JP4254387B2 - Hierarchical evaluation value measuring method, its supporting device and program - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a technique for measuring the high value that a purchaser recognizes for the specification of a product such as an automobile.
[0002]
[Prior art]
Companies provide profits that meet the needs of buyers by providing products with specifications that buyers such as consumers need.
The values of buyers are diversifying and companies are struggling to understand the product specifications that many buyers need. For products such as automobiles, which have a mixture of practicality and hobbies, it is difficult to accurately know the specifications of the products that many buyers need, and products that were introduced with high expectations May end up disappointing, or products that did not expect much may be sold more than expected.
If the specifications required by many purchasers can be accurately grasped in advance, it is possible to reduce the development cost of useless products and provide the products required by purchasers at a low price. A technology for accurately grasping in advance the specifications of a product required by the purchaser is required not only by the company but also by the purchaser.
[0003]
Methods have been developed to investigate product specifications that many buyers will choose. One of them is conjoint analysis. In the conjoint analysis, the product attribute that the purchaser selects as a product selection factor and an evaluation value (also referred to as a partial utility coefficient) that the purchaser recognizes for the level are measured. The details of the conjoint analysis are described in, for example, “Seven Product Planning Tools, Chapter 9, Editor: Noriaki Kanda, Nikka Giren Publishing Co., Ltd.”.
For example, in the case of a product called an automobile, the purchaser selects a purchase vehicle type in consideration of factors such as manufacturer, displacement, fuel consumption, and price. Therefore, the manufacturer, displacement, fuel consumption, price, and the like are attributes.
The level in the conjoint analysis indicates what the attribute is specifically for each attribute. For example, for the attribute of displacement, a level of 1.2 to 1.4 liters, There is a level of 1.4 to 1.6 liters.
The purchaser determines whether the level of 1.2 to 1.4 liters is preferable for the attribute of displacement or whether the level of 1.4 to 1.6 liters is preferable. Judge whether the company level is preferable or whether the level of company II is preferable, and for the attribute of price, for example, determine whether the level of 900 to 1,200,000 yen is preferable, or whether the level of 1,200 to 1,400,000 yen is preferable Then, after evaluating them comprehensively, one car is selected.
[0004]
The strength of the purchaser's product selection process is not constant, and many consumers select products with an emphasis on the level of 1.4 to 1.6 liters. While the level of 4 liters is not preferred, it may not be important whether the manufacturer is Company I or Company II. In this case, the level of 1.4 to 1.6 liters is a level that works strongly and favorably when selecting a product, and the level of 1.2 to 1.4 liters is a level that acts strongly and disadvantageously when selecting a product. It can be said that the level of being made and the level of being manufactured by II company is a neutral level. The level of importance at which a level functions when selecting a product can be quantified, for example, a positive level can be expressed as a positive level, a zero level as a neutral level, and a negative level as a negative level. In this case, a large absolute value is given according to the degree of strong work. The numerical standard may be arbitrarily set, and the level that works most disadvantageously may be zero, and may have a large positive value in the order of neutral level, advantageous level, and strong advantageous level. In the present specification, a value obtained by quantifying the importance of a level when selecting a product is referred to as an evaluation value or a partial utility coefficient. It can be said that the attribute in which the evaluation value or the partial utility coefficient varies greatly depending on the level to which it belongs, that is, the attribute of displacement in the previous example, is an important attribute when selecting a product. It can be said that the attribute whose evaluation value or partial utility coefficient does not vary greatly depending on the level to which it belongs, that is, the attribute “maker” in the previous example, is an attribute that is not important in selecting a product. The importance of attributes at the time of product selection can also be quantified. Importance can also be indicated in% Here, a value obtained by quantifying the importance of an attribute at the time of product selection is referred to as a contribution rate. In the previous example, the contribution rate of the attribute of displacement is high, and the contribution rate of the attribute of manufacturer is low.
[0005]
Various methods have been developed for questionnaire surveys (contract surveys of buyers) for conjoint analysis. Typically, a virtual product group in which attributes and levels are variously combined is created, and combinations of attributes and levels of each product are shown on the card. Prepare one card for one product. Then, ask the surveyees to give me the card group and rank the cards according to the order of the items they want to select. When multiple survey subjects are ranked, and the results are entered, the evaluation values for the attributes and levels that the survey subjects recognize (sometimes simply called evaluation values or partial utility coefficients of the levels) are measured. There are programs on the market. In order to measure the evaluation values that the purchaser accepts for the attributes and levels that exist in many types with as few cards as possible, the orthogonal table method used in the experimental design method is adopted.
A paired comparison method may be used instead of the card method. In the paired comparison method, a plurality of pairs of virtual products are assumed. A pair of virtual merchandise is shown to the survey target person, and the merchandise with the strong demand to be selected is selected. This is repeated for a plurality of combinations. The paired comparison method can be executed at low cost in a short time using the Internet. There are commercially available programs that use computers to measure the evaluation values for the attributes and levels recognized by the survey subjects from the survey results. Details of the questionnaire item setting method, questionnaire implementation method, and commercial program for conjoint analysis are explained in the above-mentioned “Seven Product Planning Tools, Chapter 9.4 Items”.
[0006]
[Problems to be solved by the invention]
There are many types of attributes that buyers consider when selecting products. For example, in the case of automobiles, the purchaser is the manufacturer, displacement, fuel consumption, price, vehicle size, number of passengers, body type, drive system, navigation system, aero equipment, audio equipment, interior specifications, seat fabric, door mirror type, lamp type Determine the model to be purchased after considering the above.
If the number of attributes increases, the number of cards and the number of questions used for conjoint analysis will increase, making it impossible to investigate in practice. In addition, the burden on the survey subject becomes excessive.
In addition, survey subjects are required to consider levels related to basic attributes of products such as displacement and vehicle size, and levels related to detailed attributes such as specifications of door mirrors and lamps. There is also a problem in that it makes it difficult to make the right choice because the survey is uncomfortable. At present, evaluation values and contribution rates are measured by handling each attribute from the basic specifications of the product such as manufacturer, displacement, vehicle size, etc. to detailed attributes such as specifications of door mirrors and lamps, The reliability of the evaluation value for the level of the detailed attribute is low. “Vehicles with 1.2-1.4 liters, manufacturer I, seat fabric jacquard moquette, and price of 1.1 million yen” and “displacement 1.4-1.6 liters, manufacturer II If you choose to select a vehicle model with a fabric price of 1,000,000 yen, you will be buried in the difference in basic attributes, and for the seat fabric belonging to the detailed attributes, the valuation and fabric of the level that is a jacquard moquette In many cases, the selection that reflects “a certain level of evaluation value” is not made.
[0007]
In the current conjoint analysis, basic and detailed attributes of products cannot be handled separately, making it difficult to conduct accurate surveys. Also, depending on the attribute, the survey result may not be reliable.
The present invention realizes a measurement technique that can handle the basic specification and the detailed specification of a product separately, and can make a survey by first selecting roughly according to the basic specification and then selecting further according to the detailed specification. According to this technology, survey subjects can respond with no sense of incongruity and can conduct accurate surveys. Further, the present invention realizes a technique for obtaining an effective evaluation value across attribute groups from separately conducted surveys. In order to make it possible to compare and evaluate the evaluation value of the attribute belonging to the basic specification of the product and the evaluation value of the attribute belonging to the detailed specification, a technique for converting (normalizing) them to the same scale is realized.
[0008]
[Means and Actions for Solving the Problems]
The present invention considers the level related to the basic attributes of the product such as the displacement and the vehicle size when the purchaser selects the product, such as an automobile, and the level related to the detailed attributes such as the specifications of the door mirror and the lamps. Further, the present invention relates to a method of measuring an evaluation value that a purchaser recognizes for a product attribute and its level for a product selected in consideration of the above.
The measurement method created by the present invention is characterized in that an attribute group is divided into a plurality of layers and analyzed. In this method, a product attribute that is selected by a purchaser as a product selection factor by referring to a database that stores in advance whether each product attribute belongs to a basic attribute, an intermediate attribute, or a detailed attribute by a computer. An attribute classification processing program for classifying groups into basic attributes, intermediate attributes, and detailed attributes is executed. Then, as a first step, a computer is used for a survey result obtained by setting a level group for each attribute of the attribute group belonging to the basic attribute and the intermediate attribute and conducting a buyer's value survey. The basic attribute and intermediate attribute evaluation values obtained by executing the conjoint analysis program are stored in the basic / intermediate attribute evaluation value database. As a second step, a conjoint analysis is performed by the computer on the survey results obtained by conducting a survey of buyers' values by setting a level group for each attribute group belonging to intermediate attributes and detailed attributes. A step of storing the intermediate attribute value and the detailed attribute evaluation value obtained by executing the program in the intermediate / detailed attribute evaluation value database is performed. The evaluation value for each intermediate attribute level stored in the basic / intermediate attribute evaluation value database in the first step and the intermediate attribute level stored in the intermediate / detailed attribute evaluation value database in the second step. The evaluation values stored in the basic / intermediate attribute evaluation value database and the evaluation values stored in the intermediate / detail attribute evaluation value database are normalized so that the evaluation values are equal to each other. The attribute evaluation values are stored in the basic / intermediate / detailed attribute evaluation value database for storing them in an integrated manner.
When normalizing here, the level of the evaluation value measured in the second step is matched with the level of the evaluation value measured in the first step, and the level of the evaluation value measured in the first step is set in the second step. Or matching the evaluation value measured in the first step and the evaluation value measured in the second step to the reference level.
