JP2010191792A - Query distribution system and query distribution method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a query distribution system capable of evaluating proficiency of an answerer in each category of queries. <P>SOLUTION: This query distribution system 100 includes an evaluation value calculating part for calculating an evaluation value (for example, proficiency s) indicating reasonability of reply processing of the answerer in each category of the queries, a correction factor calculating part for calculating a correction factor for the evaluation value, and the first answerer evaluation part for calculating the first distribution reference value a serving as an index for determining a distributed side of a new query, by correcting the evaluation value by the correction factor γ. The correction factor calculating part calculates the correction factor γ to reduce the reasonability along with the higher degree of difficulty of the reply processing in each category. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an inquiry distribution system and an inquiry distribution method for distributing an accepted inquiry to a plurality of respondents.

  Conventionally, in an inquiry distribution system that accepts an inquiry from a user and distributes the inquiry to a plurality of respondents, can the inquiry that arrives at each respondent be distributed using the current workload of the respondent as an index? Or, it was distributed to respondents according to the content of the inquiry, for example, the product type.

  However, among a plurality of respondents, both inexperienced persons and experienced persons are mixed, and it is not efficient to simply distribute inquiries based only on the workload and product type. Also, if appropriate inquiry distribution cannot be performed, the response time will be unnecessarily delayed. On the other hand, if an inquiry is to be distributed efficiently, in order to distribute the inquiry, an administrator-level expert who is familiar with the contents of all inquiries and the skills of the respondents is required.

  Patent Document 1 discloses that an inquiry is distributed based on operator information associated with operator identification information.

  Further, Patent Document 2 discloses a system in which the respondent's field of expertise is investigated in advance and an inquiry is automatically distributed to the respondent using the field.

  Further, Patent Document 3 accepts an evaluation of the answer process performed by the respondent from the questioner, and responds to the questioner's evaluation together with the question information performed by the questioner and the answer information performed by the respondent. In addition, a system for storing in a database is disclosed.

JP 2002-359686 A Japanese Patent Laid-Open No. 2005-316828 JP 2004-157733 A

  The content of the inquiry arriving at the inquiry system has a different level of difficulty in answer processing for each category, for example, for each product and corresponding content. However, the systems disclosed in Patent Documents 1 to 3 cannot evaluate the skills of respondents according to the difficulty level of answer processing for each category.

  Therefore, an object of this invention is to provide the inquiry distribution system which can evaluate the answer processing skill of a respondent according to the difficulty of the answer processing for every category.

  One aspect of the inquiry distribution system according to the present invention includes an evaluation value calculation unit that calculates an evaluation value indicating the appropriateness of answer processing of a respondent for each inquiry category, and a correction coefficient calculation unit that calculates a correction coefficient for the evaluation value And a first respondent evaluation unit that calculates a first distribution reference value a serving as an index for determining a new inquiry distribution destination by correcting the evaluation value with the correction coefficient γ. The correction coefficient calculation unit calculates the correction coefficient γ so as to decrease the appropriateness as the difficulty level of the answer processing for each category increases.

  One aspect of the inquiry distribution method according to the present invention is to calculate an evaluation value indicating the appropriateness of the respondent's answer processing for each inquiry category, calculate a correction coefficient for the evaluation value, and calculate the evaluation value by the correction coefficient. By correcting, a first distribution reference value serving as an index for determining a new inquiry distribution destination is calculated. The correction coefficient is calculated by a computer so as to decrease the suitability as the difficulty of the answer process for each category increases.

  According to one aspect of the inquiry distribution system according to the present invention, it is possible to provide a system that can quantitatively evaluate the proficiency level of respondents for each category of inquiry contents.

It is a figure which shows the structural example of the inquiry distribution system 1000 which concerns on Embodiment 1. FIG. It is a figure which shows the inquiry distribution system 100 which concerns on Embodiment 2, and its periphery. It is a figure which shows the questionnaire sent to the inquirer terminal 1 from the answer and quality evaluation management part 7 in the inquiry distribution system 100 which concerns on Embodiment 2. FIG. It is a figure which shows the further detailed structural example of the 1st respondent evaluation part 3 and the 2nd respondent evaluation part 4 in the inquiry distribution system 100 which concerns on Embodiment 2. FIG. It is a satisfaction-point conversion table of the 1st respondent evaluation part 3 in the inquiry distribution system 100 which concerns on Embodiment 2. FIG. It is a TAT-point conversion table of the 1st respondent evaluation part 3 in the inquiry distribution system 100 concerning Embodiment 2. FIG. It is a figure which shows the function F (n) for calculating the correction coefficient (gamma) hold | maintained in the 1st respondent evaluation part 3 in the inquiry distribution system 100 which concerns on Embodiment 2. FIG. 10 is a flowchart showing an operation of the inquiry distribution system 100 according to the second embodiment. It is a flowchart which shows operation | movement of the 1st respondent evaluation part 3 of the inquiry distribution system 100 which concerns on Embodiment 2. FIG. It is a figure which shows operation | movement of the 2nd respondent evaluation part 4 of the inquiry distribution system 100 which concerns on Embodiment 2. FIG. It is a figure which shows the flow of a process in the inquiry distribution system 100 which concerns on Embodiment 2. FIG. In the inquiry distribution system 100 which concerns on Embodiment 2, it is a figure which shows the inquiry information sent to the reception management part 2 from the inquirer terminal 1. FIG. It is a figure which shows the inquiry information registered into inquiry management DB6 in the inquiry distribution system 100 which concerns on Embodiment 2. FIG. It is a figure which shows the flow of a process of the 1st respondent evaluation part 3 in the inquiry distribution system 100 which concerns on Embodiment 2. FIG. It is a figure which shows the flow of a process of the 2nd respondent evaluation part 4 in the inquiry distribution system 100 which concerns on Embodiment 2. FIG. It is a figure which shows the effect of the inquiry distribution system 100 which concerns on Embodiment 2. FIG. It is a figure which shows the effect of the inquiry distribution system 100 which concerns on Embodiment 2. FIG. It is a figure which shows the effect of the inquiry distribution system 100 which concerns on Embodiment 2. FIG.

