being “luxury”, while some of the products A to E are “urban” and “youthful” ", It can be seen that there is no product that is evaluated in connection with the image of" practical ".

  The product planner can examine the image of the product plan while examining the relationship between the product and the factor plotted on the perceptual map (product / factor map) of FIG. 6 created by the above-described processing. For example, when planning the next model of product B, if planning on the same image route as product B, the image of product B (ie, “simple”, “unique”, “youthful” It can be seen that it is sufficient to plan a product that is plotted at a position close to an image that is evaluated particularly in relation to “ Or, if you are planning a new product with a concept that was not available in the past, you should have an image that is evaluated especially in relation to “urban”, “youthful”, “practical”, etc. The image concept of the planned product can be planned while considering other products and factors, such as it is preferable to plan the product you have.

According to the product planning support apparatus 2 of the present embodiment, when the image directionality of the planned product on the perceptual map in FIG. 6 is formulated, the planned product is actually plotted on the perceptual map, and its coordinate position From the above, it is possible to calculate the evaluation score for the factor group for the planned product.
When the operator (product planner) executes a predetermined operation with the mouse 8 and the keyboard 6, the arithmetic device 4 reads the evaluation score calculation program 26 from the storage device 12 and executes it. Thereby, the arithmetic unit 4 starts the process shown in the flowchart of FIG. It is assumed that when the arithmetic device 4 starts the following processing, the processing shown in FIG. 2 is finished and the perceptual map is displayed on the display device 10.

  In step S20, the arithmetic unit 4 first acquires designated coordinates of the planned product in the perceptual map. The designated coordinates of the planned product may be input by the product planner (operator) inputting a numerical value using the keyboard 6 or by plotting the planned product on the perceptual map using the mouse 8. . The arithmetic device 4 stores the acquired designated coordinates in the storage device 12 as planned product designated coordinate data 28. FIG. 8 shows a perceptual map in which planned products are plotted according to designated coordinates.

Next, in step S22, a reference product is selected from the product group. When the reference product is selected, the arithmetic device 4 reads the evaluation score (reference product evaluation score data) for the factor group for the reference product from the product group evaluation score data 14 of the storage device 12. The reference product serving as the reference for the planned product is selected from the product group plotted on the perceptual map. At this time, the reference product may be, for example, a product plotted at a position closest to the planned product on the perception map, or may be arbitrarily designated by a product planner (operator).
The position and distance in the perceptual map of the planned product and the reference product represent the relationship between the image characteristics of the planned product and the planned product, and the strength of the relationship. When the reference product is a product that is plotted at the position closest to the planned product on the perception map (for example, product C in the case of FIG. 8), the evaluation score that the reference product has for each factor is It means that it is closest to the evaluation point required for the planned product. By selecting such a product as the reference product, the amount of change from the reference product (the amount of change in the factor eigenvector when using the component corresponding to the eigenvalue as a scale) is accurately reflected in the planned product evaluation point data. be able to.
When it is desired to plot the planned product on the designated coordinates while maintaining the evaluation equivalent to that of a specific product for the predetermined factor, it is preferable that the operator arbitrarily designates the reference product. For example, if you plan to plan a product with a more “urban” image than Product E while maintaining an image of youthfulness that is almost equivalent to Product E in FIG. Product E can be specified.

When the reference product is selected, the relationship expressed by the following formula is established between the reference product and the planned product. In the following expression, n _cj is an estimated evaluation point for factor j of the planned product, n _bj is an evaluation point for factor j of the reference product, and Σβ _t x _jt is an estimated evaluation point for the factor group for the reference product Represents the amount of change (first change).