[0009]
According to the above method, the product attributes include, for example, attributes related to the basic specifications of the product such as displacement and vehicle size, attributes related to the detailed specifications of the product such as specifications of door mirrors and lamps, and the presence / absence of navigation devices and aero equipment, for example. It is divided into attributes related to intermediate specifications such as the specifications. The basic attribute refers to an attribute related to the skeleton of the product such as the displacement and the vehicle size, for example, and an attribute of the nature that the price of the product changes greatly if the level is different. The detailed attribute refers to an item related to details of a product such as a specification of a door mirror or a lamp, for example, and refers to an attribute of a property that the product price does not change much even if the level is different. The intermediate attribute refers to an intermediate attribute between a product skeleton and a detail skeleton, and an attribute with a characteristic that the price of the product slightly changes when the level is different.
[0010]
In the above method, as the first step, for the survey results obtained by setting the level group for each attribute with respect to the attribute group belonging to the basic attribute and the intermediate attribute, A step of storing the basic attribute and intermediate attribute evaluation values obtained by executing the conjoint analysis program by the computer in the basic / intermediate attribute evaluation value database is performed. In the case of automobiles, the basic attribute is how much the displacement is, how much the vehicle size is, and whether there is a navigation device that is an intermediate attribute (if any, what its specifications are) Assume a product group that variously combines the presence or absence of aero equipment (if any, the specifications are). Then, a survey of buyers' values is conducted for the assumed product group. By using the existing conjoint analysis technique, it is possible to measure the evaluation value (partial utility coefficient) that the purchaser recognizes for the attribute and level belonging to the basic attribute and the intermediate attribute. In the case of the example, the basic attribute is an evaluation value having a displacement of 1.2 to 1.4 liters, an evaluation value having a displacement of 1.4 to 1.6 liters, and a basic attribute. An evaluation value of a level that a certain vehicle size is a B segment, an evaluation value of a level that the vehicle size is a C segment, an evaluation value that has a level that no navigation device is an intermediate attribute, and a navigation device that is an intermediate attribute It is possible to measure an evaluation value or the like possessed by the level that is a DVD voice navigation device.
At this time, the purchaser has only to answer questions regarding basic attributes and intermediate attributes, and since this is done on a daily basis in the first stage of product selection, there is no sense of incongruity and the actual product selection is the same. Can answer. As a result, for the attribute group belonging to the basic attribute and the intermediate attribute and the level group related to each attribute, the evaluation value that the purchaser recognizes for the attribute and the level is relatively accurately measured.
[0011]
In this method, in addition to the first step described above, as a second step, a level group is set for each attribute group belonging to the intermediate attribute and the detailed attribute, and a purchaser's values survey is performed. A step of storing the intermediate attribute value and the detailed attribute evaluation value obtained by executing the conjoint analysis program by the computer in the intermediate / detailed attribute evaluation value database is performed on the obtained survey result. In the case of an automobile product, the presence or absence of a navigation device that is an intermediate attribute (if any, its specifications), and the presence or absence of aero equipment (if any, which specifications are) Assume a product group with various combinations of headlight specifications (specifications such as whether it is a discharge lamp or a halogen lamp) or seat skin specifications (fabric, moquette, distinction of main skin, etc.), which are detailed attributes. Then, a survey of buyers' values is conducted for the assumed product group. As a result, in the case of the example, the evaluation value that the level that there is no navigation device that is an intermediate attribute, the evaluation value that the level that the navigation device is a DVD voice navigation device, and the headlight that is a detailed attribute is a discharge lamp. It is possible to measure an evaluation value possessed by the level of, or an evaluation value possessed by the level that the sheet skin that is the detailed attribute is a moquette.
At this time, the purchaser has only to answer questions regarding the intermediate attribute and the detailed attribute, and since this is done on a daily basis in the second stage of product selection, it is the same as when actually selecting the product. Can answer. As a result, for the attribute group belonging to the intermediate attribute and the detailed attribute and the level group related to each attribute, the evaluation value that the purchaser recognizes for the attribute and the level is relatively accurately measured.
[0012]
The evaluation value group measured for the level group related to each attribute of the attribute group belonging to the basic attribute and the intermediate attribute and the evaluation value group measured for the level group related to each attribute of the attribute group belonging to the intermediate attribute and the detailed attribute are different from each other. It has a level, so there is no point in displaying the list as it is.
Therefore, in this method, the level is adjusted (normalized) using the evaluation value group measured for the level group related to the intermediate attribute included in both evaluation value groups. For example, in the evaluation value group related to the basic attribute and the intermediate attribute, the evaluation value of the level that there is no navigation device is −2, the evaluation value that the level that the navigation device is a DVD voice navigation device is +4, and the intermediate value In the evaluation value group regarding the attribute and the detailed attribute, if the evaluation value of the level that there is no navigation device is −20 and the evaluation value of the level that the navigation device is a DVD voice navigation device is +40, intermediate attributes and Since the evaluation value group related to the detailed attribute is measured with a size 10 times larger than the evaluation value group related to the basic attribute and the intermediate attribute, normalization is performed by either of the following.
(1) The evaluation value group related to the intermediate attribute and the detailed attribute is multiplied by 1/10 to match the level with the evaluation value group related to the basic attribute and the intermediate attribute.
(2) The evaluation value group related to the basic attribute and the intermediate attribute is multiplied by 10 to match the level with the evaluation value group related to the intermediate attribute and the detailed attribute.
(3) The evaluation value group related to the basic attribute and the intermediate attribute is multiplied by k (constant) × 10, and the evaluation value group related to the intermediate attribute and the detailed attribute is multiplied by k (constant) to match the levels of both.
By performing the normalization process described above, the evaluation value group possessed by the level group relating to the basic attribute, the intermediate attribute, and the detailed attribute can be displayed at a unified level and can be compared and contrasted. In this case, it is possible to accurately grasp the evaluation value groups of basic attributes and detailed attributes having different personalities.
[0013]
  In order to carry out the method described above, a device that makes it easy is required. The apparatus of the present invention was created to support the process of measuring the evaluation value that the purchaser recognizes with respect to the product attribute that the purchaser makes as a product selection factor and its level. This support deviceExecute the above described database and the above processingmeansTheI have.
[0014]
Using the above devices, the results of surveys conducted separately for basic attributes and detailed attributes for which reliable evaluation values are not measured when surveyed in a lump sum, based on the relative evaluation values for the level of basic attributes, It becomes possible to accurately measure the relative evaluation value regarding the level.
Then, using the relative evaluation value for the intermediate attribute that is commonly included in both surveys, the level of the relative evaluation value for the basic attribute and the relative evaluation value for the detailed attribute is adjusted to Can be on the same scale.
With this device, when a purchaser selects a product, such as an automobile, for example, the level related to the basic attributes of the product, such as displacement and vehicle size, is considered, and details of the product, such as specifications of door mirrors and lamps, for example, are taken into account. With respect to a product to be selected in consideration of the level related to the attribute, it is possible to accurately measure the evaluation value that the purchaser recognizes for the level from the basic attribute to the detailed attribute.
In addition, from the survey results obtained by conducting a survey of buyers' values by setting a level group for each attribute for multiple attributes, the means of measuring the evaluation value that the purchaser recognizes for the attributes and levels are already existing. Means can be used. This can be realized by installing a commercially available program in the computer.
[0015]
  The present invention is a program.IngredientsRealizedMay be.
[0016]
Using the above program, the relative evaluation value and the level of the detailed attribute regarding the level of the basic attribute are obtained from the survey results of the basic attribute and the detailed attribute for which the reliable evaluation value is not measured when surveyed collectively. It is possible to accurately measure the relative evaluation value for each.
Then, using the relative evaluation value for the intermediate attribute that is commonly included in both measurement results, the level of the relative evaluation value for the basic attribute and the relative evaluation value for the detailed attribute are adjusted to Can be on the same scale.
Using this program, when a purchaser selects a product, it considers the level related to the basic attributes of the product such as the displacement and vehicle size, and also considers the level related to detailed attributes such as the specifications of the door mirrors and lamps. From the basic attribute to the detailed attribute, it is possible to accurately measure the evaluation value that the purchaser accepts for that level.
[0017]
The hierarchized measurement method created by the present invention is suitably used for measuring the automobile attribute that the purchaser makes as a vehicle selection factor and the evaluation value that the purchaser recognizes for the level. In this method, by referring to a database that stores in advance whether each car attribute belongs to a basic attribute, an intermediate attribute, or an equipment attribute by a computer, the car attribute group is classified into a basic attribute and an intermediate attribute. Execute an attribute classification program for classifying equipment attributes. Then, as a first step, a computer is used for a survey result obtained by setting a level group for each attribute of the attribute group belonging to the basic attribute and the intermediate attribute and conducting a buyer's value survey. The basic attribute and intermediate attribute evaluation values obtained by executing the conjoint analysis program are stored in the basic / intermediate attribute evaluation value database. As a second step, a conjoint analysis is performed by a computer on the survey results obtained by setting a level group for each attribute group belonging to the intermediate attribute and equipment attribute and conducting a survey of buyers' values. A step of storing the intermediate attribute and equipment attribute evaluation values obtained by executing the program in the intermediate / detail attribute evaluation value database is performed. The evaluation value for each intermediate attribute level stored in the basic / intermediate attribute evaluation value database in the first step and the intermediate attribute level stored in the intermediate / equipment attribute evaluation value database in the second step. The evaluation value stored in the basic / intermediate attribute evaluation value database and the evaluation value stored in the intermediate / equipment attribute evaluation value database are normalized so that the evaluation value becomes equal to the basic / intermediate / equipment evaluation value database. The attribute evaluation values are stored in the basic / intermediate / equipment attribute evaluation value database that is stored in an integrated manner.