[Embodiment 1]
Embodiment 1 of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram showing a configuration example of an inquiry distribution system 1000 according to Embodiment 1 of the present invention. Here, the inquiry distribution system 1000 is a system that receives an inquiry from an inquirer and distributes the inquiry to each respondent. The inquiry distribution system 1000 only needs to have a function of evaluating at least respondents. The inquiry distribution system 1000 has a function of distributing the inquiry to the respondents based on the evaluation results of the respondents. Or may be provided in an external system.

  As illustrated in FIG. 1, the inquiry distribution system 1000 includes an evaluation value calculation unit 101, a correction coefficient calculation unit 102, and a first respondent evaluation unit 103. The evaluation value calculation unit 101 calculates an evaluation value indicating the appropriateness of the respondent's answer processing for each category of inquiry. Here, the category refers to the type of product to be inquired or the type of inquiries such as complaint handling and product description. For example, the evaluation value is set as a value that improves as the respondent's experience value increases.

  The correction coefficient calculation unit 102 calculates the correction coefficient γ so as to decrease the suitability as the difficulty level of answer processing for each category increases. As the first distribution reference value a, the first distribution reference value a serving as an index for determining a distribution destination of a new inquiry is calculated by correcting the evaluation value with the correction coefficient γ.

  Here, the inquiry arriving at the inquiry distribution system 1000 includes a reply process with a high difficulty level and a reply process with a low difficulty level depending on the type of product and the type of inquiry. Therefore, in the inquiry distribution system 1000 according to the first embodiment, the first distribution reference value a is calculated using a different correction coefficient γ according to the difficulty level of the response processing between categories, thereby performing the response processing. Even for inquiries between a plurality of categories having different levels of difficulty, the answerer's answer processing skills can be evaluated by one index.

  The correction coefficient γ may be a constant value set for each category, or may be a value that changes according to the experience value. When the correction coefficient γ is set to a value that changes corresponding to the experience value, a table in which the experience value and the correction coefficient γ are associated with each other is prepared, and the correction coefficient γ can be determined according to this table. Further, the correction coefficient γ may be determined by a function of experience values. For example, this function can be configured as a function that converges to a constant value at a different convergence speed depending on the difficulty level of answer processing for each category.

[Embodiment 2]
Next, a configuration example when the inquiry distribution system 1000 shown in FIG. 1 is implemented more specifically will be described. FIG. 2 is a block diagram illustrating a configuration example of the distribution system 100 and the periphery of the distribution system 100. As shown in FIG. 2, an inquiryer terminal 1 used by an inquiryer who makes an inquiry is connected to an inquiry distribution system 100 via a network. The inquirer terminal 1 has a function of sending inquiry information to the inquiry distribution system and a function of receiving an answer to the inquiry from the inquiry distribution system 100. In the second embodiment, the inquiry distribution system 100 will be described as a system that makes inquiries and answers by e-mail.

  The inquiry distribution system 100 includes a reception management unit 2, a first respondent evaluation unit 3, a second respondent evaluation unit 4, a respondent terminal 5, an inquiry management database (DB) 6, and an answer / quality evaluation management unit 7. I have.

  The reception management unit 2 collectively receives inquiries transmitted from the inquiry terminal 1. The reception management unit 2 has a customer information DB 21. The customer information DB 21 manages a customer number that is an identification number of the inquirer and customer information associated with the customer number. The reception management unit 2 authenticates the inquirer by comparing the customer number included in the inquiry transmitted by the inquirer with the customer number stored in the customer information DB 21.

  If the customer number included in the inquiry and the customer number stored in the customer information DB 21 do not match, the reception management unit 2 replies to the inquirer terminal 1 that the customer number cannot be received. Further, the reception management unit 2 assigns an inquiry ID for identifying an inquiry to the inquiry that has been successfully authenticated. Then, the inquiry is transferred to the second respondent evaluation unit 4 together with the inquiry ID and the customer number of the inquirer who made the inquiry.

  The first respondent evaluator 3 has the functions of the evaluation value calculator 101, the correction coefficient calculator 102, and the first respondent evaluator 103 in the first embodiment. Specifically, the first respondent evaluation unit 3 calculates the evaluation value of the reply process of the respondent and calculates the correction coefficient γ. Then, the first respondent evaluation unit 3 calculates the distribution reference value a for each respondent by correcting the evaluation value with the correction coefficient γ. In the second embodiment, the evaluation value is a proficiency level s calculated based on a questionnaire answered by the inquirer. Specifically, the first respondent evaluator 3 obtains an answer to a questionnaire registered in the inquiry management DB 6, and calculates the proficiency level s by converting the answer into a preset point.

