Product sales prediction system, product sales prediction method, and program

Abstract

A product sales prediction system 10 includes: an acquisition unit 111 that acquires sales data of a product in each period from a first period to an n-th period (n is a natural number of 2 or more); a linear calculation unit 112 that, on the basis of the sales data, calculates a straight line indicating the relationship between sales of the product and the number of corresponding periods, by using regression analysis; a tapering unit 113 that deforms the straight line into a curve in which, in and after the n-th period, the amount of change tapers at a constant rate from the amount of change in the straight line; and a predicted sales calculation unit 114 that, on the basis of the curve, calculates predicted sales of the product in a prescribed period in the future.

Description

Product Sales Prediction System, Product Sales Prediction Method, and Program Cross - reference to Related Applications

[0001] This application is based on Japanese Patent Application No. 2025 - 005461 filed on January 15, 2025, the content of which is incorporated herein by reference.

[0002] This disclosure relates to a product sales prediction system, a product sales prediction method, and a program.

[0003] In the development of products, marketing strategies, formulation of production plans, etc., predicting the sales of existing products is very important.

[0004] Therefore, an object of this disclosure is to provide a novel system for predicting the future sales of products.

[0005] A product sales prediction system according to an aspect of this disclosure includes: an acquisition unit that acquires sales data of a product for each period from the first period to the nth period (n is a natural number of 2 or more); a straight - line calculation unit that calculates, by regression analysis, a straight line indicating the relationship between the sales of the product and the number of periods corresponding thereto based on the sales data; a decreasing unit that, after the nth period, transforms the straight line into a curve such that the amount of change decreases at a certain ratio from the amount of change in the straight line; and a predicted sales calculation unit that calculates the predicted sales of the product for a predetermined future period based on the curve.

[0006] A product sales prediction system according to an aspect of this disclosure causes a computer to realize: an acquisition function that acquires sales data of a product for each period from the first period to the nth period (n is a natural number of 2 or more); a straight - line calculation function that calculates, by regression analysis, a straight line indicating the relationship between the sales of the product and the number of periods corresponding thereto based on the sales data; a decreasing function that, after the nth period, transforms the straight line into a curve such that the amount of change decreases at a certain ratio from the amount of change in the straight line; and a predicted sales calculation function that calculates the predicted sales of the product for a predetermined future period based on the curve.

[0007] A product sales forecasting method according to one aspect of this disclosure involves acquiring product sales data for each period from the first period to the nth period (where n is a natural number of 2 or more), calculating a straight line showing the relationship between the sales of the product and the corresponding number of periods by regression analysis based on the sales data, transforming the straight line into a curve such that the amount of change decreases at a constant rate from the amount of change in the straight line after the nth period, and calculating the forecast sales of the product for a predetermined future period based on the curve.

[0008] This disclosure makes it possible to provide a novel system for predicting future sales of a product.

[0009] This document shows the hardware configuration of the product sales forecasting system 10 according to this disclosure. This document shows the functional blocks of the product sales forecasting system 10 according to the first embodiment. This document shows the flowchart of the product sales forecasting system 10 according to the first embodiment. This document shows the functional blocks of the product sales forecasting system 10 according to the second embodiment. This document shows the flowchart of the product sales forecasting system 10 according to the second embodiment. This document shows the flowchart of the product sales forecasting system 10 according to the second embodiment.

[0010] The forms for implementing this disclosure will be described in detail with reference to the drawings.

[0011] 1. Hardware Configuration Diagram 1 is a diagram showing an example of the hardware configuration of the product sales forecasting system 10 according to this disclosure. The product sales forecasting system 10 has a processor 11 such as a CPU (Central Processing Unit) or GPU (Graphics Processing Unit), a storage device 12 such as memory (for example, RAM (Random Access Memory) or ROM (Read Only Memory)), an HDD (Hard Disk Drive) and / or SSD (Solid State Drive), a communication IF (Network Interface) 13 for wired or wireless communication, an input device 14 for receiving input operations, and an output device 15 for outputting information. The input device 14 is, for example, a keyboard, a touch panel, a mouse and / or a microphone. The output device 15 is, for example, a display, a touch panel, and / or a speaker.