[0018]
This method is a method in which the target product is limited to an automobile, and the evaluation value that the purchaser recognizes in various specifications of the automobile is measured.
Here, the measurement is divided into basic attributes of the automobile such as displacement and vehicle size and equipment attributes such as specifications of door mirrors and lamps. According to this measurement method, the relative evaluation value related to the basic attribute and the relative evaluation value related to the equipment attribute are measured based on the measurement results of the basic attribute and the measurement related to the equipment attribute that cannot be measured reliably when measured together. Can be measured accurately.
Then, using the relative evaluation value for the intermediate attribute that is commonly included in both measurement results, the level of the relative evaluation value for the basic attribute and the relative evaluation value for the equipment attribute is adjusted to both levels. Can be on the same scale.
By using this measurement method, it is possible to accurately measure the evaluation value that the purchaser recognizes for the level from the basic attribute of the automobile to the equipment attribute.
[0019]
The present invention can also be embodied in a purchaser grouping method. In this method, a program for grouping evaluation value distributions stored for each purchaser in a basic / intermediate / detail attribute evaluation value database by a computer to group purchasers, and the process Storing the relationship between the purchaser and the group obtained in the above in a group information database.
According to this method, the purchaser group is grouped with reference to the evaluation values of the basic / intermediate / detailed attributes. Therefore, the purchaser group is included in a group having the same attribute and level as important at the time of purchase, that is, a group having a common preference. Can be classified.
[0020]
The present invention can also be embodied in a method for supporting the process of determining the attribute level of a planned product. In this method, the sales volume index calculation program is executed by a computer for each group grouped by the grouping method described above with reference to a database storing the attributes and levels of the planned products, and the sales volume of the planned products. The third step for obtaining the index, the fourth step for storing the sales volume index of the planned product for each group obtained in the third step in the sales volume index database, and the planned product for each group stored in the sales volume index database By referring to the sales volume index, the attribute level of the planned product is corrected by the computer so as to improve the sales volume index of the group whose sales volume index does not meet the predetermined standard (for example, increase the sales volume or the sales share). 5th process and 6th process which memorize | stores the level corrected in 5th process in the database which memorize | stores the attribute of the said planned goods, and its level. To have.
[0021]
“Sales volume index” refers to an index indicating sales volume in some form, and includes the sales volume itself. The “planned product sales volume index” includes the ratio of the planned product sales volume to the total sales volume of the planned product and other products (the sales share of the planned product).
According to this method, it is possible to determine the attribute level of the planned product more appropriately as compared with the case where the attribute level of the planned product is determined by referring to only the sales volume index for all the groups.
[0022]
DETAILED DESCRIPTION OF THE INVENTION
Preferred embodiments of the present invention will be listed.
(Form 1) A database of attributes and levels is prepared for a known automobile. Cars are classified according to vehicle type / grade / option. For example, it is memorized that the attribute D of the automobile of the option C of the clade B of the vehicle type A is level E and the attribute F is level G.
The vehicle type market for which the evaluation value (partial utility coefficient) recognized by the purchaser is to be measured is designated by the vehicle type name stored in the database. Multiple vehicle types can be specified.
When multiple car model names are specified, the level group is organized for each attribute using the database. For example, when the vehicle types A1 and A2 are designated, the level E1 (vehicle type A1) and the level E2 (vehicle type A2) exist for the attribute D. At this time, even if the name of the same model is different depending on the clade and options, the level will be arranged in consideration of that. For example, if the vehicle type A1 has clades B1 and B2, the level of attribute F of grade B1 is G1, and the level of attribute F of grade B2 is G2, the level G1 (grade B1) and level G2 with respect to attribute F Arrange that (grade B2) exists.
As described above, the attribute and level distribution table for preparing cards used for conjoint analysis is arranged. A program for this is available.
(Mode 2) The attribute and level distribution table prepared in mode 1 is divided into a level distribution table for basic attributes and intermediate attributes, and a level distribution table for intermediate attributes and detailed attributes. A program for this is available.
(Form 3) The measurer can correct the attribute and level distribution table arranged in form 2. For example, it is possible to set a level that does not exist or to delete an attribute or level that seems unnecessary. A program for this is available.
(Embodiment 4) A conjoint analysis is performed using each of the two types of classification tables of Embodiment 2 or 3.
(Form 5) For each survey subject, first, a value survey is performed based on the level distribution regarding basic attributes and intermediate attributes, and then a value survey is performed based on the level distribution regarding intermediate attributes and detailed attributes. .
(Form 6) The purchaser's values survey is performed using the Internet.
(Form 7) The purchaser's values survey is based on a paired comparison method. The paired comparison method mentioned here is one described in “Item 9 of Chapter 9 of Product Planning Seven Tools, Editor: Noriaki Kanda, Nikka Giren Publishing Co., Ltd.”.
(Mode 8) The purchaser's values survey is executed by a virtual negotiation method.
(Embodiment 9) The purchaser's values survey is executed in a virtual negotiation method using the Internet and the one-to-one comparison method.
[0023]
(Mode 10) In the method for supporting the determination of the attribute level of the planned product, in the third step, in addition to the attribute of the planned product and its level, a database for storing the attribute of the other product and its level is referred to. The sales share is calculated as a sales volume index.
(Mode 11) In the method for supporting the determination of the attribute level of the planned product, the third step to the sixth step are repeated in that order,
In the third step, the sales volume index of the planned product is obtained based on the level of the product attribute stored in the immediately preceding sixth step.
According to this, it is easy to obtain a more appropriate level of product attributes.
(Mode 12) In the method of mode 11, the sales volume index obtained in the third step based on the level of the attribute of the planned product stored each time the sixth step is repeatedly performed is the best (for example, sales The method further includes a step of outputting the level of the attribute of the planned product having the largest quantity or sales share) to the output device of the computer.
(Mode 13) The purchaser grouping method and the planned product attribute level determination support method can be understood as an apparatus or program invention.
[0024]
【Example】
A first embodiment of the present invention will be described. In this embodiment, the evaluation value (partial utility coefficient) measurement support apparatus shown in FIG. 1 is used. This device is a device that supports the process of measuring an evaluation value that the purchaser recognizes for the product attribute that the purchaser uses as a product selection factor and its level. This device is configured by a computer device, and the hardware configuration is the same as a normal one. This device includes a processing device (CPU), a storage device (an optical storage medium, a magnetic storage medium, or a semiconductor memory such as a RAM or a ROM), an input device, an output device, and the like. In this apparatus, a program for executing the processing of the present embodiment is installed. The apparatus includes a classification unit 62, a first unit 66, a second unit 70, a normalization unit 74, and databases 64, 68, 72, and 76.
[0025]
The classifying means 62 refers to the database 64 that stores in advance whether the product attribute group that the purchaser selects as a product selection factor, whether each product attribute belongs to a basic attribute, an intermediate attribute, or a detailed attribute. And classify them into basic attributes, intermediate attributes, and detailed attributes. The classification means 62 is realized by executing an attribute classification processing program on a computer device.
[0026]
The first means 66 sets the level group for each attribute with respect to the attribute group belonging to the basic attribute and the intermediate attribute of the product and conducts the purchaser's value survey to obtain the basic information from the purchaser. Measure the evaluation values allowed for the attribute and intermediate attribute levels. The measured evaluation values of the basic attribute and the intermediate attribute are stored in the evaluation value database 68 of the basic / intermediate attribute.
The second means 70 sets the level group for each attribute with respect to the attribute group belonging to the intermediate attribute and the detailed attribute of the product and conducts the purchaser's value survey, and the purchaser selects the intermediate group Measure the evaluation value allowed for the level of attributes and detailed attributes. The measured intermediate attribute and detailed attribute evaluation values are stored in the intermediate / detailed attribute evaluation value database 72.
The first means 66 and the second means 70 are realized by executing a conjoint analysis program on a computer device.
[0027]
The normalizing means 74 uses the first means 66 so that the evaluation value for each level of the intermediate attribute measured by the first means 66 is equal to the evaluation value for each level of the intermediate attribute measured by the second means 70. The measured evaluation value and the evaluation value measured by the second means 70 are normalized. The normalized evaluation values are stored in the basic / intermediate / detail attribute evaluation value database 76 that stores the evaluation values of the basic / intermediate / detail attributes in an integrated manner.
The normalizing means 74 is realized by executing a normalization processing program on a computer device.
[0028]
A second embodiment of the present invention will be described. In this embodiment, the evaluation value measurement support apparatus shown in FIG. 2 is used. This device is configured by a computer device as in the first embodiment, and includes a processing device, a storage device, an input device, an output device, and the like. Various databases store information that can be used in a computer. Various processing devices are realized by a computer operating according to a program. 2 and 3 illustrate the case where the present invention is applied to a product called an automobile. The application target of the present invention is not limited to automobiles, and can be widely applied to products that are provided in a variety of types because there are various demands.
[0029]
The database 2 stores the attributes and levels of products (in this case, automobiles). FIG. 4 exemplifies the storage contents of the database 2, and attributes include category, manufacturer, size, engine, price, displacement, fuel consumption, fuel used, drive system, transmission, navigation device, seat skin, audio device, A list of factors to be taken into consideration when the purchaser selects a purchased item, such as an aero equipment, a wheel, an air conditioner, a headlight, and a sunroof, is stored. A level is stored for each attribute. For example, the level of E1 company, E2 company, etc. is stored in the attribute of manufacturer, and the level of exhaust volume is stored in the attribute of displacement. In the attribute, the level that there is no navigation device, the level that has DVD navigation and the level that has DVD voice navigation are stored, and the level such as 15-inch aluminum wheel and 15-inch steel wheel is stored in the attribute of wheel In the attribute of sunroof, a level that there is no sunroof, a level that has an electric sunroof, and the like are stored.
