JP6584018B2 - Reward prediction calculation device, reward prediction calculation server, reward prediction calculation program, and reward prediction calculation method - Google Patents

Description

  The present invention relates to prediction for setting a reasonable remuneration in a membership system introduction / sales organization.

  There are various forms of membership-based referral and sales organizations, but in the basic form, those who purchased products sold by the headquarters through member referrals joined the organization as new members. It pays a member a bonus (bonus).

  Members do not sell products themselves, but can receive rewards from the headquarters through referrals. In addition, the headquarters company can regard the member organization formed through the introduction as an infrastructure for selling products without stores. For this reason, both are considered to have advantages, and for a period of time it has been widely spread.

  However, in light of these advantages, the headquarters formulates an excessive remuneration plan, and as a result, the management plan is changed frequently in order to avoid a bankruptcy in cases where the bankruptcy or management failure occurs. In many cases, members were worried and the membership organization was dissolved. As a result, it was subject to legal regulations, and the number of headquarters was small.

  The cause of such a situation can be found in a compensation plan that tends to be excessive, but it is not easy for the supervising company to set a compensation plan appropriately. If more members can be quickly organized, the operations of the headquarters can continue even if the remuneration is increased. The more rewards, the more members and hands, but the more rewards, the higher the risk of bankruptcy due to the payment of rewards. On the contrary, if the reward is reduced, the attractiveness of attracting the member is poor, so the number of members tends to decrease.

  Whether the compensation plan of the headquarters is appropriate depends on various factors such as the price and attractiveness of the product itself, the number of members, and economic trends. For this reason, it is difficult to predict how a reward will be determined if an appropriate plan is obtained. There is a strong social demand for the headquarters to formulate an appropriate compensation plan. However, in the past, the actual situation was that the prediction was based on the personal experience of the plan creator, and no prediction with an effective quantitative evaluation was made.

In this situation, the inventors also introduced a new member by one of the members, and these new members introduced a new member and branched into two (branched). In the past, trials and errors have been carried out in a quantitative evaluation of the compensation plan in the member-based introduction sales distribution organization that constitutes the sales organization.
However, the method previously proposed by the inventors introduced a dividend rate (sweeping rate D (n)) in the compensation plan to limit the maximum amount of bonus received in a certain period. It was difficult to create (Patent Document 1).

  Another method is to limit the maximum amount of bonus received at different dividend rates (sweeping rate F (n)), but this proposal also makes it difficult to create a flexible compensation plan (patents). Reference 2).

  Therefore, even if n becomes larger and the hierarchy becomes deeper, the dividend rate always converges, and it has the effectiveness and versatility that contributes to flexibly determining an appropriate and reasonable compensation plan for members from the calculated dividend rate A reward prediction calculation device and prediction method have been desired.

JP 2008-90804 A JP 2013-47959 A

  An object of the present invention is to provide a reward prediction calculation device, a prediction server, a prediction program, and a prediction method for setting an appropriate and reasonable reward plan in a member system introduction sales organization.

<1> In order to achieve the above object, the present invention provides:
A remuneration prediction calculation server that calculates a remuneration amount for members of a member system introduction sales distribution organization and predicts a dividend rate thereof,
It can communicate with the terminal device via the network,
A condition receiving unit that receives transmission of a variable condition for calculating the remuneration amount from the terminal device;
Remuneration decision rule for convergence of the dividend rate, evaluation rule for quantitative evaluation of the dividend rate related to the prediction, change rule for changing the variable condition according to the result of the quantitative evaluation, and variable condition for the reception and change A storage unit for storing
In accordance with the received fluctuation condition or the change condition related to the change, a prediction evaluation is performed in which the reward amount and the dividend rate are predicted based on the reward determination rule and the evaluation rule, and the dividend rate related to the prediction is quantitatively evaluated. A control unit to execute;
A change unit that changes the change condition according to the change rule with the change condition related to the reception as an initial value, and repeatedly executes the prediction evaluation of the control unit;
A decision support unit that communicates with the terminal device to execute display including a dividend rate related to the prediction evaluation and provision of an input unit for a decision command related to the display;
When receiving the determination command transmitted from the terminal device, an output unit that transmits the dividend rate according to the determination command and the variation condition according to the display as determination data;
With
The reward determination rule is:
[1] Each member has arrangement information mapped so as to form a hierarchy that is virtually arranged and filled in each node of the binary tree data structure,
[2] Regarding the amount of compensation for each member,
Based on the total number of products purchased, which is the total number of products purchased by members belonging to lower tiers that branch from the member based on the placement information, a given maximum limit, and a given reward base,
Calculate the provisional compensation amount that is discretized and evaluated by the compensation base for the total number of product purchases for each tier, and calculate the compensation amount so that the provisional compensation amount does not exceed the maximum limit amount,
The provisional compensation amount is
Regarding the remuneration amount of members belonging to the i-th hierarchy of the binary tree data structure having a hierarchy of depth n,
The maximum limit is M yen, the basic reward amount is h yen, the reward base is g, each member purchases one product, the total product purchase is S (i), and the provisional reward is P ( i) Yen, INT {} is an operator that takes an integer value of the number in {}, * is an integration symbol,
P (i) = h * INT {S (i) / g}
Because
The amount of reward determined is
M ≦ P (i) and k + i = n + 1
, The case is classified according to the m-th layer counted from the n-th layer that is the minimum value of k,
[Case 1]:
For members in the hierarchy of i ≦ n + 1−m, the remuneration amount shall be the maximum limit of M yen,
[Case 2]:
For members in the hierarchy of i> n + 1−m, the remuneration amount is calculated by assuming that the remuneration amount is P (i) = h * INT {(2 ^{(n + 1−i)} −2) / g} yen, which is the provisional remuneration amount. Is configured to do.

Because of this configuration, the payout rate always converges even when the hierarchy is infinitely deep, so quantitative evaluation under various variable conditions is provided to the terminal device through the network based on the converged payout rate. be able to.
In addition, a parameter called a reward base is introduced to calculate a provisional reward amount obtained by discretizing and evaluating the total number of product purchases by the reward base. As a result, the total number of product purchases can be rounded down to a stepwise rounded value. Therefore, in calculating the remuneration amount and the payout rate, the trial calculation can be simplified by changing the remuneration number.
Furthermore, old members who are shallow in the membership organization limit the maximum amount of remuneration, and relatively new members who are deep in the hierarchy are rewarded according to the total number of items purchased. Therefore, members who receive relatively high remuneration are limited to the maximum limit, but there is no reduction factor for members who do not reach the maximum limit and do not have many referrals. Is easy to keep, and can be presented to members in advance, so it is easy to accept. In addition, the headquarters can easily evaluate the dividend rate when the maximum limit and the number of rewards are moved simultaneously, so a flexible compensation plan can be considered.
Here, the dividend rate refers to a ratio between a source of dividends such as a sales amount related to the product received by the membership introduction sales distribution organization and a reward amount (the same applies in this specification). This dividend resource includes not only the sales amount of the product itself, but also income that becomes a dividend resource for the organization, such as a membership fee. Further, not only the total sales amount but also a part of the necessary deductions may be used.
In addition, since the converged payout rate corresponding to the changing condition (parameter) is obtained even when the hierarchy is infinitely deep, it is possible to quantitatively evaluate the reward plan.
In addition, since the basic compensation amount is multiplied by the discretized evaluation value to obtain the provisional compensation amount, the dividend amount can be changed with the three main parameters being the maximum limit, the number of compensation bases, and the basic compensation amount. Since it is possible, the compensation plan can be determined flexibly.