  The second respondent evaluation unit 4 is configured to calculate the distribution reference value b based on the distribution reference value a calculated by the first respondent evaluation unit 3. The second respondent evaluation unit 4 calculates the distribution reference value b based on the distribution reference value a and the inquiry coefficient k that is the number of inquiries currently being handled by each respondent. The distribution reference value b is a reference value for distributing a newly arrived inquiry.

  The respondent terminal 5 includes a plurality of respondent terminals 51 to 53. The respondent uses each of the respondent terminals 51 to 53 to receive an inquiry through the reception management unit 2 and the distribution control unit 3, and creates an answer to the inquiry. In order to identify respondents who use the respective respondent terminals 51 to 53, the respondents are referred to as respondents 51 to 53.

  The answer / quality evaluation management unit 7 receives the answers transmitted from the respondent terminals 51 to 53 in a lump, and transfers the answers to the inquiry terminal 1 that made the inquiry. In addition, the answer / quality management unit 7 sends a questionnaire for performing quality evaluation on the answer to the inquirer terminal 1 when transferring the answer. The answer / quality evaluation management unit 7 records the answer of the questionnaire sent from the inquirer in the inquiry management DB 6.

  FIG. 3 is a diagram illustrating an example of a questionnaire transmitted from the answer / quality evaluation management unit 7 to the inquirer terminal 1. The questionnaire is configured to evaluate the quality of the answer according to the satisfaction level felt by the inquirer with respect to the answer processing performed by the answerer. For example, the questionnaire includes items that can be selected by the inquirer such as very satisfied, satisfied, normal, dissatisfied, or very dissatisfied. In addition, regarding the causes that have affected satisfaction, it is possible to select specific situations such as the time to obtain a solution, the appropriateness of explanation at the window, and the appropriateness of interim reporting of the investigation status. ing.

  As shown in FIG. 2, the answer / quality evaluation management unit 7 receives from the inquirer terminal 1 a questionnaire result in which the satisfaction level of the inquirer terminal 1 is reflected. The answer / quality evaluation management unit 7 stores the questionnaire result transmitted from the inquirer terminal 1 in the inquiry management DB 6 in association with the inquiry content and the answer to the inquiry result.

  The reception management unit 2, the inquiry management DB 6, the respondent terminal 5, and the like do not have to be provided inside the inquiry distribution system 100, and may be provided outside the inquiry distribution system 100.

  Next, configuration examples of the first respondent evaluation unit 3 and the second respondent evaluation unit 4 in the inquiry distribution system 100 according to Embodiment 2 will be further described. FIG. 4 is a diagram illustrating a more detailed configuration example of the first respondent evaluation unit 3 and the second respondent evaluation unit 4.

  The first respondent evaluation unit 3 calculates the distribution reference value a according to the quality evaluation acquired from the inquiry management DB 6 and a preset correction coefficient γ. Specifically, the first respondent evaluation unit 3 performs the answer processing based on the satisfaction level of the inquirer's answer processing obtained from the questionnaire result obtained from the inquiry management DB 6 and TAT (Turn Around Time). The proficiency level s is calculated for each category. Here, the proficiency level s is an index indicating the answer skill of the respondent. The first respondent evaluator 3 calculates the sorting reference value a by multiplying the proficiency level s by a preset correction coefficient γ.

  The TAT indicates a period of time from when the inquirer terminal 1 transmits an inquiry to the inquiry distribution system 100 until a response is received by the inquirer terminal 1. The category of answer processing includes the type of product such as products A, B, and C that are the subject of answer processing, the type of processing content such as complaint handling, product description, and the like. Here, for explanation, the category is described as a product.

  The first respondent evaluation unit 3 has a satisfaction-point conversion table shown in FIG. The satisfaction level-point conversion table is a conversion table for reflecting the satisfaction level in the proficiency level s, and stores the satisfaction level and points acquired from the inquiry management unit 6 in association with each other. Specifically, as shown in FIG. 5, the satisfaction-point conversion table is 5 points for very satisfied, 3 points for satisfied, 0 points for normal, -3 points for unsatisfactory, and very unsatisfactory. On the other hand, -5 points are stored in association with each other.

  Moreover, the 1st respondent evaluation part 3 is further provided with the TAT-point conversion table shown in FIG. In the TAT-point conversion table, TAT and points are associated with each other. Specifically, TAT is associated with 0 points for day 0, TAT is associated with 6 points for day 1,.

  The first respondent evaluation unit 3 converts the satisfaction and TAT acquired from the inquiry management DB 6 into points using the satisfaction-point conversion table and the TAT-point conversion table. Then, a value obtained by adding the points obtained by converting the satisfaction and the TAT is defined as a proficiency level s. This proficiency level s is calculated for each combination of respondent and product. In the following, the combination of the respondent and the product is represented as (respondent, product).

  As shown in FIG. 4, the first respondent evaluator 3 includes a proficiency level table 30. The proficiency level table 30 holds the proficiency level s calculated by the above method for each (respondent, product) combination.

  Moreover, the 1st respondent evaluation part 3 is provided with the experience value table 31, as shown in FIG. The experience value table 31 holds an experience value n for each combination of (respondent, product). Here, the experience value n is the number of cases in which the respondent has completed the answer process regarding the target product within a predetermined period in the past. For example, when the respondent 51 has completed 50 reply processes for the product A, 50 is entered in the column of (respondent 51, product A).

  Further, the first respondent evaluation unit 3 holds a correction coefficient γ set in advance for each category of answer processing. FIG. 7 is a diagram showing a function F (n) for calculating the correction coefficient γ held in the first respondent evaluation unit 3. The function F (n) is a function that is different for each product. The function F (n) is a function in which the correction coefficient γ is uniquely determined according to the value of the experience value n. This function F (n) is a function that converges from 0 or more to 1 as the experience value n increases.