[0012] 2. First Embodiment A first embodiment of the present disclosure will be described. Figure 2 is a diagram showing an example of the functional block configuration of a product sales forecasting system 10 according to the first embodiment of the present disclosure. The product sales forecasting system 10 includes an acquisition unit 111, a linear calculation unit 112, a reduction unit 113, a forecast sales calculation unit 114, a product sales data storage unit 121, and a calculation formula storage unit 122. The product sales data storage unit 121 and the calculation formula storage unit 122 can be realized using a storage device 12 provided by the product sales forecasting system 10. The acquisition unit 111, the linear calculation unit 112, the reduction unit 113, and the forecast sales calculation unit 114 can be realized by the processor 11 of the product sales forecasting system 10 executing a program stored in the storage device 12. The program can be stored in a storage medium. The storage medium storing the program may be a computer-readable non-temporary storage medium. The non-temporary storage medium is not particularly limited, but may be a USB (Universal Serial Bus) memory or a CD-ROM (Compact Disc Read-Only Memory), for example. The product sales forecasting system 10 does not need to consist of a single computer, but may consist of multiple computers and storage devices distributed over a communication network. For example, it may be implemented by a cloud service such as SaaS (Software as a Service), in which case the product sales data storage unit 121 is managed on a cloud server, and each functional module (acquisition unit 111, linear calculation unit 112, reduction unit 113, and forecast sales calculation unit 114) is executed on the cloud server.

[0013] The product sales forecasting system 10 according to the first embodiment calculates projected sales for a predetermined period in the future. The product can be any product, but products with low price elasticity of demand, products with controlled prices, products with small price differences depending on the sales location are preferred. Examples of such products include pharmaceuticals, luxury goods such as tobacco, infrastructure services such as electricity and gas, and services of public transportation and national / public schools. The period can be of any length; for example, it may be from April to March of the following year, or it may be a year from January to December, or it may be a quarter divided into four periods. If the year from April to March of the following year is considered as one period, the predetermined period refers to April to March of the following year in a predetermined fiscal year, and the projected sales for the predetermined period refers to the total projected sales from April to March of the following year in a predetermined fiscal year. Furthermore, sales may be sales amount or sales quantity, but sales quantity is preferred.

[0014] The product sales data storage unit 121 stores product sales data for each period from the first period to the nth period (where n is a natural number of 2 or more), which has been acquired by the acquisition unit 111. The first period to the nth period are consecutive periods; for example, if the first period refers to January to December of a given fiscal year, the second period refers to January to December of the following fiscal year.

[0015] The calculation formula storage unit 122 stores formulas representing straight lines calculated by the straight line calculation unit 112, and formulas representing straight lines (curves) transformed by the reduction unit 113, etc.

[0016] 2-1. Calculation of Predicted Product Sales The calculation of predicted product sales using the product sales forecasting system 10 will be explained using the flowchart in Figure 3.

[0017] First, the acquisition unit 111 acquires product sales data for each period from the first period to the nth period (step S101). The acquisition unit 111 may acquire product sales data from external devices such as business equipment used by businesses that sell products, devices that aggregate sales data for each product, cloud servers on a network, websites, or the company's own database, or it may acquire the data by receiving product sales data from users using the product sales forecasting system 10. When the acquisition unit acquires product sales data from external devices, cloud servers on a network, websites, or the company's own database, these may be connected to the product sales forecasting system 10 via a communication network. The product sales data acquired by the acquisition unit 111 is stored in the product sales data storage unit 121 (step S102).

[0018] The line calculation unit 112 calculates a line showing the relationship between product sales and the corresponding number of periods by regression analysis based on the product sales data for each period stored in the product sales data storage unit 121 (step S103). The line showing the relationship between product sales and the corresponding number of periods may be, for example, a line with the number of periods on the x-axis and product sales on the y-axis. In that case, the line calculated by regression analysis is a line calculated by regression analysis based on each plotted point (x, y) where the number of periods x and the corresponding sales y are plotted for each period from the first period to the nth period. The regression analysis is preferably a linear regression analysis. The linear regression analysis may be, for example, a simple linear regression analysis. The equation showing the line calculated by the line calculation unit 112 is stored in the calculation formula storage unit 122 (step S104).