<2> In addition, in order to calculate the amount of remuneration for members of the member system introduction sales distribution organization and predict the dividend rate,
A condition input means for receiving an input of a variable condition for calculating the remuneration amount;
A reward determination rule for converging the dividend rate, an evaluation rule for quantitatively evaluating the dividend rate according to the prediction, a change rule for changing the variable condition according to a result of the quantitative evaluation, and the input and change Storage means for storing such variation conditions;
Based on the variable condition related to the input or change, based on the reward determination rule and the evaluation rule, the reward amount and the dividend rate are predicted, and a prediction evaluation is performed to quantitatively evaluate the dividend rate related to the prediction. Control means;
Repetitive means for changing the fluctuation condition according to the change rule with the fluctuation condition related to the input as an initial value, and repeatedly executing the prediction evaluation of the control means;
A decision support means for executing a display including a dividend rate relating to the prediction evaluation, and providing an input means for a decision command relating to the display;
When the determination command is input, a reward prediction calculation program for causing the payout rate according to the determination command and the variation condition according to the display to function as output means for outputting as determination data,
The reward determination rule is:
[1] Each member has arrangement information mapped so as to form a hierarchy that is virtually arranged and filled in each node of the binary tree data structure,
[2] Regarding the amount of compensation for each member,
Based on the total number of products purchased, which is the total number of products purchased by members belonging to lower tiers that branch from the member based on the placement information, a given maximum limit, and a given reward base,
For each tier, calculate the provisional reward amount that is discretized and evaluated by the reward base for the total number of product purchases,
Calculate the compensation amount so that the provisional compensation amount does not exceed the maximum limit amount,
The provisional compensation amount is
Regarding the remuneration amount of members belonging to the i-th hierarchy of the binary tree data structure having a hierarchy of depth n,
The maximum limit is M yen, the basic reward amount is h yen, the reward base is g, each member purchases one product, the total product purchase is S (i), and the provisional reward is P ( i) Yen, INT {} is an operator that takes an integer value of the number in {}, * is an integration symbol,
P (i) = h * INT {S (i) / g}
Because
The amount of reward determined is
M ≦ P (i) and k + i = n + 1
, The case is classified according to the m-th layer counted from the n-th layer that is the minimum value of k,
[Case 1]:
For members in the hierarchy of i ≦ n + 1−m, the remuneration amount shall be the maximum limit of M yen,
[Case 2]:
For members in the hierarchy of i> n + 1−m, the remuneration amount is calculated by assuming that the remuneration amount is P (i) = h * INT {(2 ^{(n + 1−i)} −2) / g} yen, which is the provisional remuneration amount. It is configured to be.

Because of this structure, the payout rate always converges even when the hierarchy is infinitely deep. Therefore, a computer program for quantitative evaluation under various fluctuation conditions based on the converged payout rate is provided. Can do.
In addition, a parameter called a reward base is introduced to calculate a provisional reward amount obtained by discretizing and evaluating the total number of product purchases by the reward base. As a result, the total number of product purchases can be rounded down to a stepwise rounded value. Therefore, in calculating the remuneration amount and the payout rate, the trial calculation can be simplified by changing the remuneration number.
Furthermore, old members who are shallow in the membership organization limit the maximum amount of remuneration, and relatively new members who are deep in the hierarchy are rewarded according to the total number of items purchased. Therefore, members who receive relatively high remuneration are limited to the maximum limit, but there is no reduction factor for members who do not reach the maximum limit and do not have many referrals. Is easy to keep, and can be presented to members in advance, so it is easy to accept. In addition, the headquarters can easily evaluate the dividend rate when the maximum limit and the number of rewards are moved simultaneously, so a flexible compensation plan can be considered.
In addition, since the converged payout rate corresponding to the changing condition (parameter) is obtained even when the hierarchy is infinitely deep, it is possible to quantitatively evaluate the reward plan.
In addition, since the basic compensation amount is multiplied by the discretized evaluation value to obtain the provisional compensation amount, the dividend amount can be changed with the three main parameters being the maximum limit, the number of compensation bases, and the basic compensation amount. Since it is possible, the compensation plan can be determined flexibly.

<3> In addition,
In the reward prediction calculation method that calculates the amount of compensation for members of the member system introduction sales distribution organization and predicts the dividend rate,
A condition input step for inputting a variable condition for calculation of the reward amount;
A first storage step of storing a reward determination rule for converging the dividend rate, an evaluation rule for quantitatively evaluating the dividend rate related to the prediction, and a change rule for changing the variation condition according to the result of the quantitative evaluation When,
A second storage step for storing a variation condition related to the input and change;
Based on the variable condition related to the input or change, based on the reward determination rule and the evaluation rule, the reward amount and the dividend rate are predicted, and a prediction evaluation is performed to quantitatively evaluate the dividend rate related to the prediction. Control steps;
The variable condition according to the input is changed to an initial value, the variable condition is changed according to the change rule, and the display including the payout rate related to the prediction evaluation is repeatedly executed to repeatedly execute the prediction evaluation of the control step, A decision support step for providing the input means;
When the determination command is input, an output step of outputting the dividend rate according to the determination command and the variation condition according to the display as determination data;
A reward prediction calculation method including
The reward determination rule is:
[1] Each member has arrangement information mapped so as to form a hierarchy that is virtually arranged and filled in each node of the binary tree data structure,
[2] Regarding the amount of compensation for each member,
Based on the total number of products purchased, which is the total number of products purchased by members belonging to lower tiers that branch from the member based on the placement information, a given maximum limit, and a given reward base,
Calculate the provisional compensation amount that is discretized and evaluated by the compensation base for the total number of product purchases for each tier, and calculate the compensation amount so that the provisional compensation amount does not exceed the maximum limit amount,
The provisional compensation amount is
Regarding the remuneration amount of members belonging to the i-th hierarchy of the binary tree data structure having a hierarchy of depth n,
The maximum limit is M yen, the basic reward amount is h yen, the reward base is g, each member purchases one product, the total product purchase is S (i), and the provisional reward is P ( i) Yen, INT {} is an operator that takes an integer value of the number in {}, * is an integration symbol,
P (i) = h * INT {S (i) / g}
Because
The amount of reward determined is
M ≦ P (i) and k + i = n + 1
, The case is classified according to the m-th layer counted from the n-th layer that is the minimum value of k,
[Case 1]:
For members in the hierarchy of i ≦ n + 1−m, the remuneration amount shall be the maximum limit of M yen,
[Case 2]:
For members in the hierarchy of i> n + 1−m, the remuneration amount is calculated by assuming that the remuneration amount is P (i) = h * INT {(2 ^{(n + 1−i)} −2) / g} yen, which is the provisional remuneration amount. It is configured to be.

Because of this structure, the payout rate always converges even when the hierarchy is infinitely deep, so rewards that allow the computer to perform quantitative evaluation under various variable conditions based on the converged payout rate A prediction calculation method can be provided.
In addition, a parameter called a reward base is introduced to calculate a provisional reward amount obtained by discretizing and evaluating the total number of product purchases by the reward base. As a result, the total number of product purchases can be rounded down to a stepwise rounded value. Therefore, in calculating the remuneration amount and the payout rate, the trial calculation can be simplified by changing the remuneration number.

Furthermore, old members who are shallow in the membership organization limit the maximum amount of remuneration, and relatively new members who are deep in the hierarchy are rewarded according to the total number of items purchased. Therefore, members who receive relatively high remuneration are limited to the maximum limit, but there is no reduction factor for members who do not reach the maximum limit and do not have many referrals. Is easy to keep, and can be presented to members in advance, so it is easy to accept. In addition, the headquarters can easily evaluate the dividend rate when the maximum limit and the number of rewards are moved simultaneously, so a flexible compensation plan can be considered.
In addition, since the converged payout rate corresponding to the changing condition (parameter) is obtained even when the hierarchy is infinitely deep, it is possible to quantitatively evaluate the reward plan.
In addition, since the basic compensation amount is multiplied by the discretized evaluation value to obtain the provisional compensation amount, the dividend amount can be changed with the three main parameters being the maximum limit, the number of compensation bases, and the basic compensation amount. Since it is possible, the compensation plan can be determined flexibly.

  In the above invention, “output” includes not only display and printing, but also the case of giving to other programs and devices as data.

  Further, “input” refers to that which passes through an interface for giving at least a variation condition to the CPU. “Input unit” and “input means” are not only interfaces with humans such as keyboards, mice, and voice input devices, but also interfaces with other programs such as interface circuits and interface programs, and other computers. Including those that take In the embodiment, the keyboard 10 corresponds to this.

  The “program” includes not only a program that can be directly executed by a computer but also a program that can be executed by being installed on a hard disk or the like. It also includes the case where it is compressed or encrypted.