[0030]
The level is stored for each automobile provided in the market, and the automobile is specified by the model name, grade, and option type. In FIG. 4, the car model name is A1, the grade is B1, and the option is C1 format, the category is D1 (for example, sedan, there are other categories such as minivans and compact cars) The manufacturer is E1, the size is F1 (for example, the size of the B segment, there are other sizes such as C segment and D segment), and the engine is G1 (for example, in-line 6-cylinder DOHC) Other than that, there are in-line 4-cylinder DOHC, V-type 6-cylinder DOHC, etc.), the price is H1 yen, the displacement is I1 liter, and the navigation device is J1 (DVD navigation device) In addition to this, there are specifications that do not have a navigation device, etc.).
This database is available for all cars on the market.
[0031]
The person who wants to investigate the values accepted by the purchaser assumes the vehicle model market to be investigated first. Then, a search condition is input from the search condition input unit 6 (FIG. 2). For example, a search condition such that the category is a sedan and the price is Hc to Hd is input.
Then, the vehicle type search processing unit 4 (FIG. 2) searches the database 2 to search for a corresponding product, and displays the name of the vehicle type having the searched product on the selected vehicle type / grade / option display unit 8 (FIG. 2). . At this time, the grades and options included in the searched model name are also displayed. The price is also displayed here.
A person who wants to investigate the values accepted by the purchaser is included in the survey object using the selected vehicle type / grade / option input unit 10 while referring to the vehicle type name / grade / option type / price displayed on the display unit 8. Select a product group. Specifically, the vehicle type name / grade / option displayed on the display unit 8 is selected by clicking. The number of products to be selected can be arbitrarily selected. If you select by car model name, all grades and option types belonging to that car model will be selected. If you select options, you can select only options.
[0032]
When a product is selected, the selected vehicle type / grade / option / attribute / level list creation processing unit 12 operates to create a list of attributes / levels of the selected product. At this stage, a list in which only the product group selected from the table of FIG. 4 is extracted is created.
When the attribute / level list of the selected product group is created, the attribute / level distribution list creation processing unit 14 operates to organize the level distribution for each attribute of the selected product group. FIG. 5 shows an example. The selected product group category is D1-D4, the manufacturer has E1-E4, the size is F1-F4, the engine is G1-G4, and the price is It can be seen that H1 to H4, displacements are I1 to I4, navigation specifications are J1 to J4, seat skins are K1 to K4, and audio specifications are L1 to L4.
[0033]
The database 16 stores information indicating whether each attribute included in the database 2 is a basic attribute, an intermediate attribute, or a detailed (equipment) attribute. For example, it is stored that the attributes related to the basic specifications of the product such as the displacement and the vehicle size are basic attributes, and the attributes related to the detailed specifications of the product such as the specifications of the door mirror and the lamps are detailed (equipment) attributes. For example, intermediate attributes such as navigation device and aero equipment specifications are stored as intermediate attributes. The basic attribute refers to an attribute related to the skeleton of the product such as the displacement and the vehicle size, for example. The detailed attribute refers to an attribute related to the details of the product such as the specifications of the door mirror and the lamps, for example. The intermediate attribute is an intermediate attribute between the product skeleton and the details, and the product price changes slightly when the level is different. The classification of basic attributes, intermediate attributes, and detailed (equipment) attributes is previously determined by a person and stored in the database 16.
[0034]
The basic attribute & intermediate attribute / level distribution list creation processing unit 18 (FIG. 2) creates a level distribution relating to attributes stored in the database 16 as basic attributes or intermediate attributes. The left side of FIG. 6 shows an example in which the level distribution relating to the basic attribute and the intermediate attribute is extracted from the level distribution list of FIG. The level distribution for detailed attributes has been eliminated. A list of level distributions related to basic attributes and intermediate attributes shown on the left side of FIG. 6 is displayed on the screen display unit 20.
Similarly, the intermediate attribute & detailed attribute / level distribution list creation processing unit 26 creates a level distribution related to attributes stored in the database 16 as intermediate attributes or detailed attributes. The right side of FIG. 6 shows an example in which the level distribution relating to the intermediate attribute and the detailed attribute is extracted from the level distribution list of FIG. The level distribution for basic attributes has been eliminated. The level distribution of intermediate attributes (in this case navigation-related attributes) is commonly included in both lists. The level distribution list regarding the intermediate attribute and the detailed attribute shown on the right side of FIG. 6 is displayed on the screen display unit 28.
The search vehicle type / grade / option display unit 8, the level distribution list display unit 20 for basic attributes and intermediate attributes, and the level distribution list display unit 28 for intermediate attributes and detailed attributes are physically the same display. By switching the display screen in time series, the search vehicle type / grade / option display unit 8, the level distribution list display unit 20 regarding basic attributes and intermediate attributes, For example, the level distribution list display unit 28 may be used for detailed attributes.
[0035]
In conjoint analysis, it is necessary to treat all attributes deleted from the level distribution list as the same. In the level distribution list regarding the basic attribute and the intermediate attribute on the left side of FIG. 6, it is necessary that the level of the deleted detailed attribute is the same. Therefore, in actuality, it is assumed that the distribution of detailed attributes on the right side of FIG. 6 is arranged at a standard level. It is assumed that the sheet skin, which is a detailed attribute, is actually a fabric (when this is a standard) when the level distribution on the left side of FIG. As a result, the level distribution relating to the price shown on the left side of FIG. 6 needs to be corrected from the actual price of FIG. This is because if the actual leather specification products are aligned with the fabric specification, the price of the genuine leather specification products will be reduced. The prices H1a to H4a shown in the level distribution regarding the price on the left side of FIG. 6 indicate the prices of products in which the level of the detailed attribute is aligned with the standard level.
Similarly, in the level distribution list regarding the intermediate attribute and the detailed attribute on the right side of FIG. 6, it is necessary that the level of the deleted basic attribute is the same. Therefore, in practice, the basic attributes on the left side of FIG. 6 are assumed to be distributed at a standard level as illustrated. Actually, even if the engine specification is G1 or G2, when the level distribution on the right side of FIG. 6 is used, it is uniformly G1 (when this is a standard). As a result, the level distribution regarding the price shown on the right side of FIG. 6 needs to be corrected from the actual price of FIG. This is because the price of a product changes if a product having a G2 engine is actually replaced with a product having a G1 engine. The prices H1b to H4b shown in the level distribution relating to the price on the right side of FIG. 6 indicate the prices of products whose basic attributes are set to the standard level.
[0036]
The person who conducts the survey can add, delete, or modify the level while viewing the level distribution (shown on the left side of FIG. 6) regarding the basic attribute and the intermediate attribute displayed on the list display unit 20 (FIG. 2). Means 22). The surveyor will remove some levels if the number of questions to the buyer becomes excessive. In this case, the investigator is assisted by displaying the effect of the deletion. For example, the level adjustment operation is assisted by displaying that the vehicle type D1 will be excluded from the investigation target if I1 is deleted from the displacement attribute. Researchers can add a level and virtually create a product that doesn't actually exist.
Similarly, the person who conducts the survey can add, delete, or modify the level while viewing the level distribution (shown on the right side of FIG. 6) regarding the intermediate attribute and the detailed attribute displayed on the list display unit 28. (Means 30 of FIG. 2). Here too, a display for assisting the addition / deletion / correction of the level is made.
[0037]
When the list of attribute groups exemplified on the left side of FIG. 6 and the level group of each attribute are set, the partial utility coefficient (evaluation value) for each attribute and level is investigated using a known conjoint analysis method. It becomes possible to measure.
Usually, a virtual product group with various levels is created by a computer. In order to measure the evaluation value (simply called evaluation value or partial utility coefficient of the level) that the purchaser accepts for many types with as few questions as possible The method of the orthogonal table of the method is adopted. The details are described in “Item 9.4 of Chapter 9 of Seven Product Planning Tools”.
Here, it is assumed that two virtual commodities are used for the paired comparison method, which is easy to be familiar with the investigation of the virtual negotiation method using the Internet. An example is shown in FIG. 7, and products 44 and 46, products 50 and 52, and the like, which are selected by the purchaser, are created. The level of the product to be created is selected using the orthogonal table method so that many kinds of evaluation values (partial utility coefficients) exist with a small number of products.
[0038]
The basic attribute & intermediate attribute survey program creation processing unit 24 uses the list on the left side of FIG. 6 and the orthogonal table method to generate pairs of products to be questioned one after another (products 44 and 46, products 50 and 52, etc.). Create.
Similarly, the intermediate attribute & detailed attribute investigation program creation processing unit 32 uses the list on the right side of FIG.
In the means 34 (FIG. 3) for conducting a value survey on basic attributes and intermediate attributes using the Internet, pairs of products created by the survey program creation processing unit 24 (products 44 and 46, products 50 and 52, etc.) Investigate the buyer's values using the virtual negotiation method. The purchaser who accesses using the Internet shows the products 44 and 46 illustrated in FIG. 7 and clicks the answer column 48 to obtain an answer. Next, the products 50 and 52 are shown and the answer column 54 is clicked to obtain an answer. It is executed one after another for a pair of products set in advance.