As described above, the estimated evaluation point of the planned product can be represented by the first change amount from the reference product when the evaluation point n _bj of the reference product and the component t corresponding to the eigenvalue t are used as a scale. When representing the estimated evaluation point planning items in the type described above, the estimated evaluation point n _cj planning products are relationships that differ by the amount of the evaluation point n _bj a first variation of the reference product is established. The first change amount becomes smaller as the evaluation points of the reference product and the planned product are similar (that is, the closer the reference product and the planned product are plotted on the perceptual map).
Further, the first change amount can be expressed as a first change amount parameter with a factor coordinate corresponding to the eigenvalue associated with the perceptual map as a scale. Since the evaluation point and the factor coordinate value of the reference product are stored in the storage device 12, an assumed first change parameter is assumed (step S24), and the above-described model formula is used using the assumed first change parameter. And the first change amount can be calculated for each factor of the factor group (step S26). When the first change amount is calculated, a hypothetical evaluation point for the factor group of the planned product in the change amount is calculated (step S28).
For example, if the component (factor coordinates) corresponding to the eigenvalue associated with the perceptual map of FIG. 8 is used as a scale, the one-dimensional assumed first change parameter is obtained by the processing from step S24 to step S28 described above. β ₁ and a two-dimensional assumed first change parameter β ₂ are assumed, and the first change amount and the assumed evaluation point of the planned product are calculated when these assumed first change parameters are used. For example, when an assumed evaluation score for factor j is described, the assumed evaluation score for factor j of the planned product is expressed as n _cj = n _bj + β ₁ x _j1 + β ₂ x _j2 .
As described above, the assumed first variation parameter is assumed on the basis of the component corresponding to the eigenvalue associated with the perceptual map. From this, if two or more components are associated with the perceptual map, a hypothetical first variation parameter is assumed in addition to the above-described β ₁ and β ₂ , and from these hypothetical first variation parameters, One change amount and an assumed evaluation point of the planned product are calculated.

When the assumed evaluation score is calculated, an assumed standardized evaluation score for the planned product is calculated in step S30. The assumed standardized evaluation point a _cj of the planned product is calculated from the original evaluation point of the product group described by the product group evaluation point data 14 and the assumed evaluation point of the planned product calculated in step S28. For example, the standardized hypothesis evaluation point of the planned product can be calculated using a technique such as a chi-square test method.
In this step, for example, an assumed standardized evaluation score of the planned product can be calculated by processing the following equation. In the following equation, n _cj is a hypothetical evaluation point of the planned product in factor j, and n _c. Is the sum of the evaluation score of the product group and the assumed evaluation score of the planned product, n _{. j} is the sum of evaluation scores for each factor, and n _{. .} Is the sum of evaluation points of product group, planned product and factor item.

By executing the above processing, the assumed standardized evaluation point a _cj of the planned product is obtained by removing the effects specific to the product (row element) and the factor (column element) from the assumed evaluation point n _cj , respectively. It has been converted to a numeric value that represents the relationship.

In step S32, the assumed coordinates y _ct of the planned product on the perception map are calculated. The planned product assumption coordinate y _ct is stored as the assumed standardized evaluation score of the planned product calculated in step S30 and the product group dimension reduction data 18, and corresponds to the eigenvalue associated with the perceptual map and the eigenvalue. It can be calculated from factor eigenvectors with components as scales.
The planned product assumption coordinates y _ct are calculated by processing the following equation, for example. In the following equation, y _ct is a hypothetical coordinate of the planned product with the component corresponding to the t-dimensional eigenvalue as a scale, and u _ct is the eigenvector of the planned product with the component corresponding to the t-dimensional eigenvalue as a scale, λ _t is a t-dimensional eigenvalue, w _jt is an eigenvector of factor j having a component corresponding to the t-dimensional eigenvalue as a scale, and a _cj is an assumed standardized evaluation point of the planned product in factor j.