[0019] The reduction unit 113 transforms the straight line into a curve such that, from the nth period onward, the amount of change decreases at a constant rate from the amount of change in the straight line, based on the straight line equation stored in the calculation formula storage unit 122 (step S105). The amount of change in the straight line is the absolute value of the slope of the straight line, and the reduction unit 113 transforms the straight line so that the absolute value of the slope decreases at a constant rate. The rate of reduction can be appropriately determined between 0% and 100% depending on the characteristics of the product and the sales environment, but for example, the amount of change may continue to decrease at a rate of 10%, 20%, 30%, 40%, or 50%. The interval at which the amount of change decreases can be any length and can be appropriately determined depending on the characteristics of the product and the sales environment, but for example, the amount of change may decrease every period. In that case, for example, if the amount of change in the nth period is 10 and the amount of change is to decrease at a rate of 10%, the decreasing unit 113 deforms the line so that the amount of change continues to decrease, with the amount of change in the (n+1)th period being 9, the amount of change in the (n+2)th period being 8, the amount of change in the (n+3)th period being 7, and so on. Specifically, when the number of periods is x, the predicted sales of the product is y, the sales of the product in the nth period is I, the slope of the line is S, and the decreasing coefficient (0 ≤ t ≤ 1) indicating the rate of decrease is t, the equation showing the curve obtained after deformation by the decreasing unit 113 from the nth period onward may be shown by the following equation (Equation 1).

[0020]

[0021] The equation representing the curve obtained by the reduction unit 113 is stored in the calculation formula storage unit 122 (step S106).

[0022] The forecast sales calculation unit 114 calculates the forecast sales of a product for a predetermined future period based on the curve equation stored in the calculation formula storage unit 122 (step S107). The predetermined period is any period from the (n+1)th period onward. When calculating forecast sales using the above formula 1, x is substituted with the number of periods that is the difference between the predetermined period and the nth period. That is, if you want to calculate the forecast sales for the (n+5)th period, you can calculate the forecast sales y by substituting (n+5)-n=5 for x. Alternatively, the forecast sales y can also be calculated based on a curve equation that has been modified as appropriate by the provider or user of the product sales forecasting system 10. For example, if it is known in advance that the increase in sales will decrease significantly in a certain period in the future, the curve equation may be modified as appropriate. The calculated forecast sales are provided to the user from the product sales forecasting system 10 (step S108). They may also be provided by outputting the forecast sales to the user terminal.

[0023] 3. Second Embodiment A second embodiment of the present disclosure will now be described. Figure 4 is a diagram showing an example of the functional block configuration of the product sales forecasting system 10 according to the second embodiment of the present disclosure. The product sales forecasting system 10 includes an acquisition unit 111, a population adjustment unit 1113, a linear calculation unit 112, a reduction unit 113, a forecast sales calculation unit 114, a forecast sales amount calculation unit 115, a product sales data storage unit 121, and a calculation formula storage unit 122. The product sales data storage unit 121 and the calculation formula storage unit 122 can be realized using a storage device 12 provided by the product sales forecasting system 10. Furthermore, the acquisition unit 111, the population adjustment unit 1113, the linear calculation unit 112, the reduction unit 113, the forecast sales calculation unit 114, and the forecast sales amount calculation unit 115 can be realized by the processor 11 of the product sales forecasting system 10 executing a program stored in the storage device 12.

[0024] The product sales forecasting system 10 according to the second embodiment calculates the projected sales quantity of a pharmaceutical product over a predetermined period in the future. The pharmaceutical product may be any pharmaceutical product, and may be an oral drug or a topical drug, or an over-the-counter drug or a prescription drug. Furthermore, the pharmaceutical product in this embodiment may be a drug whose first nine digits of the drug price standard listed drug code are common.

