JP2017027632A - Demand prediction device, demand prediction system, control method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a demand prediction device, a demand prediction system, a control method, and a program capable of efficiently performing demand prediction of products which have a poor demand past record required for predicting demand of the products.SOLUTION: A demand past record related to a similar product of a prediction object product (new product) is collected, and demand prediction can be performed according to an each time period of sale, in a manner that the prediction object product is before sale, is in an initial state of sale and is in a fluctuation period, or has already been sold and a predetermined period has passed since start of sale.SELECTED DRAWING: Figure 14

Description

  The present invention is an invention of an information processing device that performs demand prediction of products, and in particular, an information processing device that can perform demand prediction even if there is no information about demand results before release, such as a new product, The present invention relates to a prediction method and a program.

  In the case of renewal of existing products, a certain degree of prediction is possible by extracting trends and cycles from demand data of products before renewal and applying them to the future.

  However, in the case of a completely new product, since there is no or little past demand data, it can be said that the demand prediction is very difficult. In particular, there are two points in forecasting the demand for new products: estimating the initial sales scale and capturing subsequent demand movements.

Therefore, when forecasting demand for new products, we introduce the concept of product groups that are expected to show similar order fluctuations, and use demand fluctuations of the product groups to which they belong even for products for which past orders have been insufficient. Thus, an invention for predicting the demand for the product is disclosed (for example, see Patent Document 1).
JP 2001-243401 A

  However, the invention described in Patent Document 1 is shown for products for which sales have started, but for which past demand results are insufficient, and products for which there is no actual demand, that is, new products to be released in the future. There is no description or suggestion about forecasting demand for products (for example, paragraph numbers “0037” to “0044”, etc.), so it is necessary to predict demand after launching such new products. There may be a problem such as delay in arrangement of inventory.

  Estimating the initial sales scale of a new product is a very difficult problem, and the most common practice is that the person in charge estimates it based on past experience and various factors. It is necessary to make a comprehensive judgment based on the results of the products, superiority or inferiority of the products of competitors, and sales promotion.

  However, if the method based on the experience of the person in charge is used, when estimating the demand amount, the work load on the person in charge for analysis of various factors, etc. is high, and it may be attributed to the experience. It is not certain if it will be required.

  On the other hand, grasping demand trends after the launch of new products is also a very difficult problem.

  Even if the initial sales scale after the release is determined, it is necessary to grasp the demand trend such as whether the sales volume will increase, become constant, decrease, or how fast it will decrease.

  In any case, there is no doubt that it is difficult to predict new products.

  The present invention has been made in view of such a problem, and is a demand prediction device capable of efficiently performing demand prediction of a product with a poor demand record required for performing demand prediction of the product, demand An object is to provide a prediction system, a control method, and a program.

  A first invention for achieving the above object is a demand prediction device for predicting a demand amount of a new product, and a demand result storing means for storing a demand result including a demand amount of the product and a date when the demand is present; A similar product information setting unit that selects a product similar to the new product, a calculation unit that calculates a base demand amount of the new product from a demand amount of the similar product selected by the similar product information setting unit, and the calculation unit And a prediction execution means for obtaining a demand amount after the sales start date of the new product using the base demand amount of the new product calculated by the above.

  A second invention for achieving the above object is a demand prediction system in which a client terminal and an information processing apparatus are connected via a network, and the client terminal selects a product similar to a new product. A similar product information setting unit, wherein the information processing apparatus stores a demand result storage unit that stores a demand record including a demand amount of a product and a date when the demand is present, and a similar product selected by the similar product information setting unit. A calculation means for calculating a base demand amount of the new product from a demand amount, and a prediction execution means for obtaining a demand amount after the start date of the new product using the base demand amount of the new product calculated by the calculation means; A demand forecasting system characterized by comprising:

  A third invention for achieving the above object is a control method for a demand prediction apparatus for predicting a demand amount of a new product, wherein the demand prediction apparatus selects similar product information for selecting a product similar to the new product. The base demand of the new product from the demand amount stored in the demand result storage means for storing the demand result including the demand amount of the product and the demand date of the similar product selected by the setting means and the similar product information setting means A calculation unit that calculates an amount; and a prediction execution unit that calculates a demand amount after the start date of release of the new product using the base demand amount of the new product calculated by the calculation unit. It is a control method of a demand prediction device.

  A fourth invention for achieving the above object is a program that can be read and executed by a demand prediction device that predicts a demand amount of a new product, and selects a product similar to the new product for the demand prediction device. From the demand amount stored in the demand result storage means for storing the demand result of the similar product selected by the similar product information setting means and the similar product selected by the similar product information setting means, including the demand amount of the product and the date when there was a demand. A calculation unit that calculates a base demand amount of the product, and a prediction execution unit that calculates a demand amount after the start date of release of the new product by using the base demand amount of the new product calculated by the calculation unit. It is a program characterized by this.

  A fifth invention for achieving the above object is a method for controlling a demand prediction system in which a client terminal and an information processing apparatus are connected via a network, wherein the client terminal is a product similar to a new product. A similar product information setting unit that selects the demand, and the information processing apparatus stores a demand record including a demand amount of the product and a date when there was a demand for the similar product selected by the similar product information setting unit. Demand after the sales start date of the new product using the calculation means for calculating the base demand amount of the new product from the demand amount stored in the result storage means, and the base demand amount of the new product calculated by the calculation means A control method for a demand prediction system, characterized by executing a prediction execution means for obtaining a quantity.