In the above formula, the eigenvalue and factor eigenvector (product group dimension reduction data 18) common to the planned product and the product group are used. This makes it possible to obtain assumed coordinates on the perceptual map corresponding to the evaluation points assumed for the planned product.
In step S34, it is determined whether or not the calculated assumed coordinate is the assumed coordinate closest to the designated coordinate acquired in step S20. In this step, for example, a threshold value is set for the distance between the designated coordinates and the assumed coordinates, and the determination can be made based on whether or not the distance calculated between the assumed coordinates and the designated coordinates is equal to or less than the threshold value. If it is determined that the assumed coordinate is closest to the designated coordinate (YES in step S34), the assumed evaluation point of the planned product when the assumed coordinate is taken is selected as the estimated evaluation point of the planned product, and the planned product is selected. The estimated evaluation score data 30 is stored in the storage device 12 (step S36). On the other hand, if it is determined that it is not the assumed coordinate closest to the designated coordinate (NO in step S34), the process from step S24 to step S34 is repeated assuming a different assumed first change parameter.

In step S24, assuming the component corresponding to the eigenvalue associated with the perceptual map as a scale and changing the assumed first variation parameter assumed for each, the assumed coordinates of the planned product are changed for each component on the perceptual map. Can be changed independently. For example, if the assumed first change parameter β ₁ for the component corresponding to the one-dimensional eigenvalue (first component) is changed, the assumed coordinates of the planned product are moved in the scale direction represented by the first component. Can do. Alternatively, if the assumed first change parameter β ₂ for the component corresponding to the two-dimensional eigenvalue (second component) is changed, the assumed coordinates of the planned product are moved in the scale direction represented by the second component. Can do. From this, when it is determined that the calculated assumed coordinate is not the assumed coordinate closest to the designated coordinate (NO in step S34), the assumed first change parameter assumed next is the designated coordinate and the assumed coordinate. May be determined in consideration of the positional relationship.

  For example, when planning the next model (planned product) of a product for which the current model exists, if you want to create an image concept that inherits the image of the current model, As in the case of planning an image, there are cases where the coordinate position of the planned product is specified on the perceptual map, and the image represented by the predetermined factor is specified to take a predetermined value. In such a case, the estimated evaluation score for the predetermined factor calculated by the above-described series of processing does not necessarily score a desired evaluation score. For this reason, in the following process, the process of calculating the evaluation score of the planned product is executed by reflecting the intention of the product planner in the estimated evaluation score.

  In step S38, a planned product target evaluation score is acquired for a predetermined factor. The target evaluation score is input to the arithmetic device 4 by operating the keyboard 6 and the mouse 8. A target evaluation point refers to an evaluation point that is a target for a planned product for a predetermined factor in a group of factors. The planned product target evaluation point can be expressed, for example, in a format as shown in the following equation.

In this step, as a target evaluation point of the planned product, a fixed value designated as an evaluation point for a predetermined factor may be acquired, or each evaluation point (fixed value) specified for a plurality of factors may be acquired. May be. The acquired target evaluation points are stored in the storage device 12 as planned product target evaluation point data 22.
For example, if it is desired that the product maintain a youthful image equivalent to the product E in FIG. 4, in this step, the product planner sets the target evaluation point of the planned product for the factor “youthful”. The same evaluation score “64” as the product E in FIG. 4 can be designated. By specifying the target evaluation point, the product planner can reflect his intention on the evaluation point.

In many cases, there is a discrepancy between the estimated evaluation point described by the planned product estimated evaluation point data 30 and the target evaluation point described by the planned product target evaluation point data 22 for a predetermined factor. Therefore, it is necessary to execute processing for approximating the estimated evaluation point to the target evaluation point by changing a parameter having a component corresponding to the eigenvalue as a scale. However, changing the first change amount parameter with the component corresponding to the eigenvalue associated with the perceptual map as a scale is not preferable because the coordinates of the planned product on the perceptual map change from the designated coordinates. Therefore, among the components of the factor eigenvector, the component corresponding to the eigenvalue not associated with the perceptual map, that is, the component that is contracted when generating the perceptual map and is not reflected on the perceptual map, is used as a scale. An estimated evaluation point related to a predetermined factor of the planned product is brought close to the target evaluation point using a value expressing the second change amount (second change amount parameter). Changing the second change amount parameter does not affect the coordinate position where the planned product is plotted on the perceptual map.
For example, the estimated evaluation point can be brought close to the target evaluation point by processing the following expression. In the following expression, n _bj ′ is an estimated evaluation point in the factor _j ′ for which the target evaluation point is designated, and Σβ _t x _j ′ _t is a change amount from the estimated evaluation point to the target evaluation point (second change amount). N ^fix is a target evaluation point set for the factor j ′.