[0025] If a drug formulation consists of two or more Stock Keeping Units (SKUs), for example, if it consists of a brand-name drug and a generic drug, the acquisition unit 111 further comprises a brand-name drug data acquisition unit 1111 and a generic drug data acquisition unit 1112, and the product sales forecasting system 10 calculates the total predicted sales quantity of the brand-name drug and the generic drug. Here, if a drug formulation consists of a brand-name drug and a generic drug, it means that the generic drug was launched before the nth period in which the sales quantity data is acquired. If the drug formulation consists of only one SKU, the acquisition unit 111 functions, and if the drug formulation consists of two or more SKUs, the brand-name drug data acquisition unit 1111 and the generic drug data acquisition unit 1112 function. In addition to the total predicted sales quantity of the brand-name drug and the generic drug, the system may also calculate the individual predicted sales quantities of the brand-name drug and the generic drug. When further calculating the projected sales quantities for both original drugs and generic drugs, the projected sales calculation unit 114 further comprises a convergence share setting unit 1141, an original drug sales share curve calculation unit 1142, an original drug projected sales share calculation unit 1143, a generic drug projected sales share calculation unit 1144, an original drug projected sales quantity calculation unit 1145, and a generic drug projected sales quantity calculation unit 1146.

[0026] The product sales data storage unit 121 stores the following: sales quantity data of pharmaceutical products for each period from the first period to the nth period, acquired by the acquisition unit 111; sales quantity data of original drugs for each period from the first period to the nth period, acquired by the original drug data acquisition unit 1111; sales quantity data of generic drugs for each period from the first period to the nth period, acquired by the generic drug data acquisition unit 1112; predicted sales share of original drugs calculated by the original drug predicted sales share calculation unit 1143; predicted sales share of generic drugs calculated by the generic drug predicted sales share calculation unit 1144; predicted sales quantity of original drugs calculated by the original drug predicted sales quantity calculation unit 1145; and predicted sales quantity of generic drugs calculated by the generic drug predicted sales quantity calculation unit 1146. Here, the sales quantity may also be the quantity used, in which case the product sales forecasting system 10 calculates the predicted quantity used.

[0027] The calculation formula storage unit 122 stores the convergence share set in the convergence share setting unit 1141, the formula showing the curve calculated by the original drug sales share curve calculation unit 1142, and so on.

[0028] 3-1. Calculation of total projected sales volume of original and generic drugs The calculation of the total projected sales volume of original and generic drugs, when a drug formulation consists of original and generic drugs, using the product sales forecasting system 10, will be explained using the flowchart in Figure 5.

[0029] First, the original drug data acquisition unit 1111 and the generic drug data acquisition unit 1112 each acquire sales quantity data for original drugs and generic drugs for each period from the first period to the nth period (step S201). The original drug data acquisition unit 1111 and the generic drug data acquisition unit 1112 may acquire sales quantity data for pharmaceuticals from, for example, an external device that aggregates sales data for pharmaceuticals, a cloud server on a network, a site, or an in-house database, or they may acquire the data by receiving sales data for pharmaceuticals from a user using the product sales forecasting system 10. The acquired sales quantity data for original drugs and generic drugs may be, for example, data disclosed by the National Database (NDB) of the Ministry of Health, Labour and Welfare, or data extracted from the pharmaceutical master, which is the basic master of the claims processing system of the Social Insurance Medical Fee Payment Fund. When the original drug data acquisition unit 1111 and the generic drug data acquisition unit 1112 acquire sales data for pharmaceutical products from external devices, cloud servers on a network, websites, company databases, etc., these may be connected to the product sales forecasting system 10 via a communication network. The sales quantity data for original drugs and generic drugs acquired by the original drug data acquisition unit 1111 and the generic drug data acquisition unit 1112 are stored in the product sales data storage unit 121 (step S202). The data stored in step S202 is the total sales quantity data for original drugs and generic drugs, as well as the sales quantity data for original drugs and generic drugs individually. In addition, the original drug data acquisition unit 1111 and the generic drug data acquisition unit 1112 may acquire sales quantity data for original drugs and generic drugs by genre, for example, by gender, age, and / or age group. In that case, the product sales forecasting system 10 calculates the predicted sales quantity of pharmaceutical products by gender, age, and / or age group. Furthermore, the sales quantity data of formulations by gender, age, and / or age group, obtained by the original drug data acquisition unit 1111 and the generic drug data acquisition unit 1112, may be adjusted by dividing them by the respective populations for each gender, age, and / or age group before proceeding to step S202 and beyond.When adjusting by dividing by each population, the acquisition unit 111 may further include a population adjustment unit 1113.