[0039]
When the conduct of the value survey relating to the basic attribute and the intermediate attribute is completed, the value survey relating to the intermediate attribute and the detailed attribute is subsequently carried out on the same survey subject (means 38 in FIG. 3). In this case, the same survey method is utilized, and a purchaser who accesses using the Internet shows the products 44 and 46 illustrated in FIG. Next, the products 50 and 52 are shown and the answer column 54 is clicked to obtain an answer. It is executed one after another for a pair of products set in advance. In this case, the question is repeated according to the product pair created by the intermediate attribute & detailed attribute survey program creation processing unit 32.
When answers are obtained from a considerable number of purchasers, the partial utility coefficient measurement unit 36 (FIG. 3) for each level of the basic attribute and intermediate attribute collects the values collected by the value survey execution means 34 regarding the basic attribute and intermediate attribute. Using the result, the evaluation value (partial utility coefficient) for each level of the basic attribute and intermediate attribute is measured. An example of the measurement result is illustrated on the left side of FIG. The small letter of the alphabet in the figure indicates the evaluation value (partial utility coefficient), and the purchaser recognizes that the value of g1 is recognized when the engine is G1, for example, and the value of g2 is recognized when the engine is G2. . For the prices H1a to H4a, an evaluation value as to whether or not it is worth the expenditure is obtained. For example, an evaluation that is worth the price of H1a yen is obtained at a rate of h1a, and an evaluation that is worth the price of H2a yen is h2a It can be seen that it is obtained at a ratio of Here, the subscript a is an evaluation value obtained by both the basic attribute and intermediate attribute investigation and the intermediate attribute and detailed attribute investigation, and the evaluation value obtained by the basic attribute and intermediate attribute investigation. It is shown that. The evaluation value or partial utility coefficient is a positive value for the level that favors the purchaser's product selection, has a large value that works favorably, and does not work favorably or disadvantageously when the purchaser selects the product The level is zero, the level that works adversely when the purchaser selects a product is negative, and the level that works harder disadvantageously is smaller (the absolute value is larger). The partial utility coefficient at the level that is most disadvantageous may be set to zero, and all other partial utility coefficients may be set to zero or more.
[0040]
Some attributes have a wide distribution of partial utility coefficients depending on the level, while others have a narrow distribution range. It can be seen that the attribute in which the partial utility coefficient is widely distributed depending on the level is an important attribute that greatly affects the purchaser's product selection. A large contribution is given to this attribute. It can be seen that an attribute whose partial utility coefficient does not change much depending on the level is an insignificant attribute that does not significantly affect the purchaser's product selection. A small contribution is given to this attribute. In the lowermost stage on the left side of FIG. 8, contribution rates (Da to Ja) are shown.
From the list of attributes and levels shown in FIG. 6, a pair of products used in the pair comparison method shown in FIG. 8 is created, and an evaluation value for each attribute and level is calculated from the obtained answer, and the contribution ratio of the attribute is calculated. Software that executes processing up to the calculation is commercially available, and the software can be used. For example, an adaptive conjoint analysis program can be obtained from Sawtooth.
[0041]
Similarly, the partial utility coefficient measuring unit 40 (FIG. 3) for each level of the intermediate attribute & detailed attribute uses the survey result collected by the value survey execution means 38 regarding the intermediate attribute & detailed attribute to obtain the intermediate attribute & Measure the evaluation value for each level of detailed attributes. An example of the measurement result is illustrated on the right side of FIG. The small letter of the alphabet in the figure indicates the evaluation value, and the purchaser recognizes that the value of k1b is recognized when the sheet skin is K1, for example, and the value of k2b is recognized when the sheet skin is K2. The subscript b indicates an evaluation value obtained by examining the intermediate attribute and the detailed attribute. The evaluation value or partial utility coefficient is a positive value for the level that favors the purchaser's product selection, has a large value that works favorably, and does not work favorably or disadvantageously when the purchaser selects the product The level is zero, the level that works adversely when the purchaser selects a product is negative, and the level that works harder disadvantageously is smaller (the absolute value is larger). In the lowermost row on the right side of FIG. 8, the contribution ratios (Jb to Hb) for each of the intermediate attribute and the detailed attribute are shown.
[0042]
The evaluation value shown on the left side of FIG. 8 has a meaning indicating the relative value within the range of the attribute and level shown on the left side, and the evaluation value shown on the right side of FIG. 8 is within the range of the attribute and level shown on the right side. However, it is meaningless to directly compare the evaluation value on the left side with the evaluation value on the right side. This is because the relative evaluation on the left side is completed on the left side, and the relative evaluation on the right side is completed on the right side.
[0043]
However, in this case, the evaluation value of the intermediate attribute is commonly included in the basic attribute & intermediate attribute evaluation value list shown on the left side of FIG. 8 and the intermediate attribute & detailed attribute evaluation value list shown on the right side. Yes. Normalization processing is performed by the normalizing means 41 (FIG. 3) using the evaluation value of the intermediate attribute included in common. As a result, it is possible to obtain a list in which the evaluation value list shown on the left side of FIG. 8 and the evaluation value list shown on the right side, which are essentially unrelated, are integrated on the same scale. The evaluation values and the contribution rates of the levels of the basic attribute, the intermediate attribute, and the detailed attribute thus normalized are stored in the database 42 illustrated in FIG. 3 and can be displayed as a list by the list display unit 43.
[0044]
FIG. 9 shows levels J1 (no navigation is set), J2 (equipped with CD navigation), J3 (DVD navigation) of intermediate attributes J (navigation attributes) that are commonly included in the left and right of FIG. Is shown), and the evaluation value of level J4 (equipped with DVD voice navigation) is graphically displayed.
The evaluation values at each level include the evaluation values (subscript a) surveyed in the basic attribute & intermediate attribute world shown on the left side of FIG. 8, and the intermediate attribute & detailed attribute world shown on the right side of FIG. The evaluated value (subscript b) exists. The former is indicated by a black circle and a solid line, and the latter is indicated by an x mark and a broken line. Both do not match.
[0045]
FIG. 10 plots an evaluation value (subscript a) investigated in the world of basic attributes & intermediate attributes on one axis, and an evaluation value (subscript b) investigated in the world of intermediate attributes & detailed attributes on an axis orthogonal to the axis. ) Is plotted. The two do not agree but correlate well. A correlation equation can be obtained by applying the method of least squares. Once the correlation formula is obtained, it is possible to convert the evaluation value surveyed in the world of intermediate attributes and detailed attributes into the evaluation value surveyed in the world of basic attributes & intermediate attributes, and conversely, the world of basic attributes & intermediate attributes It is also possible to convert the evaluation value investigated in (1) into an evaluation value investigated in the world of intermediate attributes and detailed attributes.
[0046]
FIG. 11 shows the evaluation value (subscript b) investigated in the world of intermediate attribute & detailed attribute shown in the right side of FIG. 8 as the evaluation value (subscript c) investigated in the world of basic attribute & intermediate attribute. FIG. 9 shows a list that is converted and displayed in a unified manner with the evaluation values investigated in the world of basic attributes and detailed attributes shown on the left side of FIG. 8. Regarding the contribution rate, the contribution rate (subscript b) measured from the evaluation value surveyed in the world of intermediate attributes and detailed attributes is added to the contribution rate (subscript b) measured from the evaluation value surveyed in the world of basic attributes & intermediate attributes. The character c) is converted.
It has been verified that the contribution ratios of basic attributes, intermediate attributes, and detailed attributes, and the evaluation values at each level, which are unified in this way, have high accuracy.
[0047]
The price levels shown in the table of FIG. 11 are the prices H1a to H4a when the detailed level shown on the left side of FIG. 6 is the standard, and the basic levels shown on the right side of FIG. Prices H1b to H4b are set. In this case, 8 levels (H1 to H8) are set for convenience. The evaluation values for the prices H1a to H4a are the same as the evaluation values (subscript a) obtained from the survey on basic attributes and intermediate attributes, but the evaluation values for the prices H1b to H4b are related to intermediate attributes and detailed attributes. The evaluation value (subscript b) obtained from the survey is normalized (converted) to the level of the evaluation value (subscript a) obtained from the survey on the basic attribute and the intermediate attribute according to the correlation equation shown in FIG. An evaluation value (subscript c) is adopted. As for the contribution rate, the contribution rate (subscript c) calculated from the evaluation value (subscript c) normalized (converted) to the level of the evaluation value (subscript a) obtained from the investigation on the basic attribute and the intermediate attribute Has been fixed.
[0048]
Buyers who surveyed values do not necessarily represent all buyers. In particular, if you use the Internet to conduct a value survey, you can conduct a quick and inexpensive survey, but the survey target is biased toward purchasers who are familiar with the Internet. The values obtained from the survey cannot be used as the average value of all buyers.
In FIG. 12, according to a value survey using the Internet, males recognize a partial utility coefficient of 0.7 against the level that the manufacturer is manufactured by Company I, whereas females have a partial utility coefficient of 0.4. For example, males only accept a partial utility factor of 0.3 against the level of being manufactured by Company II, while women have a partial utility factor of 0.6.
If the composition ratio of men and women in the region is 60% and 40%, the weighted average partial utility coefficient is calculated by the formula shown in FIG. 12, and the average purchaser in the region is made by company I. A partial utility coefficient of 0.58 is recognized against the standard, and a partial utility coefficient of 0.42 is recognized against the level of being manufactured by Company II.