Using the above model formula, the second change amount can be expressed by using a second change amount parameter having a component corresponding to the eigenvalue t not associated with the perceptual map as a scale. For this reason, in step S40, an assumed second variation parameter is assumed. In this step, hypothetical second change parameters for all eigenvalues t not associated with the perceptual map may be assumed, or eigenvalues assuming hypothetical second change parameters may be designated. For example, in the case of the eigenvalues shown in FIG. 5A, the assumed second variation parameters β ₃ and β ₄ are respectively set for eigenvalues not associated with the perceptual map, that is, for the three-dimensional eigenvalue and the four-dimensional eigenvalue. Can be assumed. As shown in FIG. 5C, the coordinates of each product and each factor can be calculated for the component corresponding to the eigenvalue t that is not associated with the perceptual map. As shown in the above formula, in this embodiment, the second change amount is calculated using the factor coordinates even when the component corresponding to the eigenvalue t that is not associated with the perceptual map is used as a scale.

If the assumed second change parameter is assumed, in step S42, the change amount (second change amount) from the estimated evaluation point for each factor to the target evaluation point is calculated. The second change amount represents the difference between the estimated evaluation point of the planned product and the target evaluation point, using the evaluation point for each factor group as a scale.
When the second change amount is calculated, the arithmetic unit 4 calculates the difference between the estimated evaluation point and the target evaluation point using the above-described model formula (step S44), and the calculated estimated evaluation point and the target evaluation point are calculated. It is determined whether or not the difference is minimum (step S46).
If it is determined that the difference between the estimated evaluation score and the target evaluation score is not minimum (NO in step S46), the processing from step S40 to step S46 is repeated assuming a different assumed second variation parameter. On the other hand, if it is determined that the difference between the estimated evaluation score and the target evaluation score is the smallest (YES in step S46), the second change when the difference between the estimated evaluation score and the target evaluation score becomes that value. An amount (that is, a value obtained by multiplying each factor coordinate corresponding to the three-dimensional and four-dimensional eigenvalues by the amount of the second variation parameter) is added to the estimated evaluation point to calculate the evaluation point of the planned product (step S48). For example, the evaluation point n _Cj of the planned product is expressed by n _Cj = n _cj + β ₃ x _j3 + β ₄ x _j4 . The estimated evaluation point n _cj already includes a change in evaluation point based on a difference in position on the perceptual map, which is represented by the first change amount parameters β ₁ and β ₂ . By adding the second variation calculated using the second variation parameters β ₃ and β ₄ to the estimated evaluation point, the target evaluation is performed while maintaining the position of the planned product on the perceptual map at the designated coordinates. An evaluation point closer to the point can be calculated. The calculated evaluation score is stored in the storage device 12 as planned product evaluation score data 32.
FIG. 10 is a diagram showing the calculated evaluation score of the planned product. For the factor “youthful”, when the target evaluation score is set to the same evaluation score “64” as that for the product E, the planned product is “urbanized” in order to be plotted at the position specified in the perception map. “It is necessary to receive a particularly high evaluation for the two factors“ practical ”, and at the same time, it is necessary to design the image so that the image for the factor“ simple ”can have a slightly higher evaluation. Understood. The product planner can examine the image concept of the planned product based on the obtained planned product evaluation point.

  The planned product evaluation score data 32 calculated by the above-described series of processing includes information on the characteristics of the component corresponding to the eigenvalue not associated with the perceptual map (not reflected in the perceptual map). According to the product planning support device 2, the arithmetic device 4 calculates the evaluation score for the factor group for the planned product from the designated coordinates of the planned product plotted on the perceptual map, and outputs it to the storage device 12, the display device 10, etc. Can do.