[0030] The linear calculation unit 112 calculates a linear line showing the relationship between the sales quantity of a drug and the corresponding number of periods by regression analysis, based on the total sales quantity data of original and generic drugs for each period stored in the product sales data storage unit 121 (step S203). The formula showing the calculated linear line is stored in the calculation formula storage unit 122 (step S204). Steps S203 and S204 are the same as steps S103 and S104, respectively.

[0031] The reduction unit 113, based on the linear equation stored in the calculation formula storage unit 122, transforms the linear equation into a curve such that the amount of change decreases at a constant rate from the amount of change in the linear equation from the nth period onward (step S205). The equation representing the curve obtained by the transformation by the reduction unit 113 is stored in the calculation formula storage unit 122 (step S206). Steps S205 and S206 are the same as steps S105 and S106, respectively.

[0032] The forecast sales calculation unit 114 calculates the total forecast sales quantity of original and generic drugs for a predetermined future period of the product based on the curve formula stored in the calculation formula storage unit 122 (step S207). The calculated forecast sales quantity is provided to the user from the product sales forecasting system 10 (step S208).

[0033] 3-2. Calculation of Predicted Sales Volumes for Original and Generic Drugs The calculation of the predicted sales volumes for original and generic drugs, respectively, using the product sales forecasting system 10, when the formulation consists of original and generic drugs (when the generic drug was launched before the nth period in which sales volume data is acquired), will be explained using the flowchart in Figure 6.

[0034] The convergence share setting unit 1141 sets the convergence share for when the sales share of the original drug converges to a certain value (step S211). The convergence share may be set or changed as appropriate by the provider or user of the product sales forecasting system 10. The convergence share is 0 or more and less than 100 when the total share is expressed as a percentage, and 0 or more and less than 1 when the total share is set to 1. The value of the convergence share set by the convergence share setting unit 1141 is stored in the calculation formula storage unit 122 (step S212).

[0035] In steps S201 and S202, the original drug sales share curve calculation unit 1142 calculates an original drug sales share curve (step S213) that shows the relationship between the original drug sales share, which includes information on the market share of original drugs, and the number of years since the launch of generic drugs, based on the sales quantity data of generic drugs stored in the product sales data storage unit 121 and the convergence share stored in the calculation formula storage unit 122.

[0036] For example, if a generic drug is launched between the first period for which sales volume data was obtained and the nth period (i.e., the generic drug is launched in the mth period (where m is a natural number less than or equal to n)), the sales share curve of the original drug can be calculated by performing steps 1) to 5) below. 1) Create rectangles for periods m through n, where the length of the shorter side is the sales share (≤1) for the p-th period (where p is a natural number between m and n, inclusive), and the length of the longer side is 1. 2) When the generic drug is launched, divide the m-th period into four parts, and b=1 for the first quarter, b=2 for the second quarter, b=3 for the third quarter, and b=4 for the fourth quarter. Arrange each rectangle in the region x ≤ 1 / 8 + 1 / 4*(4-b) + (p-m) and y ≤ 1, such that the longer side touches y=1 and the shorter side touches x = 1 / 8 + 1 / 4*(4-b) + (p-m). 3) Transform the rectangle for the m-th period into an area-equal triangle with a base of 1 / 8 + 1 / 4*(4-b) such that its sides touch y=1 and the rectangle for the (m+1)th period of the generic drug. 4) Transform the rectangles into trapezoids with two sides perpendicular to y=1 as the upper and lower bases, so that the curves connecting the vertices of the triangles and rectangles that do not touch y=1 converge to the line y=converging share. 5) Calculate the equations of the curves connecting the vertices of the triangles and trapezoids that do not touch y=1.