  In the function F (n), the value of the experience value n that starts to converge to 1 differs depending on the difficulty level of the answer processing of the product (category). Specifically, when the answer processing becomes increasingly difficult in the order of product A, product B, and product C, the value of experience value n that starts to converge to 1 in the order of product A, product B, and product C is It gets bigger.

  The first respondent evaluation unit 3 obtains the correction coefficient γ by substituting the experience value n held in the experience value table 31 into F (n). Each time the data in the inquiry management DB 6 is updated, the first respondent evaluation unit 3 acquires the satisfaction level, the TAT, and the experience value n, and sequentially updates the proficiency level table 30 and the experience value table 31. Then, the correction coefficient γ is obtained from the experience value n from the function F (n). And the 1st respondent evaluation part 3 calculates the distribution reference value a from the following formula | equation (1) based on the proficiency s, the experience value n, and the correction coefficient γ.

(Expression 1) Correction coefficient γ × skill level s / experience value n = distribution reference value a

  The first respondent evaluation unit 3 sends the distribution reference value a calculated for each (respondent, product) combination to the second respondent evaluation unit 4.

  The second respondent evaluation unit 4 includes a distribution table 40 and a final distribution order table 41. In the distribution table 40, for each combination of (respondent, product), the number k of correspondence items and the distribution reference value b are stored in association with each other. In the distribution order table 41, the distribution order set based on the distribution table 40 is stored for each respondent. The second respondent evaluation unit 4 distributes the inquiry transferred from the reception management unit 2 to appropriate respondents by referring to the distribution order table 41.

  The number k of correspondence is the number of inquiries for which the answer processing is incomplete among the inquiries distributed to the respondents 51 to 53. Here, the result that the second respondent evaluation unit 4 distributes to each respondent is also transmitted to the inquiry management DB. The inquiry management DB 6 calculates, for each respondent, the number of inquiries for which no answer has been registered among the inquiries registered in the self, and transmits the calculated number of inquiries k to the second respondent evaluation unit 4. Note that the handling number k is the number of all inquiries distributed to the respondent regardless of the product. The number k of correspondences can be obtained directly from the inquiry terminals 51 to 53, for example, and the answer registered in the inquiry management DB 6 is completed from the number assigned to each respondent in the second respondent evaluation unit 4 It can also be obtained by subtracting the number of inquiries made.

  The second respondent evaluator 4 calculates the distribution reference value b according to the expression (2) based on the distribution reference value a and the corresponding number of cases k input from the first respondent evaluator 3. The distribution reference value b is calculated for each combination of (respondent, product).

(Formula 2) Corresponding number of cases k / sorting reference value a = sorting reference value b

  The second respondent evaluation unit 4 adds the distribution reference value b calculated for each (respondent, product) for each respondent. Then, the total of the distribution reference values b calculated for each respondent is compared, and the distribution rank is given in ascending order from the smallest of the distribution reference values b. Specifically, in the order of respondent 51, respondent 52, and respondent 53, when the total of the distribution reference values b gradually increases, the respondent 51 ranks first and the respondent 52 ranks second. The respondent 53 is given a ranking of 3rd place.

  The second respondent evaluation unit 4 distributes the inquiry transferred from the reception management unit 2 to the respondents 51 to 53 (respondent terminals 51 to 53) based on the sequentially updated distribution order table 41.

  Next, the operation of the inquiry distribution system 100 configured as described above will be described. FIG. 8 is a flowchart showing the operation of the inquiry distribution system 100.

  The inquirer makes an inquiry such as a question about the product to the inquiry distribution system 100 using the inquirer terminal 1 (step S1). At the time of this inquiry, information related to the inquiry, such as the customer number that is the identification number of the inquirer, the name of the inquirer, the category of the inquiry content, the inquiry content, and the response deadline, is transmitted from the inquiry terminal 1 to the reception management unit 2. Is done. Here, the inquiry content category is, for example, information for identifying an inquiry product, for example, a product name or a product number.

  The reception management unit 2 determines whether the inquiry sent from the inquiry terminal 1 can be received by the inquiry distribution system 100 (step S2). Specifically, it is determined whether or not the customer number transmitted from the inquirer terminal 1 exists in the customer management DB 21.

  Here, when the customer number transmitted from the inquirer terminal 1 does not exist in the customer management DB 21 (impossible in step S2), a reply indicating that the customer number cannot be accepted is returned to the inquirer terminal 1 (step S4).

  On the other hand, if the customer number stored in the customer management DB 21 matches the customer number transmitted from the inquirer terminal 1 in step S2 (possible in step S2), identification information for identifying the inquiry Is assigned to the inquiry and notified (step S3). The second respondent evaluator 4 forwards the inquiry to the second respondent evaluator 4 together with the inquiry ID (step S5).

  In addition to the inquiry information transmitted from the inquiry terminal 1 such as the customer number, the name of the inquirer, the category of the inquiry content, the inquiry content, and the response deadline, the reception management unit 2 receives the inquiry ID, customer information, and inquiry reception date and time. Are registered in the inquiry management DB 6 (step S6).

  The second respondent evaluator 4 selects a respondent to deal with the transferred inquiry from the respondents 51 to 53 based on the distribution order table 41 of the second respondent evaluator 4 (step S7). . The second respondent evaluation unit 4 transmits an inquiry to the selected respondent (step S8).