Specific examples of the present invention have been described in detail above, but these are merely examples and do not limit the scope of the claims. The technology described in the claims includes various modifications and changes of the specific examples illustrated above.
For example, in this embodiment, the first change amount and the second change amount are calculated by assuming the first change amount parameter and the second change amount parameter, respectively, but the first change amount is calculated and associated with the perceptual map. Calculating a first variation parameter with a component corresponding to the determined eigenvalue as a scale, and calculating a second variation parameter with a second variation parameter as a scale corresponding to the component corresponding to the eigenvalue not associated with the perceptual map. You may calculate.
The target evaluation point described by the planned product target evaluation point data 22 may be a constraint condition that restricts predetermined factor items of the planned product. For example, (1) a constraint condition that an evaluation score for a predetermined factor of a planned product is fixed to an arbitrary fixed value (real number), and (2) an evaluation score for a predetermined factor of the planned product is stored in the product group. You may prescribe | regulate the constraint condition fixed to the same value as the evaluation score of the predetermined goods included, and you may memorize | store it in the memory | storage device 12. FIG.
In step S22 of FIG. 3, the reference product is selected from the product group plotted on the perceptual map. However, in this step, the reference product may be selected based on the origin of the perceptual map.
Further, the product planning support device 2 calculates the coordinates of the planned product from the evaluation points of the planned product calculated in step S48 of FIG. 3, and executes processing for verifying the accuracy of the evaluation points by comparing with the designated coordinates. Also good.

  The technical elements described in this specification or the drawings exhibit technical usefulness alone or in various combinations, and are not limited to the combinations described in the claims at the time of filing. In addition, the technology illustrated in the present specification or the drawings achieves a plurality of objects at the same time, and has technical utility by achieving one of the objects.

It is a figure showing the outline | summary structure of a product plan assistance apparatus. It is a flowchart showing the process which produces a perceptual map. It is a flowchart showing the calculation process of the evaluation point about the factor group with respect to the plan goods which designated the coordinate in the perception map. It is a figure showing the information which merchandise group evaluation score data describes. It is a figure showing the information which the merchandise group dimensional reduction data and perception map data describe. It is a figure showing the perceptual map created from the perceptual map data. It is a figure which shows the designated coordinate of the planned product which the planned product designated coordinate data describes. It is a figure showing the perceptual map by which the plan goods were plotted according to the designated coordinate. It is a figure showing the evaluation score of the planned product described by the planned product evaluation score data.

Explanation of symbols

2: Product planning support device 4: Computing device 6: Keyboard 8: Mouse 10: Display device 12: Storage device 14: Product group evaluation point data 16: Product group standardized evaluation point data 18: Product group dimension reduction data 20: Perception Map data 22: Planned product target evaluation point data 24: Correspondence analysis program 26: Evaluation point calculation program 28: Planned product designation coordinate data 30: Planned product estimated evaluation point data 32: Planned product evaluation point data

Claims (5)