[0049]
When the attributes / levels / partial utility coefficients (evaluation values) as shown in FIG. 11 are arranged, the attribute / level data of the actual product stored in the database 2 is used together to occupy the competitive product group. Shares by product can be calculated. As shown in FIG. 13, if the partial utility coefficient of attribute j at level j is βi, j and the level of a product is a level marked with a circle in the figure, the number of sales of the product is It is known that it is proportional to the exponential value of the sum of the partial utility coefficients with circles. If the number of sales of all competing products is calculated, the share of each product in the competing product group can be calculated.
[0050]
In addition to the above, various measures can be taken to estimate the number of sales. In storing the partial utility coefficient for each attribute level, the partial utility coefficient for each attribute level can be stored for each purchaser. As shown in FIG. 14, the partial utility coefficient β for each attribute level of the purchaser 1 is stored, the partial utility coefficient χ for each attribute level of the purchaser 2 is stored, and for each attribute level of the purchaser 3. If the partial utility coefficient λ is stored, it is possible to calculate the value that the purchaser recognizes for the specific product by inputting the attribute and level setting of the specific product. For example, it is possible to know the comprehensive evaluation value that the purchaser 1 recognizes for the product 1, the comprehensive evaluation value that the purchaser 1 recognizes for the product 2, and the like. The comprehensive evaluation value can be calculated by the equation of FIG.
When calculating the value of each competitive product for a single buyer, the product that the purchaser recognizes the highest value is revealed, and the product that the purchaser will select when selecting the product is known. it can. By executing this for all survey subjects, the share by product among all survey subjects can be calculated. For each combination of purchaser and product, if the value that the purchaser recognizes for the product is quantified, the share for each product can be estimated and calculated by processing that value as the purchaser's product selection probability. .
[0051]
By comparing the share calculated from the evaluation value (partial utility coefficient) with the actual share, the accuracy of the evaluation value can be evaluated. When the share by product is calculated using the unified evaluation value list illustrated in FIG. 11, a result that matches the actual share is obtained. It can be seen that this technique can accurately measure the evaluation value for each level from basic attributes to detailed attributes.
[0052]
The technique can be implemented in multiple stages. Basic attribute-intermediate attribute-detailed attribute analysis By applying the unified technology to what was previously a detailed attribute, the basic attribute-intermediate attribute-detailed attribute-intermediate attribute-further detailed attribute chain Obtainable. In this case, it is possible to accurately measure the contribution ratio of the attribute and the evaluation value of the level from the basic attribute having a large contribution ratio to further detailed attributes having a small contribution ratio.
[0053]
In addition, the specification regarding an aero equipment other than the specification regarding a navigation apparatus can be employ | adopted for an intermediate attribute. In addition, for example, when the displacement is distributed within a fine and narrow range of 1.3, 1.4, and 1.5 liters, the displacement can be set as an intermediate attribute. The number of attributes to be intermediate attributes is not limited to 1 and may be 2 or more.
[0054]
A third embodiment of the present invention will be described. In this embodiment, an apparatus that supports the determination of the level of product attributes shown in FIG. 15 is used. This device is configured by a computer device as in the first embodiment, and includes a processing device, a storage device, an input device, an output device, and the like. In this apparatus, by executing a program for performing the processing of the present embodiment, the purchaser grouping means 102, the sales share calculating means 106, and the product attribute level correction determining means 110 are mainly executed by the processing apparatus. The product attribute level correcting means 112, the product attribute level initial setting means 116, and the product attribute level determining means 118 are realized. A group information database (DB) 104, a sales share database 108, and a product attribute and level database 114 are realized by the storage device. Information stored in each database 104, 108, 114 can be output to an output device (display device or the like) 120 as necessary. This embodiment also illustrates the case where the present invention is applied to a product called an automobile, as in the second embodiment.
[0055]
Next, processing contents of the product attribute level determination support apparatus shown in FIG. 15 will be described with reference to the flowchart of FIG. Further, a specific example (see FIG. 17) will be described. This specific example is a simplified example for easy understanding of the contents.
[0056]
The purchaser grouping means 102 in FIG. 15 performs a grouping process using the normalized evaluation values (partial utility coefficients) for each level of the basic attribute, the intermediate attribute, and the detailed attribute to group the purchasers ( (S10 in FIG. 16). Evaluation values for attributes and levels are stored for each purchaser, and the distribution of evaluation values differs for each purchaser. In order to group the evaluation value distribution for each purchaser, a cluster analysis method can be used. Cluster analysis generally means a general analysis method for classifying a plurality of individuals (in this embodiment, purchasers) into groups having similar properties. Cluster analysis is a kind of multivariate analysis technique and is used in fields such as marketing research.
The evaluation value can be obtained from the evaluation value database 76 (see FIG. 1) described in the first embodiment or the evaluation value database 42 (see FIG. 3) described in the second embodiment. Further, from the databases 76 and 42, an evaluation value given by the purchaser for each attribute level can be obtained for each purchaser. Cluster analysis is performed using the evaluation value data. Specifically, for example, it is possible to perform cluster analysis by obtaining the contribution rate of each attribute from this evaluation value data for each purchaser and using the distribution of the contribution rate. By calculating the contribution rate from the normalized evaluation value data, the contribution rate is normalized between the attributes as a result.
As described above, an attribute whose evaluation value varies greatly depending on the level to which it belongs has a high contribution rate. An attribute whose evaluation value does not vary much depending on the level to which it belongs has a low contribution rate. That is, the magnitude of the contribution rate depends on the fluctuation range of the evaluation value when the level is changed. In the second embodiment, since the contribution rate has already been obtained and stored in the database 42 (see FIG. 3), the contribution rate can be used.
[0057]
An example of cluster analysis will be described. For example, as shown in FIG. 18, a coordinate space is prepared in which the contribution ratio of the product attribute is plotted on the horizontal axis and the contribution ratio of the displacement is plotted on the vertical axis. Next, each purchaser is plotted at the corresponding location in the coordinate space. For example, the purchaser P1 has a contribution ratio of the price X1 and a contribution ratio of the exhaust amount Y1, so the purchaser P1 plots it at the position (X1, Y1). In the case of FIG. 18, the Euclidean distance ((X1-X2) between purchasers P1 and P2.²+ (Y1-Y2)²)^1/2Is, for example, the Euclidean distance between purchasers P1 and P3 ((X1-X3)²+ (Y1-Y3)²)^1/2Shorter than For this reason, the purchasers P1 and P2 are grouped. Further, the Euclidean distance between the purchasers P3 and P4 is shorter than the Euclidean distance between the purchasers P1 and P3, for example. Therefore, purchasers P3 and P4 are grouped. In this example, the group including the purchasers P1 and P2 has a property that the contribution ratio of the price is large and the contribution ratio of the displacement is small. In other words, this group can be said to be a price-oriented group. Considering the same, the group including purchasers P3 and P4 can be said to be a group that emphasizes displacement. In this way, purchasers are grouped.
However, for simplification of explanation, the case of two attributes of price and displacement has been described as an example, but actually, cluster analysis is performed in a situation where there are more attributes.
[0058]
According to the above-described grouping process, the purchasers can be grouped according to the category of which product attribute is important rather than the category of gender and age. The element of emphasizing the product attribute directly affects the sales volume of the product rather than factors such as gender and age. Therefore, by using the information obtained by this grouping, it becomes easier to plan a product with a larger sales volume than in the past.
[0059]
As a result of cluster analysis, not only groups that emphasize basic attributes, but also groups that emphasize intermediate attributes and detailed attributes may be formed. For example, audio is a detailed attribute, but those who like music are likely to place importance on audio performance. As a result, a group that places importance on audio may be formed.
[0060]
As a specific example, it is assumed that the purchaser is classified into four groups Q1 to Q4 as a result of the cluster analysis (see FIG. 17A). It is assumed that the group Q1 places importance on the price, the group Q2 places importance on the displacement, the group Q3 places importance on the size, and the group Q4 places importance on the audio. Price, displacement, and size are basic attributes, but audio is a detailed attribute. When the total number of purchasers belonging to groups Q1 to Q4 is 1, the ratio of purchasers belonging to group Q1 is assumed to be 0.4. Similarly, it is assumed that the ratios of purchasers belonging to the groups Q2, Q3, and Q4 are 0.3, 0.2, and 0.1, respectively.
[0061]
The purchaser grouping means 102 generates other information about the group as necessary (S12). For example, an average value in group units of an evaluation value of a certain product attribute level is generated, or an average value in group units of a contribution rate possessed by a certain product attribute is generated.
[0062]
Information about the group obtained by the processing of the grouping means 102 is stored in the group-specific databases 104a, 104b, 104c,. In addition to information about which buyers belong to the group, information about the group includes information about the buyers belonging to the group (not only individual purchaser information but also statistical information of all purchasers belonging to the group) Included).
[0063]
19 and 20 show a part of the data stored in the group information database 104 in a graph format. The vertical axis in FIG. 19 indicates the evaluation value. The horizontal axis in FIG. 19 indicates the level of product attributes (H1, I1, etc.) corresponding to the evaluation value on the vertical axis for each attribute (price, displacement, etc.). FIG. 19 shows the average evaluation value of all the groups and the average evaluation value of the price-oriented group. FIG. 20 corresponds to FIG. 19, with the vertical axis representing the contribution ratio and the horizontal axis representing the product attribute.
FIG. 19 shows that in the price-oriented group, the fluctuation range of the evaluation value when the price level is changed to H1, H2,... Is larger than the average of all the groups. This means that, as shown in FIG. 20, in the price-oriented group, the contribution ratio of the price is larger than the average of all the groups.
[0064]
Further, as shown in FIGS. 19 and 20, in the price-oriented group, the fluctuation range or the contribution rate of the evaluation value of the fuel efficiency is larger than the whole. For this reason, it can be said that the price-oriented group is a group that places importance on fuel efficiency. In this way, one group may place importance on two or more product attributes.