[0037] Furthermore, for example, if a generic drug was launched before the first period for which sales volume data was obtained, the sales share curve of the original drug may be calculated by performing the following steps 1) to 4). 1) Create rectangles for each of the first to n periods, where the length of the shorter side is the sales share (≤1) of the q-th period (where q is a natural number less than or equal to n) and the length of the longer side is 1. 2) Arrange each rectangle in the region x ≤ q and y ≤ 1 such that the longer side touches y = 1 and the shorter side touches x = q. 3) Transform each rectangle into an area-equal trapezoid with two sides perpendicular to y = 1 as the upper and lower bases, such that the curves connecting the vertices of each rectangle that do not touch y = 1 converge to the straight line of y = convergent share. 4) Calculate the equation of the curves connecting the vertices of each trapezoid that do not touch y = 1. If there is little increase or decrease in the sales volume of generic drugs between the first and second periods, the original drug sales share curve approximates the straight line of y = convergent share.

[0038] Specifically, if x is the number of years a generic drug has been around, y is the sales share of the original drug, C is the convergence share, and a is an arbitrary coefficient such that 0 < a < 1, then the sales share curve of the original drug is y = (1 - C) * a ｘ It may also be expressed as +C (Equation 2).

[0039] The formula for the original drug sales share curve calculated by the original drug sales share curve calculation unit 1142 is stored in the calculation formula storage unit 122 (step S214).

[0040] The original drug sales share calculation unit 1143 calculates the predicted sales share of the original drug for a predetermined future period based on the original drug sales share curve formula stored in the calculation formula storage unit 122 (step S215). When calculating the predicted sales share of the original drug using the above formula 2, x is substituted with the number of periods that are the difference between the predetermined future period and the period in which the generic drug was launched if the generic drug was launched between the first period and the nth period, and with the number of periods that are the difference between the predetermined future period and the first period if the generic drug was launched before the first period. In addition, the predicted sales y can also be calculated based on the original drug sales share curve formula that has been modified as appropriate by the provider or user of the product sales forecasting system 10. For example, the original drug sales share curve formula may be modified as appropriate if the sales share of the generic drug for a certain period in the future is known in advance. The calculated predicted sales share of the original drug is stored in the product sales data storage unit 121 (step S216).

[0041] The generic drug forecast sales share calculation unit 1144 calculates the forecast sales share of generic drugs by subtracting the forecast sales share of the original drug stored in the product sales data storage unit 121 from the total share (step S217). If the total share or the forecast sales share of the original drug is expressed as a percentage, the forecast sales share is subtracted from 100. If the total share or the forecast sales share of the original drug is expressed as a number with the total share set to 1, the forecast sales share is subtracted from 1. The calculated forecast sales share of generic drugs is stored in the product sales data storage unit 121 (step S218).

[0042] The pioneer drug predicted sales quantity calculation unit 1145 and the follow-up drug predicted sales quantity calculation unit 1146 each multiply the total predicted sales quantity of the pioneer drug and the follow-up drug by the predicted sales share of the pioneer drug and the follow-up drug stored in the product sales data storage unit 121 to calculate the predicted sales quantity of the pioneer drug and the follow-up drug (step S219). Here, the total predicted sales quantity of the pioneer drug and the follow-up drug is obtained by executing steps S201 to S207. The calculated predicted sales quantity of each of the pioneer drug and the follow-up drug is stored in the product sales data storage unit 121 (step S220).

[0043] 3-3. Calculation of the predicted sales amount of the preparation The calculation of the predicted sales amount of the preparation by the product sales prediction system 10 will be described using the flowchart of FIG. 7.