  The respondents 51 to 53 confirm the content of the inquiry received from the second respondent evaluation unit 4 and create an answer to the inquiry. Then, the respondent terminals 51 to 53 transmit the answers created by the respondents 51 to 53 to the answer / quality evaluation management unit 7 (step S9). At this time, the respondent terminals 51 to 53 transmit the inquiry ID, the identification information of the respondent, and the content of the answer to the answer / quality evaluation management unit 7.

  The answer / quality evaluation management unit 7 forwards the answer received from the respondent terminals 51 to 53 to the corresponding inquirer terminal 1 based on the received inquiry ID (step S10). At this time, the data transmitted from the answer / quality evaluation management unit 7 to the inquirer terminal 1 includes the questionnaire information shown in FIG. 3 in addition to the contents of the answer sent from the answerer terminals 51 to 53.

  The respondent reflects the answer to the inquiry in the questionnaire information and sends it back to the answer / quality evaluation management unit 7 as the questionnaire result.

  When the answer / quality evaluation management unit 7 receives the questionnaire result transmitted from the inquirer terminal 1, the answer content sent from the respondent terminals 51 to 53, the corresponding questionnaire result, and the TAT (Turn Around Time) are displayed. And registered in the inquiry management DB 6 (step S11).

  When the data regarding the inquiry is registered in the inquiry management DB 6, the first respondent evaluation unit 3 acquires data including the quality evaluation of the answer made by the inquirer from the inquiry management DB 6 (step S12). The inquiry management DB 6 transmits, as information related to the questionnaire results, the satisfaction degree with respect to the answer processing performed by the inquirer such as satisfaction or dissatisfaction and the TAT.

  The first respondent evaluator 3 calculates the distribution reference value a based on the data sent from the inquiry management DB 6 and sends it to the second respondent evaluator 4 (step S13).

  Next, a specific operation of the first respondent evaluation unit 3 will be described. FIG. 9 is a flowchart showing the operation of the first respondent evaluation unit 3. The first respondent evaluator 3 acquires TAT and satisfaction for each inquiry ID from the data registered in the inquiry management DB 6 (step S21).

  Subsequently, the first respondent evaluation unit 3 converts the TAT days acquired from the inquiry management DB 6 into points according to the TAT-point conversion table shown in FIG. 6 (step S22). Fluctuating, the first respondent evaluator 3 converts the satisfaction into set points according to the satisfaction-point conversion table shown in FIG.

  The first respondent evaluation unit 3 sums up the converted points of the TAT and the quality evaluation of the response, and calculates the value as the proficiency level s (step S23). Then, the first respondent evaluation unit 3 updates the proficiency level table 30 and reflects the calculated proficiency level s in the proficiency level table 30 (step S24).

  Moreover, the 1st respondent evaluation part 3 updates the experience value table 31 based on the data sent from inquiry management DB6 (step S25). Specifically, the experience value of the product corresponding to the data sent from the inquiry management DB 6 is counted up by one.

  The first respondent evaluation unit 3 obtains a correction coefficient γ from the function F (n) based on the experience value n that is sequentially updated. Then, the first respondent evaluation unit 3 calculates the distribution reference value a from the equation (1) based on the proficiency level s, the experience value n, and the correction coefficient γ (step S26).

  The first respondent evaluation unit 3 transmits the calculated distribution reference value a to the second respondent evaluation unit 4 (step S27).

  Next, a specific operation of the second respondent evaluation unit 4 in the inquiry distribution system 100 will be described. FIG. 10 is a flowchart showing the operation of the second respondent evaluation unit 4. The second respondent evaluator 4 acquires the distribution reference value a and the number k of correspondences from the first responder evaluator 3 and the respondent terminal 5 (step S31). The second respondent evaluation unit 4 calculates the distribution reference value b from the acquired distribution reference value a and the number of corresponding cases k using Expression (2).

  The second respondent evaluation unit 4 calculates the sum of the distribution reference values b for each product for each respondent. Then, the second respondent evaluation unit 4 compares the total of the distribution reference values b, and assigns the distribution rank in ascending order from the smallest distribution reference value b. The second respondent evaluator 4 updates the distribution table 40 and the distribution order table 41 (step S33). Then, the newly arrived inquiry is distributed to appropriate respondents according to the distribution order table 41 (step S34).

  Next, the operation of the inquiry distribution system 100 described in FIG. 8 will be described along a specific example. FIG. 11 is a diagram illustrating a processing flow of the inquiry distribution system 100 according to the second embodiment. As indicated by the arrow Y1, the inquirer uses the product A, and inquires the inquiry distribution system 100 when a problem or a question about the product A occurs (step in FIG. 8). S1). When making an inquiry, as shown in FIG. 12, information relating to the inquiry such as the customer number, the name of the inquirer, the name of the product to be inquired, the contents of the inquiry, and the response deadline is sent from the inquirer terminal 1 to the inquiry distribution system 100 Sent.

  As shown in FIG. 11, the reception manager 2 determines whether or not the customer number in the inquiry information transmitted from the inquirer terminal 1 exists in the customer management DB 21 (step S2 in FIG. 8). When the customer number exists in the customer management DB 21, the reception management unit 2 gives an inquiry ID to the inquiry transmitted from the inquirer terminal 1 (step S3 in FIG. 8). Then, as shown by the arrow Y2 in FIG. 11, the received inquiry is transferred to the second respondent evaluation unit 4 (step S5 in FIG. 8).