[0065]
The initial setting means 116 performs initial setting of product attribute levels (S14). The initially set product attribute level is stored in the database 114a of the product and product attribute and level in the database 114. The initial setting data may be input by an operator using an input device.
As a specific example, it is assumed that the price of a product is 2 million yen, the displacement is 2.0 L, the size is C segment, and the audio is MD and is initially set (see FIG. 17B).
[0066]
The group-specific sales share calculation unit 106a of the sales share calculation means 106 calculates and determines the sales share for each group based on the set level of the product attribute (S16). The obtained sales share for each group is stored for each group in the databases 108a, 108b, 108c,.
The sales share calculation unit 106a for each group may calculate and obtain the sales share of the new product of each company according to the method described with reference to FIG. 13 or 14 in the second embodiment. As data necessary for using these methods, it is possible to obtain the contribution ratio of each group attribute and the evaluation value of the level from the group information database 104. The attribute and level data of the company product can be obtained from the database 114a of the attribute and level of the company product. From the database 114b of the attributes and levels of the other company's products, the data of the attributes and levels of the other company's products can be obtained.
[0067]
As a specific example, when the total number of sales of the company's products and other companies' products in group Q1 is 1, the ratio (sales share) of the number of sales of the company's products in group Q1 is calculated to be 0.1. . Similarly, it is assumed that the sales shares of the groups Q2, Q3, and Q4 are calculated as 0.2, 0.4, and 0.4, respectively (see FIG. 17C).
[0068]
Based on the ratio of purchasers belonging to each group and the sales share of the company's own products for each group, the total sales share calculation unit 106b calculates and determines the total sales share of the company's products for all groups. The obtained total sales share is stored in the database 108x of the sales share database 108 (S18).
[0069]
In the above example, group Q_iOf group buyers and group Q_iThe product of the sales share of the company's own products is calculated for the case where i is changed from 1 to 4, and when these are totaled, the total sales share is obtained. That means
0.4 × 0.1 + 0.3 × 0.2 + 0.2 × 0.4 + 0.1 × 0.4 = 0.22
Is the total sales share (see FIG. 17D).
[0070]
The merchandise attribute level correction determination means 110 obtains total sales share data from the database 108. The correction determination unit 110 determines whether or not to correct the product attribute level by a predetermined determination method (S20). As an example of a determination method, if the product attribute level has not been corrected a predetermined number of times, it is determined that the product attribute level is to be corrected (YES in S20), and a processing command is issued to the product attribute level correction unit 112 (see S22). ). On the other hand, if the product attribute level has already been corrected the predetermined number of times, it is determined that the product attribute level has not been corrected (NO in S20), and a processing command is issued to the product attribute level determining means 118 (see S26). . Whether or not the correction has been performed a predetermined number of times may be determined based on the number of times the correction determination unit 110 has obtained the data on the total sales share from the database 108.
[0071]
In a specific example, when the following correction processing of S22 and S24 is performed three times, NO is determined in S20 and the process proceeds to S26.
[0072]
The determination and correction unit 112a of the product attribute level correction means 112 determines a product attribute that requires a level correction based on the group-specific sales share input from the sales share database 108, and sets the attribute level. It corrects (S22). The corrected level is stored in the attribute and level database 114a of the company product.
Specifically, the following processing is performed. (1) Search for a group whose group sales share is smaller than the total sales share. (2) Next, among the groups searched in (1), a group having the largest purchaser belonging is further searched. In the above specific example, the group Q1 corresponds. (3) Next, it is determined that the attribute emphasized by the group is an attribute that requires level correction. In the above specific example, the group Q1 is a group placing importance on the price, and it is determined that the price level needs to be corrected, that is, the price needs to be lowered. (4) Next, the level of the attribute is corrected. In the above specific example, it is assumed that the price level is lowered by one step, from 2 million yen to 1.8 million yen (see FIG. 18 (e)).
[0073]
In the above specific example, although the ratio of purchasers belonging to the group Q1 is as large as 0.4, the sales share of the group Q1 is as small as 0.1. Therefore, there is a high possibility that the total sales share can be increased by correcting the level of the product attribute so as to increase the sales share of the group Q1. Since group Q1 is a group that places importance on price, it is effective to lower the price in order to increase the sales share of group Q1.
[0074]
The level resetting unit 112b of the correcting unit 112 resets the level of other attributes of the product according to the determination and correction by the correcting unit 112a (S24). The reset level is stored in the database 114a of the attributes and levels of the company's products.
For example, if the price (sale price) of a product is lowered, the production cost must be lowered accordingly. Along with this, other attribute levels also need to be reset. The computer apparatus shown in FIG. 15 also stores cost data (not shown). This cost data includes data stored in association with each attribute level. For example, the cost of audio with MD is stored, and the cost of audio without MD is also stored. By referring to this cost data, the resetting unit 112b resets the level of each attribute so that the production cost of the entire product is equal to or lower than the allowable production cost.
[0075]
In the above example, as the price is reduced from 2 million yen to 1.8 million yen, the displacement is reduced from 2.0L to 1.8L, the size is changed from the C segment to the smaller B segment, and the audio is sent. Assume that the setting is reset from MD to MD (see FIG. 17F).
[0076]
The group-based sales share calculation unit 106a calculates and determines the sales share of the company's own product for each group based on the reset product attribute level (S16).
In the above specific example, as a result of the calculation, the sales shares of the in-house products of the groups Q1, Q2, Q3, and Q4 are 0.25, 0.15, 0.3, and 0.3, respectively (FIG. 17 (g )reference).
[0077]
The total sales share calculation unit 106b calculates and determines the total sales share of the company's products that combine all the groups (S18).
In the above example,
0.4 × 0.25 + 0.3 × 0.15 + 0.2 × 0.3 + 0.1 × 0.3 = 0.235
Is the total sales share (see FIG. 17 (h)).
[0078]
Thus, compared to the total sales share (0.22, see FIG. 17D) calculated before the correction processing of S22 and S24), the total sales share (0 .235, FIG. 17 (h)) is higher.
[0079]
As described above, in the specific example, the correction processing of S22 and S24 is performed three times. As a result of performing the second correction process, it is assumed that an overall sales share of 0.232 is obtained in the subsequent process of S18. As a result of performing the third correction process, it is assumed that an overall sales share of 0.227 is obtained in the subsequent process of S18. That is, if the total sales share calculated before the correction process is included, four total sales share values of 0.22, 0.235, 0.232, and 0.227 are obtained.
[0080]
The product attribute level determining means 118 determines and outputs the product attribute level that maximizes the total sales share (S26). The data group of the total sales share is input to the determination unit 118 from the database 108x. The determination unit 118 searches for the maximum value from these data groups. Then, the determination means 118 acquires the product attribute level corresponding to the maximum total sales share from the database 114a, and outputs this to the output device 120 as the attribute level of the new product.
[0081]
In the above specific example, the total sales share value is the largest 0.235 obtained in the calculation of S18 after the first correction process. Product attribute level that realizes this total sales share value (as mentioned above, price: 1.8 million yen, displacement: 1.8L, size: B segment, audio: no MD, see FIG. 17 (f)) Is obtained from the database 114 a and output to the output device 120.
[0082]
In the above specific example, the case where the correction process is performed only three times has been described in order to simplify the description. However, the correction process may be performed several hundred times, several thousand times, or more. Of course.
[0083]
In the above specific example, the number of correction processes is used as a reference as a criterion for determining whether or not the correction is performed in S20, but is not limited thereto. For example, if the value of the total sales share input to the correction determination unit 110 is smaller than a predetermined value, the correction is made (YES in S20), and if it is greater than the predetermined value, no correction is made (NO in S20). A discrimination criterion such as) may be adopted. In the above specific example, for example, when the total sales share value obtained by calculation is smaller than 0.23, S20 may be set as YES, and when 0.23 or more, NO may be set as S20. In this case, the second and subsequent correction processes are not performed.
[0084]
Further, as a result of the correction being made a plurality of times, if the value of the obtained total sales share gradually increases and then the value of the total sales share gradually decreases, no further correction is performed at that time (S20). And NO) may be adopted. In the above specific example, the total sales share value calculated after the first correction process is the peak (0.235), and the total sales share value calculated after the second and third correction processes is 0. It becomes 232, 0.227, and it decreases gradually. In this case, in S20 after the third correction process (the processes in S22 and S24), it may be determined that no further correction is performed.
[0085]
Further, the process in S22 is not limited to the above process. In S22, for example, a group having a sales share of a predetermined value or less in the group group or a level of attributes emphasized by a group having the smallest sales share may be corrected. Further, the levels of a plurality of groups may be corrected. Further, the level correction amount may be changed according to the sales share value.
[0086]
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.
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 can achieve a plurality of purposes at the same time, and has technical utility by achieving one of the purposes.
[Brief description of the drawings]
FIG. 1 is a diagram showing a configuration of an evaluation value measurement support device used in a first embodiment.
FIG. 2 is a diagram (part 1) illustrating a configuration of an evaluation value measurement support device used in a second embodiment.
FIG. 3 is a diagram showing a configuration of an evaluation value measurement support apparatus used in the second embodiment (part 2);
FIG. 4 is a diagram showing an example of a database that stores products, attributes, and levels.
FIG. 5 is a diagram showing an example of a distribution of levels of selected product groups.
FIG. 6 is a diagram illustrating an example of a distribution of levels of basic attributes and intermediate attributes and an example of a distribution of levels of intermediate attributes and detailed attributes at the same time.