In the perceptual map obtained as a result of the perceptual map creation method based on the evaluation score for the factor group for the product group, the evaluation score for the factor group of the planned product corresponding to the coordinates specified on the perceptual map is calculated. A device,
Original evaluation point data describing evaluation points for the factor group for the product group, eigenvalue data describing eigenvalues obtained by performing a perceptual map creation method based on the original evaluation point data, and the eigenvalues Product eigenvector data describing eigenvectors for the corresponding product group, factor eigenvector data describing eigenvectors for the factor group corresponding to the eigenvalues, and product coordinates describing coordinates in the perceptual map for each of the product groups Means for storing data and factor coordinate data describing coordinates in a perceptual map for each of said factor groups;
Means for acquiring planned product designated coordinate data describing coordinates (designated coordinates) designated in the perception map of the planned product;
Means for selecting a reference product (reference product) from the product group, extracting an evaluation score for a factor group for the reference product from the original evaluation score data, and outputting as reference product evaluation score data;
Based on the planned product designation coordinate data, the standard product evaluation point data, the original evaluation point data, the eigenvalue data, and the factor eigenvector data, an estimated evaluation point (estimated evaluation point) for the factor group for the planned product, The component corresponding to the eigenvalue associated with the perceptual map among the components of the factor eigenvector is the change amount (first variation) from the evaluation point for the factor group for the reference product to the estimated evaluation point for the factor group for the planned product. Means for calculating a value (first variation parameter) expressed as a scale, and outputting each as planned product estimated evaluation point data and first variation parameter data;
Means for acquiring planned product target evaluation point data describing a target evaluation point (target evaluation point) for a planned product for a predetermined factor in the factor group;
Based on the planned product estimated evaluation point data, the planned product target evaluation point data, and the factor-specific vector data, the change amount from the estimated evaluation point for the planned product to the target evaluation point for the planned product (second change amount) for the predetermined factor ) For calculating a value (second variation parameter) expressing the component corresponding to the eigenvalue not associated with the perceptual map among the components of the factor eigenvector, and outputting it as second variation parameter data; ,
For each factor of the factor group, add the value obtained by multiplying the estimated evaluation point for the planned product by each component corresponding to the eigenvalue that is not associated with the perceptual map among the components of the factor eigenvector, An apparatus provided with means for calculating evaluation points for factor groups for planned products and outputting them as planned product evaluation point data.   The means for calculating the second variation parameter is a component corresponding to an eigenvalue that is not associated with the perceptual map among the components of the factor eigenvector at the estimated evaluation point for the planned product for the predetermined factor. 2. The apparatus according to claim 1, wherein the second change amount parameter is calculated so that a difference between an evaluation point obtained by adding a value multiplied by and a target evaluation point for the planned product is minimized. The means for calculating the estimated evaluation point and the first change amount parameter of the planned product are:
Means for generating hypothesized first variation parameter data describing a hypothesized first variation parameter (assumed first variation parameter);
Based on the standard product evaluation point data, the assumed first variation parameter data, and the factor eigenvector data, an evaluation point (assumed evaluation point) assumed for the factor group for the planned product is calculated, and the planned product assumed evaluation point data As a means to output as
Based on the original evaluation point data and the planned product assumption evaluation point data, from the evaluation points for the product group and the factor group for the planned product, a hypothetical standardized evaluation point that standardizes the assumed evaluation point for the factor group for the planned product is calculated. , Means for outputting as planned product assumption standardized evaluation point data,
Means for calculating the assumed coordinates (assumed coordinates) of the planned product in the perceptual map based on the eigenvalue data, the factor eigenvector data, and the planned product assumption standardized evaluation point data, and outputting the calculated product assumed coordinate data;
Based on the assumed first change parameter data, the planned product assumed coordinate data, the planned product assumed coordinate data, and the planned product designated coordinate data, the assumed coordinate of the planned product closest to the designated coordinate of the planned product in the perceptual map is set. Corresponding, a hypothetical first change parameter and a hypothetical evaluation point for the factor group for the planned product are provided as means for selecting a first change parameter and an estimated evaluation point for the factor group for the planned product, respectively. The apparatus according to claim 1 or 2. In the perceptual map obtained as a result of the perceptual map creation method based on the evaluation score for the factor group for the product group, the evaluation score for the factor group of the planned product corresponding to the coordinates specified on the perceptual map is calculated. A method,