  In the present invention, the payout rate always converges even if the hierarchy becomes infinitely deep for the reward distribution of the member system introduction sales organization. Therefore, quantitative evaluation under various variable conditions is stably performed based on the converged payout rate. be able to. In addition, since the maximum amount of remuneration is determined according to the depth of each member in the organization, a rational remuneration plan that is easy for members to accept can be considered.

FIG. 1 is a block diagram showing a functional configuration of a reward prediction calculation apparatus 1 according to the first embodiment of the present invention. FIG. 2 is an example of a condition input screen generated by the condition input unit. FIG. 3 is a block diagram illustrating the data structure of the rules stored in the storage unit. FIG. 4 is an image diagram showing a decision support screen. FIG. 5 is a block diagram showing the data structure of the decision data. FIG. 6 is a diagram illustrating an example of a hardware configuration of the reward prediction calculation apparatus 1. FIG. 7 is an explanatory diagram of a binary tree data structure of the membership sales organization according to the present invention. FIG. 8 is an explanatory diagram for calculating the payout rate based on the binary tree data structure. FIG. 9 is a flowchart showing the operation of the reward prediction calculation apparatus. FIG. 10 is an overall view of a reward prediction calculation system 1000 according to the second embodiment of the present invention. FIG. 11 is a block diagram showing a functional configuration of a reward calculation prediction server 1001 according to the second embodiment of the present invention. FIG. 12 is a flowchart showing the operation of the server device 1001 and the operations of the terminal devices 1002 to 1004 corresponding thereto.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
(Organizational expansion of member system introduction sales organization, binary tree data structure)
Before describing each embodiment, a binary tree data structure to which a membership introduction / sale organization for which the reward prediction calculation apparatus of the present invention performs prediction is mapped will be described with reference to FIG. Membership referral sales organizations are mapped to a binary tree data structure for reward calculations. That is, each member is virtually arranged at each node of the binary tree data structure to form a hierarchy, and a new member is virtually arranged at any location in this data structure. Black circles shown in FIG. 7 indicate members arranged at each node of the binary tree data structure. FIG. 7 shows that members are filled from the first layer (root node) through the i-th layer to the n-th layer. This shows the sales organization developed. Some black circles are omitted.

(Operation of a member-based referral sales organization)
In the member-based referral and sales organization, the supervising company sells products in response to referrals from members and pays the introduced members according to the remuneration decision rule.
By purchasing the products handled by the headquarters, the member becomes a member of the sales organization and occupies one node of the binary tree data structure. By chaining such introductions, the purchase of the product is promoted, and more members are placed below the member's node, so that more rewards are received.
In this way, since the member receives a reward through the introduction, the motivation to introduce is generated. Similarly, each member who receives a referral can receive a reward through the referral when introducing a new member, so that the organization is similarly developed in the following manner.

  The member who joined first (also referred to as “top”) is virtually arranged in the “first layer (or“ root node ”)” of the binary tree data structure of this member system introduction sales organization. This state can be expressed as “the first member is in the first layer”. A member who receives an introduction from the first member and newly joins the organization is placed below the “second layer”. Assuming that two people are introduced, the introduced members are arranged on two branches in the second layer. Similarly, the binary tree data structure of this member system introduction / sales organization is divided into two branches at each node of each layer, and is expanded to a lower layer. As shown in FIG. 7, two introduced members are arranged at the second layer node, and four members are arranged at the third layer node.

  Of course, each member is not limited to two referrals. Members may introduce 3 or more new members. However, even in this case, the organizational development of this member system introduction sales organization maintains the “binary tree data structure”, and as shown in FIG. It is arranged at any location in the “data structure” and is expanded so as to form a hierarchy by branching from one node into two.

(Assumption 1)
In the reward prediction handled in the present invention, it is assumed that the member system introduction sales organization is mapped to such a binary tree data structure and is filled with members up to the n-th layer (Premise 1). Hereinafter, for the sake of convenience, with respect to the branches that are branched into two from each node, the left branch or the lower branch is referred to as the left, and the right branch or the upper branch is referred to as the right.

  The “binary tree data structure” described above is branched into two, so it is well suited to binary notation. When “0” is assigned to the left branch and “1” is assigned to the right branch, the position of the node where each member is arranged is uniquely determined by such a binary notation. For example, the binary number 101 corresponds to the second node from the left of the third layer of top (first layer) → left (second layer) → right (third layer). For this reason, the position where each member is arranged can be easily set by a logical operation to be handled in the database, and the prospect is good.

(Assumption 2)
This reward is determined by counting the introduction results (product sales results) over a certain period (premise 2). Based on such a “premise”, the amount of remuneration that the member should receive during the period is determined by the following “reward determination rule”, and the dividend rate is calculated.

(First embodiment: apparatus)
(Overview of device functional configuration and operation)
FIG. 1 is a block diagram showing a functional configuration of a reward prediction calculation apparatus 1 according to the first embodiment of the present invention. It is composed of a functional input unit such as a condition input unit 11, a storage unit 12, a control unit 13, a repetition unit 14, a decision support unit 15 and an output unit 16. The functional configuration unit shown here includes preinstalled firmware and the like The control program is executed by a processor of a predetermined circuit, and is realized by cooperating with various mounted devices. These programs are recorded on a computer-readable recording medium, read from the recording medium by the processor, and triggered by a user operation or reception of a signal from a device constituting the system. Executed.

  The condition input unit 11 generates a screen that accepts an input from the user such as a variation condition for performing the reward prediction calculation and a calculation command.

(Condition input screen)
FIG. 2 is an example of a condition input screen generated by the condition input unit 11. The condition input screen 20 includes a variation condition input field 21 and a calculation command input button 22.
The variable condition input field 21 includes a one-side reward base input field 23 for inputting a one-side reward base, a maximum limit input field 24 for inputting a maximum limit amount, and a basic reward amount field 25 for inputting a basic reward amount.
The calculation command input button 22 is an input button image for executing an input of a calculation command by receiving the calculation command by an action by a pointing device.

  The storage unit 12 stores a reward determination rule, an evaluation rule that quantitatively evaluates the dividend rate related to the reward prediction calculation, a change rule that changes the variable condition according to the result of the quantitative evaluation, and a variable condition that relates to the input and change. To do.

(Evaluation rules and change rules)
FIG. 3 is a block diagram illustrating the data structure of the rules stored in the storage unit. The structure is a rule ID 31, a rule name 32, and a rule 33.
The rule data shown in the column 34 where the rule ID is R1 is an evaluation rule that quantitatively evaluates the payout rate and gives an evaluation value V. Here, when the payout rate is less than 50%, the evaluation is “not attractive” When the dividend rate is 50% or more and 95% or less, an “appropriate” evaluation is given, and when the dividend rate exceeds 95%, a “risk” evaluation is given.
The evaluation rule of the present invention is not limited to the evaluation rule shown here. It can be changed as appropriate according to the nature of the member system introduction and sales organization to be applied.

The rule data shown in the column 35 of the rule ID R2 is a change rule for changing from the input variation condition. Here, when the evaluation value V shown above is “not attractive”, two change fluctuations of “a + 1” obtained by adding 1 to one-side reward radix “a” which is one of the fluctuation conditions and “a + 2” obtained by adding 2 Generate a condition. When the evaluation value V is “appropriate”, two change variation conditions of “a−1” obtained by subtracting 1 from the one-side reward base “a” and “a + 1” obtained by adding 1 are generated. When the evaluation value V is “risk”, two change variation conditions are generated: “a-1” obtained by subtracting 1 from the one-side reward radix “a” and “a-2” obtained by subtracting 2 from the value.
The change rule of the present invention is not limited to the change rule shown here. As a rule for changing the one-side reward cardinal number a, not only a rule of another method but also a rule for changing the basic reward amount h, the maximum reward limit M, and the like can be adopted.
For example, a rule for a method in which a maximum percentage amount M1, M2,... Mn is generated by multiplying the inputted maximum reward limit M0 by a predetermined percentage multiple times and applied to the change variation condition. The maximum amount of remuneration M0 that has been entered is subtracted several times in advance to generate the maximum amount of change M1, M2,. It can be adopted as appropriate according to the nature of the membership system.