[0044] The predicted sales amount calculation unit 115 multiplies the predicted sales quantity of the preparation stored in the product sales data storage unit 121 by the predicted drug price of the corresponding preparation to calculate the predicted sales amount (step S231). When the preparation consists of a pioneer drug and a follow-up drug, the predicted sales amount of each of the pioneer drug and the follow-up drug may be further calculated by multiplying the predicted sales quantity of each of the pioneer drug and the follow-up drug stored in the product sales data storage unit 121 by the corresponding predicted drug price.

[0045] Here, the predicted drug price of the preparation (including the pioneer drug or the follow-up drug) may be appropriately set and changed by the provider or user of the product sales prediction system 10, etc., but it is preferably the drug price predicted based on the drug price revision data for the nth period and the (n-1)th period. For example, assuming that the predicted drug price for the year after the nth period is the drug price for the (n-1)th period, and the predicted drug price for the year after the year after the nth period is the drug price for the nth period, the predicted drug price for the future may be determined so that the drug price for the nth period and the drug price for the (n-1)th period come alternately.

[0046] The embodiments described above are examples for illustrating the present disclosure and are not intended to limit the disclosure to those embodiments. Furthermore, the present disclosure can be modified in various ways without departing from its essence. For example, a person skilled in the art could replace the resources (hardware resources or software resources) described in the embodiments with their equivalents, and such replacements are also within the scope of the present disclosure.

Claims

1. A product sales forecasting system comprising: an acquisition unit that acquires sales data for a product for each period from the first period to the nth period (where n is a natural number of 2 or more); a line calculation unit that calculates a line showing the relationship between the sales of the product and the corresponding number of periods by regression analysis based on the sales data; a reduction unit that transforms the line into a curve such that the amount of change decreases at a constant rate from the amount of change in the line after the nth period; and a forecast sales calculation unit that calculates the forecast sales of the product for a predetermined future period based on the curve.

2. The product sales forecasting system according to claim 1, wherein when x is the number of periods, y is the predicted sales of the product, I is the sales of the product for the nth period, S is the slope of the line, and t is an arbitrary decreasing coefficient such that 0 ≤ t ≤ 1, the portion of the curve after the nth period is given by the following equation.

3. The product sales forecasting system according to claim 1 or 2, wherein the product is a pharmaceutical product, and the forecast sales calculation unit calculates the forecast sales quantity of the pharmaceutical product.

4. The product sales forecasting system according to claim 3, wherein the formulation consists of an original drug and a generic drug, and the forecast sales calculation unit calculates the total forecast sales quantity of the original drug and the generic drug.

5. The product sales forecasting system according to claim 4, wherein the forecast sales calculation unit further calculates the forecast sales quantities of the original drug and the generic drug, respectively.

6. The acquisition unit further comprises: an original drug data acquisition unit that acquires sales data of original drugs from the first period to the nth period; a generic drug data acquisition unit that acquires sales data of generic drugs from the first period to the nth period; the forecast sales calculation unit further comprises: a convergence share setting unit that sets a convergence share when the sales share of the original drug converges to a certain value if the generic drug was launched before the nth period; an original drug sales share curve calculation unit that calculates an original drug sales share curve showing the relationship between the original drug sales share, which includes information on the market share of the original drug's sales, and the corresponding number of periods, based on the sales data of the generic drug and the convergence data, if the generic drug was launched before the nth period; and an original drug forecast sales share calculation unit that calculates a forecast sales share of the original drug in a predetermined future period based on the original drug sales share curve, if the generic drug was launched before the nth period. A product sales forecasting system according to claim 5, further comprising: a generic drug forecast sales share calculation unit that calculates the forecast sales share of the generic drug by subtracting the forecast sales share of the original drug from the overall share; an original drug forecast sales quantity calculation unit that calculates the forecast sales quantity of the original drug by multiplying the total forecast sales quantity of the original drug and the generic drug calculated by the forecast sales calculation unit by the forecast sales share of the original drug; and a generic drug forecast sales quantity calculation unit that calculates the forecast sales quantity of the generic drug by multiplying the total forecast sales quantity of the original drug and the generic drug calculated by the forecast sales calculation unit by the forecast sales share of the generic drug.