FIG. 7 is a diagram showing an example of a display screen when conducting a value survey by the paired comparison method using the Internet.
FIG. 8 is a diagram showing simultaneously an example of measured evaluation values of levels belonging to basic attributes and intermediate attributes, and an example of evaluation values of levels belonging to intermediate attributes and detailed attributes;
FIG. 9 is a diagram showing evaluation values measured for levels belonging to intermediate attributes in comparison with values measured from basic attribute and intermediate attribute investigations and values measured from intermediate attribute and detailed attribute investigations. .
FIG. 10 is established between the evaluation value for the level belonging to the intermediate attribute measured from the basic attribute and intermediate attribute survey and the evaluation value for the level belonging to the intermediate attribute measured from the intermediate attribute and detailed attribute survey. The figure which illustrates the correlation to do.
FIG. 11 is a diagram showing an example in which evaluation values of levels belonging to basic attributes, intermediate attributes, and detailed attributes are unified using partial utility coefficients converted by a correlation equation.
FIG. 12 is a diagram showing an example in which the bias of the survey target person is compensated and converted into an evaluation value for an average purchaser.
FIG. 13 is a diagram showing a process of calculating the number of merchandise sales from a level partial utility coefficient.
FIG. 14 is a diagram showing an example of a process of estimating a sales share from an evaluation value of each purchaser
FIG. 15 is a diagram showing a configuration of a product attribute level determination support apparatus used in the third embodiment;
FIG. 16 is a flowchart showing the processing contents of the product attribute level determination support apparatus used in the third embodiment;
FIG. 17 is an explanatory diagram of a specific example.
FIG. 18 is an explanatory diagram of cluster analysis.
FIG. 19 is a diagram showing an evaluation value of an average value for all groups and an evaluation value of a price-oriented group average.
FIG. 20 is a diagram showing an average contribution ratio of all groups and an average contribution ratio of price-oriented groups.
[Explanation of symbols]
2: Vehicle type / grade / option / product attribute / level database
14: Attribute / level distribution list creation processing section
16: Basic attribute / intermediate attribute / detail (equipment) attribute database
18: Basic & intermediate attribute / level distribution list creation processing section
24: Basic and intermediate attribute value survey program creation processing section
26: Intermediate & detailed attribute / level distribution list creation processing section
32: Intermediate & detailed attribute value survey program creation processing part
34: Values survey implementation department for basic and intermediate attributes using the Internet
36: Partial utility coefficient calculation unit for each level of basic and intermediate attributes
38: Values survey implementation department for intermediate and detailed attributes using the Internet
40: Partial utility coefficient calculation unit for each level of intermediate & detailed attributes
41: Normalization processing unit
42: DB of partial utility coefficient and contribution rate for each level of basic, intermediate and detailed attributes normalized and unified

Claims (6)

This is a method of measuring the evaluation value that the buyer accepts for the product attribute and the level that the buyer selects the product,
By referring to a database that stores in advance whether each product attribute belongs to a basic attribute, intermediate attribute, or detailed attribute by computer, classifies the product attribute group into basic attribute, intermediate attribute, and detailed attribute Executing an attribute classification program to perform,
A conjoint analysis program is executed by a computer on the survey results obtained by conducting a survey of buyers' values by setting a level group for each attribute group belonging to basic attributes and intermediate attributes. A first step of storing the obtained basic attribute and intermediate attribute evaluation values in the basic / intermediate attribute evaluation value database;
A conjoint analysis program is executed by a computer on the survey results obtained by setting a level group for each attribute of the attribute group belonging to the intermediate attribute and the detailed attribute and conducting a survey of buyers' values. A second step of storing the obtained intermediate attribute and detailed attribute evaluation values in an intermediate / detailed attribute evaluation value database;
Evaluation value for each intermediate attribute level stored in the basic / intermediate attribute evaluation value database in the first step, and evaluation value for each intermediate attribute level stored in the intermediate / detailed attribute evaluation value database in the second step So that the evaluation value stored in the basic / intermediate attribute evaluation value database and the evaluation value stored in the intermediate / detail attribute evaluation value database are normalized so that A hierarchical evaluation value measuring method comprising: a step of storing the evaluation values in a basic / intermediate / detailed attribute evaluation value database for integrating and storing the evaluation values. Executing a program for grouping evaluation value distributions stored for each purchaser in the evaluation value database of basic / intermediate / detailed attributes by the method of claim 1 by a computer, and grouping purchasers;
A method for grouping purchasers, comprising: storing a relationship between purchasers and groups obtained in the process in a group information database. It is a method that supports the process of determining the level for each attribute of the planned product,
Referring to a database storing the attributes and levels of planned products, a sales volume index calculation program is executed by a computer for each group grouped by the method of claim 2 to obtain a sales volume index of the planned products. 3 steps,
A fourth step of storing the sales volume index of the planned product for each group obtained in the third process in the sales volume index database;
By referring to the sales volume index of the planned product for each group stored in the sales volume index database, the computer sets the attribute level of the planned product so as to improve the sales volume index of the group whose sales volume index does not meet the predetermined standard A fifth step to be corrected by
A method for supporting determination of the level of the attribute of the planned product, comprising a sixth step of storing the level corrected in the fifth step in the database storing the attribute of the planned product and the level. It is a device that supports the process of measuring the evaluation value that the purchaser accepts for the product attribute and the level that the purchaser makes the product selection factor,
Attribute classification that classifies product attribute groups into basic attributes, intermediate attributes, and detailed attributes with reference to a database that stores in advance whether each product attribute belongs to basic attributes, intermediate attributes, or detailed attributes Means for executing the program;
With respect to the fundamental attributes and intermediate attribute to set the level group for each attribute for belonging attribute group conducted a purchaser of Values Survey obtained survey result of the product, run the conjoint analysis program First means for storing the obtained basic attribute and intermediate attribute evaluation values in the basic / intermediate attribute evaluation value database ;
The intermediate attributes and the values survey of buyers by setting the level group for each attribute obtained by performed on the attribute group belonging to the Advanced Attributes survey results of goods, by running the conjoint analysis program Second means for storing the obtained intermediate attribute and detailed attribute evaluation values in an intermediate / detailed attribute evaluation value database ;
And evaluation value for each level of the intermediate attributes stored in the evaluation value database Basic / intermediate attribute in the first means, the evaluation value for each level of the intermediate attributes stored in the evaluation value database of the intermediate / advanced attributes in the second means So that the evaluation value stored in the basic / intermediate attribute evaluation value database and the evaluation value stored in the intermediate / detail attribute evaluation value database are normalized so that the A device for supporting the implementation of hierarchical evaluation value measurement, comprising means for storing the evaluation value in a basic / intermediate / detailed attribute evaluation value database for integrating and storing the evaluation values . On the computer,
Attribute classification that classifies product attribute groups into basic attributes, intermediate attributes, and detailed attributes with reference to a database that stores in advance whether each product attribute belongs to basic attributes, intermediate attributes, or detailed attributes Processing to execute the program;
Execute a conjoint analysis program on the survey results obtained by conducting a survey of buyers' values by setting a level group for each attribute group belonging to the basic attribute and intermediate attribute of the product A first process of storing the obtained basic attribute and intermediate attribute evaluation values in the basic / intermediate attribute evaluation value database ;
Execute a conjoint analysis program for the survey results obtained by conducting a survey of buyers' values by setting a level group for each attribute group belonging to the intermediate attribute and detailed attribute of the product. A second process of storing the obtained intermediate attribute and detailed attribute evaluation values in the intermediate / detailed attribute evaluation value database ;
A first evaluation value for each level of the intermediate attributes stored in the evaluation value database Basic / intermediate attribute in the process, the level each of the evaluation values of the intermediate attributes stored in the evaluation value database of the intermediate / advanced attributes in the second process So that the evaluation value stored in the basic / intermediate attribute evaluation value database and the evaluation value stored in the intermediate / detail attribute evaluation value database are normalized so that the A program for hierarchical evaluation value measurement that executes processing for storing evaluation values in a basic / intermediate / detail attribute evaluation value database that integrates and stores evaluation values. This is a method of measuring the evaluation value that the buyer accepts for the car attributes and levels that the buyer considers when choosing a car,
By referring to a database that stores in advance whether each car attribute belongs to a basic attribute, an intermediate attribute, or an equipment attribute by a computer, the car attribute group is classified into a basic attribute, an intermediate attribute, and an equipment attribute. Executing an attribute classification program to perform,
A conjoint analysis program is executed by a computer on the survey results obtained by conducting a survey of buyers' values by setting a level group for each attribute group belonging to basic attributes and intermediate attributes. A first step of storing the obtained basic attribute and intermediate attribute evaluation values in the basic / intermediate attribute evaluation value database;
A conjoint analysis program is executed by a computer on the survey results obtained by conducting a survey of buyers' values by setting a level group for each attribute group belonging to the intermediate attribute and equipment attribute. A second step of storing the obtained intermediate attribute and equipment attribute evaluation values in an intermediate / detailed attribute evaluation value database;
Evaluation value for each intermediate attribute level stored in the basic / intermediate attribute evaluation value database in the first step, and evaluation value for each intermediate attribute level stored in the intermediate / equipment attribute evaluation value database in the second step So that the evaluation value stored in the basic / intermediate attribute evaluation value database and the evaluation value stored in the intermediate / equipment attribute evaluation value database are normalized so that A method for measuring evaluation values of vehicle attributes and levels, comprising: a step of storing evaluation values in a basic / intermediate / equipment attribute evaluation value database for storing evaluation values in an integrated manner.

Images