  When the condition input unit 11 accepts the input of the change condition, the control unit 13 stores it in the change condition table file in the storage unit 12, searches the reward determination rule R1 in the storage unit 12, and changes based on the reward determination rule R1. Dividend resources such as a reward amount and sales amount corresponding to the conditions are calculated, and a dividend rate F is predicted from these. Further, an evaluation rule in the storage unit 12 is searched, and an evaluation value V of quantitative evaluation corresponding to the payout rate related to the prediction calculation is calculated based on the evaluation rule R1.

  The repetition unit 14 searches for the change rule R2 in the storage unit 12, executes the change of the change condition corresponding to the evaluation value V based on the change rule R2, and sends the changed change condition and the recalculation instruction to the control unit 13. The control unit 13 is repeatedly executed to calculate the dividend resources such as the remuneration amount and the sales amount corresponding to the changed variable condition, and to predict and calculate the dividend rate F from this.

The decision support unit 15 displays a payout rate F related to the prediction calculation, generates a decision support screen 400 provided with an input button for a decision command related to the display, and displays it on the display of this apparatus. This screen can accept an input of a determination command or the like by an action by an input device such as a keyboard described later.
As a result, the user can easily input his / her decision while viewing the information for decision determination such as the payout rate F which is a material for the determination, and thus the user's decision can be supported.
Here, an input button display for accepting an action by the input device is adopted as the input means, but the input means according to the present invention is not limited to this. An input field for receiving an input of a determination command or the like using a keyboard or the like can be appropriately employed.

(Decision support screen)
FIG. 4 is an image diagram showing a decision support screen.
Here, the change condition display 401 indicates the input change condition, and the payout rate 1 indicated by the payout rate display 404 indicates the payout rate calculated under this change condition. The decision 1 button display 407 functions as a button for inputting a decision command using the payout rate 1 and the variation condition related thereto as decision data.

The change condition display 402 shows the changed change condition, and the payout rate 2 shown in the payout rate display 405 shows the payout rate calculated under this change condition. The decision 2 button display 408 functions as a button for inputting a decision command using the payout rate 2 and the variation condition related thereto as decision data.
The change condition display 403 shows the changed change condition, and the payout rate 3 shown in the payout rate display 406 shows the payout rate calculated under this change condition. The decision 3 button display 409 functions as a button for inputting to the output unit 16 a decision command using the payout rate 3 and the variation condition related thereto as decision data.

The change condition input field display 410 is for inputting a change condition related to recalculation when the user makes a decision that none of the payout rates 1 to 3 repeatedly calculated under the displayed change conditions 401 to 403 is desirable. It is a column of. A desired variation condition can be input by a device such as a keyboard 603.
The recalculation button display 411 functions as a button for inputting to the control unit 13 a recalculation command that performs prediction calculation according to the re-input variation conditions 412 to 414 and repeats calculation such as quantitative evaluation of the dividend rate related to the prediction calculation. To do.

When receiving the input of the determination command, the output unit 16 outputs the payout rate related to the display and the variation condition as a basis thereof as determination data, and stores it in the calculation result file in the storage unit 12.
FIG. 5 is a block diagram showing the data structure of the decision data. Here, the data structure of data ID 51, one-sided reward card number 52, maximum limit 53, basic reward 54, and dividend rate 55 is adopted.

(Hardware configuration)
In FIG. 6, an example of the hardware constitutions of the reward prediction calculation apparatus 1 is shown. In FIG. 6, a memory 601, a display 602, a keyboard 603, a hard disk (HDD) 604 as a storage unit, and a DVD / CD-ROM drive 65 are connected to a CPU 600.
The hard disk 604 stores a reward prediction program 607, a reward determination rule 608, an evaluation rule 609, a change rule 610, an OS (operating system) 611, a table file 612 for storing variation conditions, and the like. A calculation result file 613 based on the payout rate is stored. The reward prediction program 607, the reward determination rule 608, the evaluation rule 609, and the change rule 610 are installed from the DVD or CD-ROM 606 via the DVD / CD-ROM drive 605. Other media may be installed by another interface.
In addition, the hard disk 604 is adopted as a storage device here, but the present invention is not limited to this, and any storage medium that can read and write to other storage media connected to the CPU 600 or the like may be used, such as an optical disk, SSD, or USB memory. There may be.

(Variation condition table file)
Table 1 shows the configuration of the variation condition table file 612.

(Reward decision rule)
The remuneration decision rule calculates the remuneration using the parameters shown in Table 1, and the product purchase is the total number of products purchased by members belonging to a lower hierarchy that branches from the member based on the member arrangement information. Based on the total number, the given maximum amount, and the given reward base, the product purchase total quantity is discretized and evaluated by the reward base for each hierarchy so that the calculation of the dividend rate converges. A provisional compensation amount is calculated, and the compensation amount is determined so that the provisional compensation amount does not exceed the maximum limit.

(Remuneration decision rule parameters)
The “reward determination rule” has a variable condition represented by the variable condition table of Table 1 as a parameter (variable). As will be described later, as a function having such parameters, the payout rate of members virtually arranged in the i-th layer having a depth of the layer up to the n-th layer is expressed by a function F (i) ( 1 ≦ i ≦ n).

The items (parameters) in the variation condition table of Table 1 will be described. Excluding the aggregation period, each parameter is an integer.
The “number of members branched from each member” (number 1) employs a binary tree data structure, so there is one for each of the left branch and the right branch, for a total of two members.
The “number of products purchased by each member” (number 2) is set to one when one member purchases one product.
“One-sided reward card” (No. 3) determines the number of reward units u by discretizing (quantizing) the total number of items purchased by the member in order to calculate the reward amount as will be described later. It is a parameter to do. “One side” means one side of the left and right branches branching from each member, and since the binary tree data structure is actually adopted, this double is the reward base g used in the calculation. In addition, * indicates that a product is taken (hereinafter the same in this specification). This is a basic parameter that has a great influence on the calculation of the dividend rate.
The “aggregation period” (number 4) is a period for calculating a reward, and the reward is calculated based on the number of items purchased in this period.
“Maximum amount” (number 5) refers to the maximum amount of remuneration received by one member during the “aggregation period”. This is M (yen).
The “basic remuneration amount” (number 6) is the remuneration amount to be paid for each remuneration unit u, and the basic remuneration amount multiplied by the remuneration unit is a provisional remuneration amount described later. This is h (yen).
“Product unit price” (number 7) is the price of one product purchased by a new member, and this is A (yen).
“Admission fee” (No. 8) is paid to the headquarters company when a new member joins. This is b (yen).
The “hierarchy depth” (number 9) refers to the number of the hierarchy counted from the root node (top) when the member-based introduction sales distribution organization is expanded into the binary tree data structure, and this is n.

Next, the outline of the reward determination rule is as follows.
The remuneration decision rule is a condition for a member to receive a remuneration, and divides a total product purchase quantity S (i), which will be described later, by a remuneration base g (= 2 * a), and uses the quotient as a remuneration unit u (i). The basic remuneration amount per unit is set to h yen, and the product of the basic remuneration amount and the remuneration unit u (i) is calculated as the provisional remuneration amount P (i). That is, the total product purchase quantity S (i) is digitized (quantized) by the reward base g (= 2 * a) and calculated in a stepwise manner. In this way, when the first step cannot be reached, that is, when the total product purchase quantity S (i) falls below the reward card number g, the reward amount becomes 0 and no reward is paid. The total product purchase quantity S (i) is not directly proportional to the reward amount, but the total product purchase quantity is discretized (quantized) and rounded down. Become. Such an evaluation is performed at each level.
In each step, the total amount of product purchases S (i) is set to pay a basic remuneration amount h yen for every integral multiple of the remuneration base g, but the size of this step is not an integral multiple of the remuneration base g. Or may be changed at each step.