Original evaluation point data describing evaluation points for the factor group for the product group, eigenvalue data describing eigenvalues obtained by performing a perceptual map creation method based on the original evaluation point data, and the eigenvalues Product eigenvector data describing eigenvectors for the corresponding product group, factor eigenvector data describing eigenvectors for the factor group corresponding to the eigenvalues, and product coordinates describing coordinates in the perceptual map for each of the product groups Storing data and factor coordinate data describing coordinates in the perceptual map for each of the factor groups;
Obtaining planned product designation coordinate data describing coordinates (designated coordinates) designated in the perception map of the planned product;
Selecting a reference product (reference product) from the product group, extracting the evaluation score for the factor group for the reference product from the original evaluation point data, and outputting as reference product evaluation point data;
Based on the planned product designation coordinate data, the standard product evaluation point data, the original evaluation point data, the eigenvalue data, and the factor eigenvector data, an estimated evaluation point (estimated evaluation point) for the factor group for the planned product, The component corresponding to the eigenvalue associated with the perceptual map among the components of the factor eigenvector is the change amount (first variation) from the evaluation point for the factor group for the reference product to the estimated evaluation point for the factor group for the planned product. Calculating a value (first variation parameter) expressed as a scale, and outputting each as planned product estimated evaluation point data and first variation parameter data;
Obtaining planned product target evaluation point data describing a target evaluation point (target evaluation point) for the planned product for the predetermined factor in the factor group;
Based on the planned product estimated evaluation point data, the planned product target evaluation point data, and the factor-specific vector data, the change amount from the estimated evaluation point for the planned product to the target evaluation point for the planned product (second change amount) for the predetermined factor ) To calculate a value (second variation parameter) expressing the component corresponding to the eigenvalue that is not associated with the perceptual map among the components of the factor eigenvector, and outputting it as second variation parameter data; ,
For each factor of the factor group, add the value obtained by multiplying the estimated evaluation point for the planned product by each component corresponding to the eigenvalue that is not associated with the perceptual map among the components of the factor eigenvector, A method comprising a step of calculating an evaluation score for a group of factors for a planned product and outputting it as planned product evaluation score data. In the perceptual map obtained as a result of the perceptual map creation method based on the evaluation score for the factor group for the product group, the evaluation score for the factor group of the planned product corresponding to the coordinates specified on the perceptual map is calculated. A program for
Original evaluation point data describing evaluation points for the factor group for the product group, eigenvalue data describing eigenvalues obtained by performing a perceptual map creation method based on the original evaluation point data, and the eigenvalues Product eigenvector data describing eigenvectors for the corresponding product group, factor eigenvector data describing eigenvectors for the factor group corresponding to the eigenvalues, and product coordinates describing coordinates in the perceptual map for each of the product groups Storing data and factor coordinate data describing coordinates in the perceptual map for each of the factor groups;
Processing for acquiring planned product designation coordinate data describing coordinates (designated coordinates) designated in the perception map of the planned product;
A process of selecting a reference product (reference product) from the product group, extracting an evaluation score for a factor group for the reference product from the original evaluation score data, and outputting as a reference product evaluation score data;
Based on the planned product designation coordinate data, the standard product evaluation point data, the original evaluation point data, the eigenvalue data, and the factor eigenvector data, an estimated evaluation point (estimated evaluation point) for the factor group for the planned product, The component corresponding to the eigenvalue associated with the perceptual map among the components of the factor eigenvector is the change amount (first variation) from the evaluation point for the factor group for the reference product to the estimated evaluation point for the factor group for the planned product. A process of calculating a value (first variation parameter) expressed as a scale, and outputting each as planned product estimated evaluation point data and first variation parameter data;
Processing for obtaining planned product target evaluation point data describing a target evaluation point (target evaluation point) for the planned product for a predetermined factor in the factor group;
Based on the planned product estimated evaluation point data, the planned product target evaluation point data, and the factor-specific vector data, the change amount from the estimated evaluation point for the planned product to the target evaluation point for the planned product (second change amount) for the predetermined factor ) To calculate a value (second variation parameter) expressing the component corresponding to the eigenvalue that is not associated with the perceptual map among the components of the factor eigenvector, and output it as second variation parameter data; ,
For each factor of the factor group, add the value obtained by multiplying the estimated evaluation point for the planned product by each component corresponding to the eigenvalue that is not associated with the perceptual map among the components of the factor eigenvector, A program that causes a computer to execute a process of calculating evaluation points for factor groups for a planned product and outputting the evaluation point data as planned product evaluation point data.

Images