  Further, as shown by an arrow Y3 in FIG. 11, the reception management unit 2 registers it in the inquiry management DB 6 in association with the inquiry ID to which information related to the inquiry transmitted from the inquirer terminal 1 is assigned (step in FIG. 8) S6). Here, as shown in FIG. 13, the information registered in the inquiry management DB 6 includes an inquiry ID, a customer number, customer information, an inquirer's name, a reception date and time, an inquiry product name, an inquiry content, an answer deadline, and the like. Yes.

  As shown by an arrow Y4 in FIG. 11, the distribution control unit 3 distributes the inquiry transferred from the reception management unit 2 to the respondent terminals 51 to 53 according to the distribution order in the distribution table 40 (FIG. 8). Step S8). As indicated by the arrow Y5, the respondents 51 to 53 who have received the inquiry create an answer to the inquiry and transmit the answer to the answer / quality evaluation management unit 7 (step S9 in FIG. 8). As shown by the arrow Y6, the answer / quality evaluation management unit 7 transmits the answer created by the respondents 51 to 53 to the inquiry terminal 1 that made the inquiry (step S10 in FIG. 8).

  Here, the answer / quality evaluation management unit 7 sends a questionnaire to the inquirer terminal 1 when sending the answer to the inquiry terminal 1 that made the inquiry.

  The inquirer receives the reply sent from the inquiry distribution system 100, answers the questionnaire information, and transmits the questionnaire result to the inquiry distribution system 100. As shown by an arrow Y7 in FIG. 11, the answer / quality evaluation management unit 7 of the inquiry distribution system 100 registers the questionnaire result received from the inquirer terminal 1 in the inquiry DB 6 (step S11 in FIG. 8).

  As indicated by the arrow Y8, the inquiry DB 6 sends information on the questionnaire result to the first respondent evaluation unit 3 (step S12 in FIG. 8). The inquiry management DB 6 also transmits TAT as information related to the questionnaire result, in addition to the satisfaction level of the answer processing performed by the inquirer such as satisfaction or dissatisfaction. As indicated by the arrow Y9, the first respondent evaluator 3 calculates the distribution reference value a and transmits it to the second respondent evaluator 4 (step S13 in FIG. 8).

  Next, the operation of the first respondent evaluation unit 3 shown in FIG. 9 will be described according to a specific example. FIG. 14 is a diagram showing a flow of processing of the first respondent evaluation unit 3. As indicated by the arrow Y11, the first respondent evaluator 3 acquires the respondent name, the inquiry product, the TAT, and the satisfaction level associated with the inquiry ID from the inquiry management DB 6 (step S21 in FIG. 9). .

  In the arrow Y12, the first respondent evaluation unit 3 converts the satisfaction degree of the inquiry information sent from the inquiry management DB 6 into points according to the satisfaction point conversion table (step S22 in FIG. 9). Specifically, since the satisfaction included in the quality evaluation is very satisfactory, the satisfaction is converted into 5 points according to FIG. Further, at the arrow Y12, the first respondent evaluation unit 3 converts the TAT of the inquiry information sent from the inquiry management DB 6 into points according to the TAT-point conversion table of FIG. 6 (step S22 of FIG. 9). Specifically, since TAT is 4 days, TAT is converted to 3 points.

  The first respondent evaluator 3 adds the value obtained by adding the points converted to the satisfaction level and the TAT to the proficiency level s of the corresponding (respondent, product) combination in the proficiency level table. Specifically, 8 points obtained by adding 5 points converted to satisfaction and 3 points converted to TAT are added to the proficiency level s before update in the product A of the respondent 51 as a target of the questionnaire result. . As a result, the proficiency level s of the product A of the respondent 51 is updated from 50 to 58 in the proficiency level table 30.

  On the other hand, the first respondent evaluation unit 3 counts up the corresponding experience value by 1 for the combination of (respondent, product) based on the information about the inquiry input from the inquiry management DB 6. Specifically, when 51 respondents are included in the information about the inquiry and the product is A, the experience value n in the column corresponding to the combination of (respondent 51, product A) is 49. 1 is added to make 50.

  The first respondent evaluation unit 3 obtains a correction coefficient γ from F (n) based on the experience value n calculated in this way. Then, as shown by the arrow Y15, the distribution reference value a is obtained by substituting the correction coefficient γ, the proficiency s, and the experience value n into the equation (1) (step S26 in FIG. 9). At the arrow Y16, the calculated distribution reference value a is transmitted to the second respondent evaluation unit 4 (step S27 in FIG. 9).

  Next, the operation of the first respondent evaluation unit 3 shown in FIG. 10 will be described according to a specific example. FIG. 15 is a diagram showing a flow of processing of the second respondent evaluation unit 4. As indicated by an arrow Y21, the second respondent evaluation unit 4 acquires the distribution reference value a calculated by the first respondent evaluation unit 3 for each combination of (respondent, product). Moreover, the 2nd respondent evaluation part 4 acquires the response | compatibility number k of each respondent 51-53 (step S31 in FIG. 10). Then, the distribution reference value b is calculated by substituting the acquired distribution reference value a and the number of correspondence cases k into Expression (2). Then, the distribution reference value b calculated for each product is added for each of the respondents 51 to 53.

  Then, the total of the distribution reference values b is compared, and the distribution order is given in ascending order from the smallest value. The second respondent evaluation unit 4 distributes the newly arrived inquiry to the respondents 51 to 53 in accordance with the sequentially updated distribution order (step S34 in FIG. 10). Specifically, inquiries are preferentially distributed from respondents with a low distribution order.