(Remuneration for i-th layer members)
Furthermore, it demonstrates in detail along a reward determination rule.
Since the binary tree data structure shown in FIG. 7 is a hierarchical structure consisting of n layers, it starts from the first layer, and below it is divided into two layers, two branches to the left and right, and two branches from the second layer to the left and right. The third layer is composed of the i-th layer branched to the left and right from the (i-1) layer, and the n-th layer branched to the left and right from the (n-1) layer. Is done.
Then, as shown in the fluctuation condition table of Table 1, each member purchases one item, so the total number of items purchased by the member group subordinate to the member in the i-th layer (that is, subordinate) If S (i) is not included in the i-layer's own member's own product purchase number), S (i) is equal to the number of introducers arranged at a lower level branching from the member's node. Then, since the total number of members of the binary tree data structure starting from the i-th layer and ending at the n-th layer is subtracted 1 from 2 ^{n- (i-1)} -1, (Equation 1) is obtained. . The total number of members of the binary tree data structure starting from the first layer and ending at the nth layer is 2 ⁰ +2 ¹ +2 ² +... +2 ⁽ⁿ⁻¹⁾ = 2 ⁿ −1.

S (i) = (2 ^{n− (i−1)} −1) −1 (pieces)
= 2 ^{n + 1-i} -2 (pieces) (Formula 1)

As a method of determining the amount of reward, a basic amount of reward is determined in advance, and the amount of reward can be determined so as to have a linear relationship with the total number of product purchases by multiplying the product by the total number of product purchases.
However, if the threshold of remuneration is raised a little and a certain number is set as a threshold and the remuneration is paid when the threshold is exceeded, it is expected that the incentive to introduce up to the threshold number will increase. If the threshold can be made variable, the incentive and reward for introduction can be determined more delicately. For this reason, the “total product purchase quantity” is discretized (quantized) by the reward radix g and converted into reward units u (i) for evaluation.
Then, as shown in (Expression 2), the reward unit u (i) is an integer obtained by dividing S (i) = 2 ^{n + 1−i} −2) (pieces) in (Expression 1) by the reward radix g. Can be represented as a part. Here, if the reward base g is 2 * a which is twice the one-side reward base a, the reward unit u (i) is an integer part obtained by dividing S (i) by 2 * a.

u (i) = INT {S (i) / g} (Formula 2)
= INT {S (i) / (2 * a)} (unit)

Here, INT {} is an operator that extracts an integer part of a numerical value in parentheses (hereinafter the same). By determining the basic reward amount for the reward unit u (i) and taking the product of these, the reward amount can be determined.
Considering this, the remuneration amount P (i) that one member of the i-th layer will receive during this counting period is expressed in the basic remuneration amount h (yen) as shown in (Equation 3). The reward unit u (i) of 2) is multiplied.

        P (i) = h * u (i) (yen) (Formula 3)

(Introduction of maximum limit)
The reward P (i) of one member of the i-th layer is obtained as shown in (Equation 3) as described above. When n is large, the amount of reward increases exponentially, so the dividend rate is There is concern about divergence. Therefore, in order not to diverge, the remuneration amount P (i) is set as a provisional remuneration amount, and the maximum limit amount M (yen) is introduced as the upper limit amount.
Since this maximum limit M limits the reward, it is desirable for the member to be fair and easy to accept. Therefore, first, the provisional remuneration amount P (i) is calculated, and then only the amount exceeding the maximum limit is limited to the maximum limit.
Assuming that the layer where the provisional reward amount P (i) reaches the maximum limit M (the layer that initially exceeds the maximum limit amount) is the mth layer from the nth layer, m is the following inequality (formula The minimum value of k satisfying 4) can be obtained.
However, since i is counted from the first layer and k is counted from the nth layer, there is a relationship of i + k = n + 1 between i and k, and considering (Equation 1), as shown in (Equation 4): The minimum value of k obtained is determined by three parameters M, h, and a (g), and does not depend on n. That is, m is a constant unrelated to n.

M ≦ P (i) = h * u (i)
= H * INT {S (i) / g)}
= H * INT {(2 ^{n + 1-i} -2) / g}
= H * INT {(2 ^k -2) / g)} (Formula 4)

In this way, m, which is the layer that reaches the maximum limit M, is obtained, and the following cases are divided to determine the member remuneration for each level. The members from the first layer to the (n + 1) -m layer limit the reward to the maximum limit M yen.
[Case 1]:
For members with i ≦ n + 1−m, the maximum remuneration is M yen.
[Case 2]:
For members with i> n + 1−m, the reward is a provisional reward amount P (i) = h * INT {(2 ^{n + 1−i} −2) / g} yen.
When the remuneration number g is twice the one-side reimbursement number a, members in both lower layers are placed in a well-balanced manner so that the number of items purchased by members belonging to the left and right branches of the member node in the binary tree data structure is a Match what is being done. Also, discretization is intuitive and easy to accept.
In accordance with such a case division, the total amount B of remuneration during the aggregation period of the entire organization is calculated separately for case 1 and case 2 below. In case 1, the reward amount is uniformly limited to the maximum amount M regardless of the layer to which the member belongs. In Case 2, the remuneration amount of one member belonging to the i-th layer is the same amount as the provisional remuneration amount P (i), and the remuneration amount differs depending on which layer it belongs to.

(Total remuneration BM in Case 1)
In case 1, as shown in FIG. 8, the reward of members from the (n + 1-m) th layer to the top first layer is a uniform maximum M yen. Therefore, as shown in the following (Formula 5), the total amount BM of case 1 may be obtained by multiplying the M yen by the total number of members from the first layer to the (n + 1-m) th layer.

(Total compensation BR in Case 2)
In case 2, the reward amount P (i) of each member of each layer does not reach the maximum limit M. The highest member of the binary tree data structure of Case 2 belongs to the (n−m + 2) th layer as shown in FIG. Therefore, the total remuneration BP (yen) of all members belonging to this member and its subordinate (subordinate) groups is calculated by adding all the remuneration amounts of each member from the (n−m + 2) layer to the n layer. 8 (each hatched portion in FIG. 8), the same as Equation 5 is considered, and the following Equation 6 is obtained, and the second Equation is derived from Equation 2 and Equation 3. When this is rewritten as nm−2 = 1, n = m−1, n−i + 1 = m−1, and i−1 = i− (n−m + 2).

As shown in FIG. 8, the BP obtained by Equation 6 is the total remuneration of one member belonging to the (n−m + 2) layer and all members belonging to the lower hierarchy branched from the member. Therefore, when this is integrated by the number of members belonging to the (n−m + 2) layer, 2 ^{n−m + 1} , the total remuneration BR of Case 2 is obtained as shown in (Expression 7). In FIG. 8, the hatched portions in case 2 are displayed in an overlapping manner, but this does not indicate that there is common data, and each hatched portion is independent in the binary tree data structure (another Wood).

BR = BP * 2 ^{n-m + 1} (Formula 7)

  Then, the total remuneration B of all members during the aggregation period is the sum of the total remuneration BM of 1 in the case of (Formula 5) and the total remuneration BR of 2 in the case of (Formula 7). It is expressed by the following (formula 8).

B = BM + BR
= M * (2 ^{n−m + 1} −1) + BP * 2 ^{n−m + 1}
= (M + BP) * 2 ^{n-m + 1} -M (Formula 8)

  Further, the total receipt amount I of the controlling company during the totaling period including the product unit price A and the admission fee (registration fee) b can be expressed by the following (formula 9).

I = A * (2 ⁿ −1) + b * 2 ⁿ⁻¹ = (2 * A + b) * 2 ⁿ⁻¹ −A (Equation 9)

  Accordingly, since (Equation 8) divided by (Equation 9) is the payout rate F (n) of the controlling company, the payout rate F (n) can be calculated as (Equation 10) shown below.

F = B / I
= {(M + BP) * 2 ^{n-m + 1} -M} / {(2 * A + b) * 2 ^n-1 -A} * 100 (%) (Formula 10)

(Convergence of dividend rate F (n))
It shows that the payout rate F (n) calculated in this way always converges. Regarding (Expression 10), the limit value where n is infinite is as shown in the following (Expression 11).

limF (n) = (M + BP) / {2 ^m−2 * (2 * A + b)} * 100 (%) (n → ∞) (Formula 11)

  The right side of (Expression 11) indicates that when n becomes infinite, it converges to a constant value unrelated to n. Therefore, if the dividend amount is determined in accordance with the reward determination rule shown in Case 1 and Case 2, (Equation 11) will not diverge, so theoretically if the dividend rate is determined from such a dividend amount, There is no bankruptcy. Therefore, since the calculated payout rate F (n) is a definite value, it can be seen that it is a highly reliable and extremely powerful index of payout. There have been no proposals for determining such a dividend amount, and it has been shown that reward prediction based on such a reward determination rule is a very powerful method.