7. If the generic drug is launched during the m-th period (where m is a natural number less than or equal to n), the original drug sales share curve calculation unit shall: 1) create rectangles for each period from the m-th to the n-th, where the length of the short side is the sales share (≤1) of the p-th period (where p is a natural number between m and n), and the length of the long side is 1; 2) when the generic drug is launched, the m-th period is divided into four quarters, and b=1 for the first quarter, b=2 for the second quarter, b=3 for the third quarter, and b=4 for the fourth quarter, and each of the rectangles shall be positioned in the region x ≤ 1 / 8 + 1 / 4 * (4 - b) + (p - m) and y ≤ 1, such that the long side touches y = 1 and the short side touches x = 1 / 8 + 1 / 4 * (4 - b) + (p - m); A product sales forecasting system according to claim 6, which calculates the original drug sales share curve by performing the following: 3) transforming the rectangle for the period m into an area-equal triangle with a base of 1 / 8 + 1 / 4 * (4 - b) such that its sides are tangent to y = 1 and the rectangle for the period (m + 1) of the generic drug; 4) transforming the rectangle into an area-equal trapezoid with two sides perpendicular to y = 1 as the upper and lower bases such that the curves connecting the vertices of the triangle and each of the rectangles that are not tangent to y = 1 converge to the line of y = the convergent share; and 5) calculating the equation of the curves connecting the vertices of the triangle and the trapezoid that are not tangent to y = 1.

8. If the generic drug was launched before the first period, the product sales forecasting system according to claim 6, wherein the original drug sales share curve calculation unit calculates the original drug sales share curve by performing: 1) creating rectangles for each of the first to n periods, where the length of the short side is the sales share (≤1) of the q period (where q is a natural number less than or equal to n) and the length of the long side is 1; 2) arranging each of the rectangles in the region x ≤ q and y ≤ 1 such that the long side touches y = 1 and the short side touches x = q; 3) transforming each of the rectangles corresponding to the first to n periods into trapezoids with two sides perpendicular to y = 1 as the upper and lower bases, such that the curves connecting the vertices of each rectangle that do not touch y = 1 converge to the straight line y = the convergent share; and 4) calculating the equation of the curve connecting the vertices of each trapezoid that do not touch y = 1.

9. When x is the number of years the generic drug has been in use, y is the sales share of the original drug, C is the convergence share, and a is an arbitrary coefficient such that 0 < a < 1, the sales share curve of the original drug is y = (1 - C) * a ｘ A product sales forecasting system according to claim 6, wherein +C.

10. The product sales forecasting system according to claim 3, further comprising a predicted sales amount calculation unit that calculates a predicted sales amount by multiplying the predicted sales quantity of the formulation by the predicted drug price.

11. The product sales forecasting system according to claim 10, wherein the predicted drug price is a drug price predicted based on drug price revision data for the nth period and the (n-1)th period.

12. The product sales forecasting system according to claim 11, wherein the predicted drug price for the year following the n-th period is the drug price for the (n-1)-th period, and the predicted drug price for the year after the nth period is the drug price for the n-th period.

13. A program that enables a computer to implement: an acquisition function for acquiring sales data for a product for each period from the first period to the nth period (where n is a natural number greater than or equal to 2); a line calculation function for calculating a line showing the relationship between the sales of the product and the corresponding number of periods using regression analysis based on the sales data; a diminishing function for transforming the line into a curve such that the amount of change decreases at a constant rate from the amount of change in the line after the nth period; and a forecast sales calculation function for calculating forecast sales of the product for a predetermined future period based on the curve.

14. A method for predicting product sales, comprising: obtaining sales data for each period from the first period to the nth period (where n is a natural number of 2 or more); calculating a straight line showing the relationship between the sales of the product and the corresponding number of periods using regression analysis based on the sales data; transforming the straight line into a curve such that the amount of change decreases at a constant rate from the amount of change in the straight line after the nth period; and calculating the predicted sales of the product for a predetermined future period based on the curve.

Images