  Next, the effect of the inquiry distribution system 100 configured as described above will be described with reference to FIG. FIG. 16 is a diagram showing the first respondent evaluation unit 3 and the second respondent evaluation unit 4 in the inquiry distribution system 100 according to Embodiment 2 of the present invention. The proficiency level table 30 of the first respondent evaluation unit 3 holds the proficiency level s for each respondent 51 to 53 for each product. Specifically, the proficiency level s of all the products A, B, and C of the respondent 51 is 640, the respondent 52 is 240, and the respondent 53 is 1000.

  The experience value table 31 holds experience values n for each product for each of the respondents 51 to 53. Specifically, the experience value n of all the products A, B, and C of the respondent 51 is 80, the respondent 52 is 40, and the respondent 53 is 100. The correction coefficient γ determined based on the experience value n is 1 for the product A, 0.9 for the product B, and 0.7 for the product C in the respondent 51. For the respondent 52, the correction coefficient γ is 1 for product A, 0.7 for product B, and 0.4 for product C. In the respondent 53, the correction coefficient γ is 1 for the product A, 0.95 for the product B, and 0.85 for the product C.

  The first respondent evaluator 3 calculates the distribution reference value a based on the proficiency level s, the experience value n, and the correction coefficient γ from the equation (1), and transmits it to the second respondent evaluator 4. The second respondent evaluation unit 4 holds the distribution reference value a. Specifically, in the respondent 51, the distribution reference value a of the product A is 8.0, the product Bha 7.2, and the product C is 5.6. In the respondent 52, the distribution reference value a of the product A is 6.0, the product B is 4.2, and the product C is 2.4. In the respondent 53, the distribution reference value a of the product A is 10.0, the product B is 9.5, and the product C is 8.5.

  In addition, in the second respondent evaluation unit 4, the number of responses k of the acquired respondents 51 to 53 is 24 for the respondent 51, 15 for the respondent 52, and 30 for the respondent 53. . In addition, according to the formula (2), the distribution reference value b calculated based on the distribution reference value a and the corresponding number of cases k is 3.0 for the product A and 3.3 for the product B in the respondent 51. Product C is 4.3. The distribution reference value b of the respondent 52 is 2.5 for the product A, 6.3 for the product B, and 12.3 for the product C. The distribution reference value b of the respondent 53 is 3.0 for the product A, 3.2 for the product B, and 9.7 for the product C.

  The total of the distribution reference values b of the respondents 51 to 53 is 10.6 for the respondent 51, 12.3 for the respondent 52, and 9.7 for the respondent 53. Accordingly, in the distribution order, the answerer 51 is second, the answerer 52 is third, and the answerer 53 is first. Therefore, the inquiry that has newly arrived at the inquiry distribution system 100 is transferred to the respondent 53 who is ranked first.

  Here, the response number k of the respondents 51 to 53 is the largest among the respondents 53, and then decreases in the order of the respondents 51 and 52, but the number of respondents to which the inquiry was transferred is 53. It can be seen that the inquiries are distributed based on the proficiency level s of the respondent without depending on the number of correspondence cases k.

  16 and 17 are diagrams further illustrating the effect of the second embodiment. From the state shown in FIG. 15, as shown in FIG. 16, it is assumed that the experience value n of all products A, B, and C is uniformly increased by 20 in all respondents 51 to 53. In addition, the proficiency level s of the respondent 51 is increased by 120 points in all the products A, B, and C, the proficiency level s of the respondent 52 is increased by 160 points, and the proficiency level s of the respondent C is decreased by 500 points. To do. That is, it is assumed that the proficiency level s of the respondent C is significantly lower than the proficiency levels s of the other respondents 51 and 52. As a result, the proficiency level table 30, the experience value table 31, and the correction coefficient γ change as shown in FIG.

  As a result, the distribution reference value a of the second respondent evaluation unit 4 is 7.6 for product A, 7.5 for product B, and 6.1 for product C in the respondent 51. The distribution reference value a of the respondent 52 is 6.7 for the product A, 5.3 for the product B, and 4.0 for the product C. The distribution reference value a of the respondent 53 is 4.2 for the product A, 4.2 for the product B, and 4.2 for the product C. As in FIG. 15, the responding case number k remains 24 for the respondent 51, 15 for the respondent 52, and 30 for the respondent 53.

  The distribution reference value b is 3.2 for the product A, 3.2 for the product B, and 4.0 for the product C in the respondent 51 according to the distribution reference value a and the number k of correspondence cases. In the respondent 52, the distribution reference value b of the product A is 2.25, the product B is 2.8, and the product C is 3.8. In the respondent 53, the distribution reference value b of the products A to C is 7.2. As a result, the total of the distribution reference values b of the respondents 51 to 53 is 10.3 for the respondent 51, 8.8 for the respondent 52, and 21.6 for the respondent 53. As a result, the respondent 51 is ranked second, the respondent 52 is ranked first, and the answerer 53 is ranked third. Therefore, the inquiry is transferred to the respondent 52 who is ranked first.

  In this way, the number of correspondence cases k remains the same as in FIG. 16, but it can be seen that the significant decrease in the proficiency level s of the respondent 53 has affected the decrease in the ranking of the respondent 53.

  As described above, according to the second embodiment, by calculating the sorting reference value a using the correction coefficient γ set in advance for each product even if the products have different difficulty levels of answer processing, The proficiency level s of the respondent can be more accurately reflected in the distribution order. Accordingly, if the inquiry is distributed according to the distribution order assigned based on the distribution reference value a, the inquiry can be transferred to a more optimal respondent.