(Calculation process)
FIG. 9 is a flowchart showing the operation of the reward prediction calculation apparatus. Hereinafter, the process flow will be described.

  (S901) A variation condition is input from an input device such as the keyboard 603. The input variation condition is stored in the hard disk (HDD) 604.

  (S902) A calculation command is input from an input device such as the keyboard 603, and calculation of the payout rate is started.

  (S903) The input fluctuation condition is read from the HDD 604, and the predicted total sales that are the predicted reward and the dividend source are calculated according to the reward determination rule described later. Next, the predicted payout rate is calculated by dividing the predicted reward by the predicted total sales.

  (S904) A value for quantitative evaluation is generated from the calculated predicted payout rate according to the evaluation rule described above.

  (S905) The change condition is changed with respect to the generated quantitative evaluation value according to the change rule described above, and the change change condition is generated.

  (S906) Based on the generated change variation condition, according to the reward determination rule described later, the predicted total sales that are the predicted reward and the dividend source are repeatedly calculated. Next, the predicted payout rate is repeatedly calculated by dividing the predicted reward by the predicted total sales.

  (S907) The display unit 602 displays the change condition and the change change condition, the corresponding payout rates, the determination button corresponding to them, and the change condition input field and the recalculation button used for recalculation.

  (S908) The user browses the display of the change condition, the change change condition, and the corresponding dividend rate, and decides whether or not there is a desired result, and if so, whether to recalculate do.

  (S909) If there is no desired result, enter the fluctuation condition related to recalculation in the fluctuation condition input field, add an action with the input device to the recalculation button, send a recalculation instruction, and from (S903) Repeat the calculation operation.

  (S910) If there is a predicted payout rate having a desired value, an action is applied to the determination button related to the desired payout rate by the input device, and a determination command is transmitted.

  (S911) The payout rate, which is a desired value, and its fluctuation condition are output to the HDD 64 and recorded.

(Second embodiment: server)
In the first embodiment, a database in which a plurality of rules, programs, and the like are recorded is stored in an apparatus that performs processing. Not limited to this, it is conceivable that these databases are recorded in another device or accessed via a communication line.
In the second embodiment, the reward determination rule, the evaluation rule, and the change rule are implemented on the server as an arithmetic program. In this case, since the variation condition is transmitted from the terminal device to the server and the calculation result is transmitted to the terminal device, the dividend condition can be easily obtained by changing the variation condition by an operation from the terminal device.
By quantitatively grasping the difference between the dividend rate obtained in this embodiment and the dividend rate based on actual results, it is possible to provide a more rational compensation plan and manage the compensation plan according to the situation. Will.

(Processing server)
A reward prediction calculation system 1000 according to this embodiment is shown in FIG. In this embodiment, a server apparatus (processing server) 1001 and terminal apparatuses 1002, 1003, and 1004 are connected via the Internet 1005. The server device 1001 stores a reward prediction calculation program, a reward determination rule, an evaluation rule, and a change rule, and functions as a web server (cloud server). The terminal devices 1002, 1003, and 1004 can use the reward prediction calculation function by accessing the server device 1001.

FIG. 11 is a block diagram showing a functional configuration of a reward calculation prediction server 1001 according to the second embodiment of the present invention. The function receiving unit 1101 includes a storage unit 1102, a control unit 1103, a repetition unit 1104, a decision support unit 1105, and an output unit 1106.
The functional configuration unit shown here is realized by executing a control program such as firmware incorporated in advance by a processor of a predetermined circuit and cooperating with various devices mounted. These programs are recorded on a computer-readable recording medium, read from the recording medium by the processor, and triggered by a user operation or reception of a signal from a device constituting the system. Executed.

(Overview of functional configuration and operation)
The condition receiving unit 1101 generates a condition input screen that accepts inputs such as fluctuation conditions and calculation instructions for performing reward prediction calculation, transmits the condition input screen to the user's terminal device, and receives input from the user.

(Condition input screen)
The condition input screen is the same as the condition input screen generated by the condition input unit in the first embodiment.

  The storage unit 1102 stores a reward determination rule, an evaluation rule that quantitatively evaluates the dividend rate related to the reward prediction calculation, a change rule that changes the variable condition according to the result of the quantitative evaluation, and a variable condition that relates to the input and change. To do.

(Evaluation rules and change rules)
The data structure of the rules stored in the storage unit is the same as the data structure in the first embodiment.

  The contents of the evaluation rule and the change rule are the same as the contents of the rule in the first embodiment.

  When the condition receiving unit 1101 receives the input of the variation condition, the control unit 1103 stores the variation condition table file in the storage unit 1102, searches for a reward determination rule in the storage unit 1102, and sets the variation condition based on the reward determination rule. Dividend resources such as the corresponding remuneration amount and sales amount are calculated, and the dividend rate F is predicted from these. Further, an evaluation rule in the storage unit 1102 is searched, and an evaluation value V of quantitative evaluation corresponding to the payout rate related to the prediction calculation is calculated based on the evaluation rule.

  The repetition unit 1104 searches for the change rule in the storage unit 1102, executes the change of the change condition corresponding to the evaluation value V based on the change rule, and transmits the changed change condition and the recalculation instruction to the control unit 1103. Then, the control unit 1103 repeatedly executes the operation of calculating the dividend resources such as the remuneration amount and the sales amount corresponding to the changed variable condition and predicting the dividend rate F from this.

  The decision support unit 1105 displays a dividend rate relating to the prediction calculation, generates a decision support screen provided with an input button for a decision command relating to the display, transmits the decision support screen to the user's terminal device, To display. This screen is the same as the screen displayed on the display of the reward prediction calculation apparatus in the first embodiment.

When the output unit 1106 receives a determination command transmitted by an operation of a determination button or the like on the terminal device, the output unit 1106 outputs a payout rate related to the determination button and a variation condition as a basis thereof as determination data, and calculates in the storage unit 1102 Store in the result file.
The data structure of the decision data is the same as the data structure in the first embodiment.

Since the terminal devices 1002, 1003, and 1004 are clients, it is not necessary to store a reward calculation prediction program or the like or perform a reward prediction calculation. At least the variable condition is transmitted to the server 1001 to calculate the reward prediction calculation. It is only necessary to be able to receive and record / hold the results.
Although three terminal devices are described here, the number of terminal devices of the present invention is not limited to this, and can be changed as appropriate according to the number of users and the like.

(Variation conditions)
The variation conditions are the same as those shown in Table 1 of the first embodiment.

(Processing flow)
FIG. 12 is a flowchart showing the operation of the server device 1001 and the operations of the terminal devices 1002, 1003, 1004 corresponding thereto.

  At the start, the server device 1001 receives access from the terminal devices 1002, 1003, and 1004, transmits the user identifier input to the terminal device, and permits the login (not shown). At that time, the control unit 1103 may check the user identifier information stored in the storage unit 1102 in advance and allow the control. For example, “0102” may be transmitted as a user identifier, and at that time, a password may be used together to ensure security.

(S1201) A condition input screen is displayed for the terminal device to which login is permitted.

(S1202) Next, the user performs input and transmission operation of the variation condition on the condition input screen, and transmits the variation condition and the calculation command from the terminal device to the server device.

(S1203) The server device calculates the predicted total sales, which are the predicted reward and the dividend resource, according to the received variable condition according to the reward determination rule. Next, the predicted payout rate is calculated by dividing the predicted reward by the predicted total sales.

(S1204) The server device generates a quantitative evaluation value for the calculated predicted payout rate according to the evaluation rule.

(S1205) The server apparatus executes the change of the change condition for the generated quantitative evaluation value according to the change rule, and generates the change change condition.

(S1206) The server device repeatedly calculates the predicted reward and the predicted total sales according to the reward determination rule described later based on the generated change variation condition. Next, the predicted payout rate is repeatedly calculated by dividing the predicted reward by the predicted total sales.

(S1207) The server device displays a change condition and a change change condition, a payout rate corresponding to them, a decision button corresponding to them, and a decision support screen displaying a change condition input field and a recalculation button used for recalculation. Is transmitted to the terminal device and displayed on the display.

(S1208) The user browses the display of the change condition and the change change condition displayed on the terminal device and the corresponding payout rate, and if there is a desired result, if any, recalculate. Make or make a decision.

(S1209) If there is no desired result, input the fluctuation condition related to recalculation in the fluctuation condition input field, add an action with the input device to the recalculation button, send a recalculation instruction, and from (S1203) The calculation operation is repeated for the server device.

(S1210) If there is a predicted payout rate having a desired value, the user applies an action to the determination button related to the desired payout rate by the input device, and transmits a determination command from the terminal device to the server device.

(S1211) The server device records the payout rate, which is a value desired by the user, and the variation condition thereof as determination data in the storage unit and transmits it to the terminal device.

  In this way, the terminal device 1002, 1003, 1004 can use the reward prediction program of the server device 1001 and the like.

(Calculation example)
Hereinafter, a calculation example in which the dividend rate is obtained in accordance with the reward determination rule of the present invention will be shown. This applies to both devices and servers.
(Calculation Example 1)
In accordance with the remuneration decision rule, the fluctuation condition is extracted from the basic form of the remuneration plan, and the dividend rate F is calculated. The fluctuation conditions are determined and calculated as shown in Table 2.
As shown in Table 2, the main parameters, ie, the reward base g (one-side reward base a), the maximum limit M, and the basic reward h are extracted and shown as follows.
g, a: One-side reward base a is 3 (reward base g is 6)
M: Maximum limit M is 2,400,000 yen
h: Basic compensation amount h is 5,000 yen

When the layer depth n = 15 under the conditions of M, h, a (when the number of members is about 32,000), the layer m reaching the maximum limit is calculated using (Equation 4). The minimum value m of k satisfying Equation 4) is
m = 12
It becomes. Then, in the conditions of m and n, the total reward BM of (Case 1) is as follows from (Equation 5):
BM = M * (2 ⁵ -1)
= 2,400,000 * 15 = 36,000,000 (yen)
It becomes. On the other hand, the BP of (Case 2) is from (Equation 6):
BP = h * 2,845
= 5,000 * 2,845 = 14,225,000 (yen)
From (Equation 7), the total compensation BR of (Case 2) is
BR = BP * 2 ⁴
= 14,225,000 * 16 = 227,600,000 (yen)
It becomes. Therefore, from (Equation 8):
Total compensation B = BM + BR
= 36,000,000 + 227,600,000
= 263,600,000 (yen)
It becomes.
On the other hand, the total receipt I of the headquarters is (Formula 9)
Total received amount I = A * (2 ⁿ −1) + b * 2 ⁿ⁻¹
= (2 * A + b) * 2 ^n-1 -A
= (2 * 10,000 + 3,000) * 2 ¹⁴ -10,000
= 376,822,000 (yen)
Therefore, by applying (Equation 10) and calculating the dividend rate F (15),
Dividend rate F (15) = (B / I) * 100 (%)
= (263,600,000 / 376,822,000) * 100 (%)
= 69.95%
It becomes.
Such a dividend can be said to be a firm dividend with reduced risk.

  In general, the payout rate F (n) may not necessarily match the actual payout rate, but this discrepancy is caused by the bias of individual member introducers (probable) Presumed to be caused by bias (multiple). Further analysis of these discrepancies will give the headquarters important business information that can be reflected in how to set up compensation plans. In this way, since the reward plan can be quantitatively evaluated as the dividend rate F (n), it becomes possible to formulate a new reward plan that is even more attractive and less risky.

(Calculation example 2: When a = 2)
Next, a calculation example in the case where the variation condition is changed and the one-side reward radix a is a = 2 is shown.
As shown in Table 3, the main parameters, ie, the reward base g (one-side reward base a), the maximum limit M, and the basic reward h are extracted and shown as follows.
g, a: One side reward base a is 2 (reward base g is 4)
M: Maximum limit M is 2,400,000 yen
h: Basic compensation amount h is 5,000 yen

Except for the case where a = 2 (two left branches and two right branches), it is the same as Calculation Example 1.
Under this condition, when the layer depth n = 15, the following result is obtained by calculating the payout rate by applying (Equation 10). In this case, if calculation is performed in the same manner as in calculation example 1, m = 11.
Dividend rate F (15) = 95.9%
It becomes.
This dividend rate figure is fairly high as a dividend assessment and may be risky for the headquarters. However, if there is a possibility that the dividend rate will be lower (for example, the actual dividend rate can be estimated to be around 50%) due to past trends or actual conditions, there is a risk but attractive appeal with strong appeal. It could be considered as one of the reward plans.

(Calculation Example 3: When a = 4)
Furthermore, an example of calculation when the variation condition is changed and a = 4 is shown. This is a case where a member purchases more products.
As shown in Table 4, the main parameters, ie, the reward base g (one-side reward base a), the maximum limit amount M, and the basic reward amount h are extracted and shown as follows.
g, a: One-side reward base a is 4 (reward base g is 8)
M: Maximum limit M is 2,400,000 yen
h: Basic compensation amount h is 5,000 yen

Except for the case where a = 4 (four left branches and four right branches, respectively), it is the same as calculation examples 1 and 2. In this case, if calculation is performed in the same manner as in calculation examples 1 and 2, m = 12.
Dividend rate F (15) = 47.6%
It becomes.
In this case, the dividend rate is less than 50%, and the risk is small compared with the second embodiment, but it will be evaluated that the compensation plan is not attractive.

  As described above, the payout rate was calculated using (Formula 10) in Calculation Examples 1 to 3. By thus changing the three parameters (variation conditions) h, M, and g (a) that determine the reward plan, it is possible to easily obtain a payout rate that always converges. In addition, since the product unit price A, the admission fee b, and the like can be changed to have different payout rates, a flexible and easy compensation plan can be predicted and evaluated.

Advantages regarding such prediction include the following.
First, by making this prediction before the start of organizational development, it is possible to flexibly determine the compensation plan based on quantitative and reliable indicators, taking into account various circumstances such as economic conditions. .
Secondly, by making this prediction after the start of organizational development, we are not only predicting the dividend rate for the entire controlling company, but also by evaluating the difference between the actual value and the predicted value of each member. It is possible to find an active base organization and to develop an overall organizational development strategy, and to provide guidelines for individual member activities.

In addition to the convergence of the reward prediction calculation in this way, the reward prediction calculation can be performed by manipulating parameters that are easy to appeal to intuition, including product unit price A, admission fee b, etc. The development can be easily considered. In addition, it is easy for the headquarters to consider realistic plans that consider the continuation and stabilization of management and the sedation of overheating of referrals, and is effective in meeting social demands. And versatile reward calculation prediction technology can be provided.

DESCRIPTION OF SYMBOLS 1 Reward prediction calculation apparatus 11 Condition input part 12 Storage part 13 Control part 14 Repeat part 15 Decision support part 16 Output part 600 CPU
601 Memory 602 Display 603 Keyboard 604 Hard disk 605 DVD / CD-ROM drive 606 CD / DVD
1000 Reward prediction calculation system 1001 Reward prediction calculation server (server device)
1002, 1003, 1004 Terminal device 1005 Internet A Product unit price a One-side reward card number B Remuneration amount determined BM Total amount of rewards restricted to the maximum limit (case 1)
BR Total amount of rewards not reaching the maximum limit (case 2)
BP Total amount of remuneration received by one member and all members belonging to the lower ranks from that member b Membership fee F (n) Dividend rate g Compensation base h Basic remuneration amount I Total amount received by supervising company M Maximum amount P (I) Provisional remuneration amount for the member in the i-th layer S (i) Total product purchase number u (i) for the member in the i-th layer Compensation for the member in the i-th layer whose total product purchase amount is S (i) Number of units V Evaluation value

Claims (3)