  In addition, this invention can also be implemented as a program which makes a computer implement the said inquiry distribution method. This program can be stored in various types of storage media, and can be transmitted via a communication medium. Here, the storage medium includes, for example, a flexible disk, a hard disk, a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD, a ROM cartridge, a battery-backed RAM memory cartridge, a flash memory cartridge, a nonvolatile RAM cartridge, and the like. . In addition, the communication medium includes a wired communication medium such as a telephone line, a wireless communication medium such as a microwave line, and the Internet.

  The present invention is not limited to the second embodiment, and can be appropriately changed without departing from the spirit of the present invention. For example, the function F (n) does not need to be set for each product, and the correction coefficient γ can be set for each category such as the type of product and the content of answer processing.

DESCRIPTION OF SYMBOLS 1 Inquirer terminal 2 Reception management part 3 1st respondent evaluation part 4 2nd respondent evaluation part 5 Answerer terminal 6 Inquiry management DB
7 Answer / Quality Evaluation Management Unit 30 Proficiency Level Table 31 Experience Value Table 40 Distribution Tables 51-53 Respondent Terminals 100, 1000 Inquiry Distribution System 101 Evaluation Value Calculation Unit 102 Correction Coefficient Calculation Unit 103 First Respondent Evaluation Unit s Proficiency n Experience value k Number of correspondence a, b Distribution standard value

Claims (19)

An evaluation value calculation unit that calculates an evaluation value indicating the appropriateness of the respondent's answer processing for each category of inquiries;
A correction coefficient calculation unit for calculating a correction coefficient for the evaluation value;
A first respondent evaluation unit that calculates a first distribution reference value that serves as an index for determining a new inquiry distribution destination by correcting the evaluation value by the correction coefficient;
The inquiry distribution system, wherein the correction coefficient calculation unit calculates the correction coefficient so as to decrease the appropriateness as the difficulty of the answer process for each category is higher. The inquiry distribution system according to claim 1, wherein the evaluation value increases as the experience value of the respondent increases. The correction factor is determined by a function of the experience value,
The inquiry distribution system according to claim 1, wherein the function is a function that converges to a constant value at a different convergence speed depending on a difficulty level of answer processing for each category. The first respondent evaluation unit determines the first distribution reference value based on a time taken for the respondent to provide an answer after an inquiry is distributed to the respondent. 4. The inquiry distribution system according to any one of three. Based on the first distribution reference value calculated by the first respondent evaluator and the number of inquiries currently being distributed to the respondent, it becomes an index for determining a new inquiry distribution destination. The inquiry distribution system according to claim 1, further comprising a second respondent evaluation unit that calculates a second distribution reference value for each category. The second response evaluation unit calculates a total of the second distribution reference values calculated for each category, and assigns a distribution rank to the respondent based on the total of the second distribution reference values. The inquiry distribution system according to claim 5. The inquiry distribution system according to claim 1, wherein the category is a product for which the inquiry is made. The inquiry distribution system according to any one of claims 1 to 7, wherein the evaluation value is calculated based on a quality evaluation for an answer provided in the past by the respondent to the inquiry. The inquiry distribution system according to claim 8, wherein the evaluation value is an inquiryer's satisfaction degree with respect to an answer process. Calculate an evaluation value that indicates the appropriateness of respondents' response processing for each category of inquiries,
Calculating a correction coefficient for the evaluation value;
By correcting the evaluation value by the correction coefficient, a first distribution reference value serving as an index for determining a distribution destination of a new inquiry is calculated.
The inquiry distribution method, wherein the correction coefficient is calculated by a computer so as to decrease the suitability as the difficulty of the answer process for each category increases. The inquiry distribution method according to claim 10, wherein the appropriateness is improved as the experience value of the respondent increases. The correction factor is determined by a function of the experience value,
The inquiry distribution method according to claim 10 or 11, wherein the function is a function that converges to a constant value at a different convergence speed in accordance with the difficulty level of answer processing for each category. 13. The first distribution reference value is determined based on a time required from when an inquiry is distributed to the respondent to when the respondent provides an answer. 13. Inquiry distribution method. Based on the first distribution reference value and the number of inquiries currently being distributed to the respondent, a second distribution reference value serving as an index for determining a new inquiry distribution destination is determined for each category. The inquiry distribution method according to claim 11, wherein the inquiry distribution method is calculated. The inquiry distribution according to claim 14, wherein a total of the second distribution reference values calculated for each category is calculated, and a distribution rank is assigned to the respondent based on the total of the second distribution reference values. Minute method. The inquiry distribution method according to claim 11, wherein the category is a product that is the target of the inquiry. The inquiry distribution method according to any one of claims 11 to 16, wherein the evaluation value is calculated based on a quality evaluation for an answer provided in the past by the respondent to the inquiry. The inquiry distribution method according to claim 17, wherein the evaluation value is an inquiryer's satisfaction degree with respect to an answer process. Calculate an evaluation value that indicates the appropriateness of respondents' response processing for each category of inquiries,
Calculating a correction coefficient for the evaluation value;
By correcting the evaluation value by the correction coefficient, a first distribution reference value serving as an index for determining a distribution destination of a new inquiry is calculated.
The correction coefficient is calculated so as to decrease the suitability as the difficulty of the answer process for each category is higher.
A program that causes a computer to execute processing.

Images