A remuneration prediction calculation server (1001) for calculating a remuneration amount for a member of a member system introduction sales distribution organization and predicting a dividend rate thereof,
  It is possible to communicate with the terminal devices (1002, 1003, 1004,...) Via the network (1005),
  A condition receiving unit (1101) for receiving a variable condition for calculating the reward from the terminal device;
  Remuneration decision rule for convergence of the dividend rate, evaluation rule for quantitative evaluation of the dividend rate related to the prediction, change rule for changing the variable condition according to the result of the quantitative evaluation, and variable condition for the reception and change A storage unit (1102) for storing
  In accordance with the received fluctuation condition or the change condition related to the change, a prediction evaluation is performed in which the reward amount and the dividend rate are predicted based on the reward determination rule and the evaluation rule, and the dividend rate related to the prediction is quantitatively evaluated. A control unit (1103) to execute;
  A repetitive unit (1104) that changes the fluctuating condition according to the change rule with the fluctuating condition related to the reception as an initial value, and repeatedly executes the prediction evaluation of the control unit;
  A decision support unit (1105) that communicates with the terminal device to execute the display including the dividend rate related to the prediction evaluation and the provision of the input unit for the decision command related to the display;
  An output unit (1106) that, upon receiving the determination command transmitted from the terminal device, transmits the payout rate according to the determination command and the variation condition according to the display as determination data;
With
  The reward determination rule is:
[1] Each member has arrangement information mapped so as to form a hierarchy that is virtually arranged and filled in each node of the binary tree data structure,
[2] Regarding the amount of compensation for each member,
  The total number of products purchased (S (i)), which is the total number of products purchased by members belonging to lower tiers branched from the member based on the arrangement information, a given maximum limit (M), Based on the given reward base (g),
  Calculate the provisional compensation amount (P (i)) that is discretized and evaluated by the compensation base for the total number of items purchased for each tier, and calculate the compensation amount so that the provisional compensation amount does not exceed the maximum limit And
  The provisional compensation amount (P (i)) is
  Regarding the remuneration amount of members belonging to the i-th hierarchy of the binary tree data structure having a hierarchy of depth n,
  The maximum limit (M) is M yen, the basic reward amount (h) is h yen, the reward card number (g) is g, each member purchases one product, and the total product purchase is S (i) And the provisional reward amount is P (i) yen, INT {} is an operator that takes an integer value of the number in {}, and * is an integration calculation symbol,
        P (i) = h * INT {S (i) / g}
Because
  The amount of reward determined is
        M ≦ P (i) and k + i = n + 1
, The case is classified according to the m-th layer counted from the n-th layer that is the minimum value of k,
[Case 1]:
        For members in the hierarchy of i ≦ n + 1−m, the remuneration amount shall be the maximum limit of M yen,
[Case 2]:
        For members in the hierarchy of i> n + 1−m, the reward amount is P (i) = h * INT {(2 ^{(n + 1-i)} -2) Calculate the reward amount as / g} yen,
Reward prediction calculation server (1001). In order to calculate the amount of remuneration for members of the member system introduction sales distribution organization and to predict the dividend rate,
  A condition input means for receiving an input of a variable condition for calculating the remuneration amount;
  A reward determination rule for converging the dividend rate, an evaluation rule for quantitatively evaluating the dividend rate according to the prediction, a change rule for changing the variable condition according to a result of the quantitative evaluation, and the input and change Storage means for storing such variation conditions;
  Based on the variable condition related to the input or change, based on the reward determination rule and the evaluation rule, the reward amount and the dividend rate are predicted, and a prediction evaluation is performed to quantitatively evaluate the dividend rate related to the prediction. Control means;
  Repetitive means for changing the fluctuation condition according to the change rule with the fluctuation condition related to the input as an initial value, and repeatedly executing the prediction evaluation of the control means;
  A decision support means for executing a display including a dividend rate relating to the prediction evaluation, and providing an input means for a decision command relating to the display;
  When the determination command is input, a reward prediction calculation program for causing the payout rate according to the determination command and the variation condition according to the display to function as output means for outputting as determination data,
  The reward determination rule is:
[1] Each member has arrangement information mapped so as to form a hierarchy that is virtually arranged and filled in each node of the binary tree data structure,
[2] Regarding the amount of compensation for each member,
  The total number of products purchased (S (i)), which is the total number of products purchased by members belonging to lower tiers branched from the member based on the arrangement information, a given maximum limit (M), Based on the given reward base (g),
  Calculate the provisional reward amount (P (i)) that is discretized by the reward base and evaluated for the total number of product purchases for each tier,
  Calculate the compensation amount so that the provisional compensation amount does not exceed the maximum limit amount,
  The provisional compensation amount (P (i)) is
  Regarding the remuneration amount of members belonging to the i-th hierarchy of the binary tree data structure having a hierarchy of depth n,
  The maximum limit (M) is M yen, the basic reward amount (h) is h yen, the reward card number (g) is g, each member purchases one product, and the total product purchase is S (i) And the provisional reward amount is P (i) yen, INT {} is an operator that takes an integer value of the number in {}, and * is an integration calculation symbol,
        P (i) = h * INT {S (i) / g}
Because
  The amount of reward determined is
        M ≦ P (i) and k + i = n + 1
, The case is classified according to the m-th layer counted from the n-th layer that is the minimum value of k,
[Case 1]:
        For members in the hierarchy of i ≦ n + 1−m, the remuneration amount shall be the maximum limit of M yen,
[Case 2]:
        For members in the hierarchy of i> n + 1−m, the reward amount is P (i) = h * INT {(2 ^{(n + 1-i)} -2) Calculate the reward amount as / g} yen,
Reward prediction calculation program. In the reward prediction calculation method that calculates the amount of compensation for members of the member system introduction sales distribution organization and predicts the dividend rate,
  A condition input step for inputting a variable condition for calculation of the reward amount;
  A first storage step of storing a reward determination rule for converging the dividend rate, an evaluation rule for quantitatively evaluating the dividend rate related to the prediction, and a change rule for changing the variation condition according to the result of the quantitative evaluation When,
  A second storage step for storing a variation condition related to the input and change;
  Based on the variable condition related to the input or change, based on the reward determination rule and the evaluation rule, the reward amount and the dividend rate are predicted, and a prediction evaluation is performed to quantitatively evaluate the dividend rate related to the prediction. Control steps;
  The variable condition according to the input is changed to an initial value, the variable condition is changed according to the change rule, and the display including the payout rate related to the prediction evaluation is repeatedly executed to repeatedly execute the prediction evaluation of the control step, and the determination instruction related to the display A decision support step for providing the input means;
  When the determination command is input, an output step of outputting the dividend rate according to the determination command and the variation condition according to the display as determination data;
A reward prediction calculation method including
  The reward determination rule is:
[1] Each member has arrangement information mapped so as to form a hierarchy that is virtually arranged and filled in each node of the binary tree data structure,
[2] Regarding the amount of compensation for each member,
  The total number of products purchased (S (i)), which is the total number of products purchased by members belonging to lower tiers branched from the member based on the arrangement information, a given maximum limit (M), Based on the given reward base (g),
  Calculate the provisional compensation amount (P (i)) that is discretized and evaluated by the compensation base for the total number of items purchased for each tier, and calculate the compensation amount so that the provisional compensation amount does not exceed the maximum limit And
  The provisional compensation amount (P (i)) is
  Regarding the remuneration amount of members belonging to the i-th hierarchy of the binary tree data structure having a hierarchy of depth n,
  The maximum limit (M) is M yen, the basic reward amount (h) is h yen, the reward card number (g) is g, each member purchases one product, and the total product purchase is S (i) And the provisional reward amount is P (i) yen, INT {} is an operator that takes an integer value of the number in {}, and * is an integration calculation symbol,
        P (i) = h * INT {S (i) / g}
Because
  The amount of reward determined is
        M ≦ P (i) and k + i = n + 1
, The case is classified according to the m-th layer counted from the n-th layer that is the minimum value of k,
[Case 1]:
        For members in the hierarchy of i ≦ n + 1−m, the remuneration amount shall be the maximum limit of M yen,
[Case 2]:
        For members in the hierarchy of i> n + 1−m, the reward amount is P (i) = h * INT {(2 ^{(n + 1-i)} -2) Calculate the reward amount as / g} yen,
Reward forecast calculation method.

Images