  A sixth invention for achieving the above object is a program that can be read and executed by a demand prediction system in which a client terminal and an information processing apparatus are connected via a network, and the client terminal is similar to a new product. Functioning as similar product information setting means for selecting a product to be stored, and storing information on the demand information including the demand amount of the product and the date of demand for the similar product selected by the similar product information setting means. A calculation means for calculating a base demand amount of the new product from a demand amount stored in the demand record storage means, and a base demand amount of the new product calculated by the calculation means, after the start date of the sale of the new product It is a program characterized by functioning as a prediction execution means for obtaining the demand amount.

  Advantageous Effects of Invention According to the present invention, it is possible to efficiently perform demand prediction from the start to the end of a product to be predicted by using a demand record of a product similar to a product for which demand prediction is performed. .

It is a block diagram which shows the function structure of the demand prediction apparatus which concerns on embodiment of this invention. It is a block diagram which shows the structure of the hardware of the demand prediction apparatus which concerns on embodiment of this invention. It is a flowchart which shows the process which performs the demand prediction of the new goods in embodiment of this invention. It is a block diagram which shows the structure of the merchandise information database in embodiment of this invention. It is a block diagram which shows the structure of the similar goods information database in embodiment of this invention. It is a block diagram which shows the structure of the similar goods seasonal index information database in embodiment of this invention. It is a block diagram which shows the structure of the demand performance database in embodiment of this invention. It is a block diagram which shows the structure of the prediction model information database in embodiment of this invention. It is a block diagram which shows the structure of the prediction result database in embodiment of this invention. It is a block diagram which shows the structure of the merchandise information setting screen in embodiment of this invention. It is a block diagram which shows the structure of the similar goods information setting screen in embodiment of this invention. It is a flowchart which shows the similar goods seasonal index calculation execution process in embodiment of this invention. In embodiment of this invention, it is explanatory drawing for demonstrating the example of the demand amount of similar goods. It is a flowchart which shows the prediction execution process in embodiment of this invention. It is a flowchart which shows the detail of the prediction execution process in the time before sale in embodiment of this invention. In embodiment of this invention, it is explanatory drawing for demonstrating the example which calculates | requires the base demand amount of a new product from the demand amount of the similar goods within a fluctuation period. In embodiment of this invention, it is explanatory drawing for demonstrating the example which shows matching a base demand amount at the time of a sales start time with an initial predicted quantity. In embodiment of this invention, it is explanatory drawing for demonstrating the example which calculates | requires the predicted quantity of the demand amount of the new goods after the fluctuation period. It is a flowchart which shows the detail of the prediction execution process after the present time after sale in embodiment of this invention. It is a flowchart which shows the detail of the prediction execution process after the fluctuation period in embodiment of this invention. In embodiment of this invention, it is explanatory drawing for demonstrating the example which calculates | requires the predicted quantity of the demand amount of the new goods after the present time in the state by which the new goods were already marketed. It is a block diagram which shows the structure of the new goods demand output screen in embodiment of this invention.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a diagram illustrating an example of a functional configuration of a demand prediction apparatus that is an example of an information processing apparatus (computer). As illustrated in FIG. 1, the demand prediction apparatus 100 according to the present embodiment includes a database, a product information setting unit 110, a similar product information setting unit 120, a similar product seasonal index calculation execution unit 130, and a prediction execution unit 140. The prediction result confirmation unit 150 is provided.

  The database (DB) includes a product information DB 101, a similar product information DB 102, a similar product seasonal index information DB 103, a demand performance DB 104, a prediction model information DB 105, and a prediction result DB 106.

  As shown in FIG. 4, the product information DB 101 includes a product code that uniquely indicates a product, a product name that indicates the name of the product corresponding to the product code, a release date that indicates the release start date of the product, a variation period (described later in detail), This is a database in which information serving as a base demand calculation source such as a variable period seasonality presence / absence flag (details will be described later), initial sales quantity (details will be described later), and the like is accumulated.

  FIG. 4 is a diagram illustrating an example of setting information registered in the product information DB 101 after the product information setting unit 110 performs product information setting. Details of the product information setting unit 110 will be described later.

  As illustrated in FIG. 5, the similar product information DB 102 indicates a product code, a product name, a product code of a product similar to the product specified from the product code, and a product name corresponding to the similar product code. This is a database in which information including a similar product name, a representative seasonal index flag (details will be described later), a release date of similar products, a sale date of similar products, an end-of-sale period of similar products, and a weight (details will be described later) is accumulated.

  FIG. 5 is a diagram illustrating an example of setting information registered in the similar product information DB 102 after the similar product information setting unit 120 performs similar product information setting. Details of similar product information setting will be described later.

  As shown in FIG. 6, the similar product seasonal index information DB 103 is a database in which similar product codes, target cycle serial numbers (details will be described later), and seasonal indices (details will be described later) are accumulated.

  FIG. 6 is a diagram illustrating an example of similar product seasonal index information registered in the similar product seasonal index information DB 103 after performing the similar product seasonal index calculation execution process. Details of the similar product seasonal index calculation execution process will be described later.

  As shown in FIG. 7, the demand record DB 104 is a database in which the past demand record values for new products and similar products by product / day are accumulated. FIG. 7 is a diagram illustrating an example of information registered in the demand record DB 104.

  As shown in FIG. 8, the prediction model information DB 105 stores information including various prediction models (for example, moving average, exponential smoothing, etc.) necessary for performing demand prediction and parameters used in each prediction model. ing. FIG. 8 is a diagram illustrating an example of information registered in the prediction model information DB 105.

  As shown in FIG. 9, the prediction result DB 106 is a database in which product codes, dates on which demand is expected, and demand prediction values are accumulated. FIG. 9 is a diagram illustrating an example of information registered in the prediction result DB 106 after performing the prediction execution process. Details of the prediction execution process will be described later.

  FIG. 2 is a diagram illustrating an example of a hardware configuration of the demand prediction apparatus 100. In FIG. 2, the CPU 201 comprehensively controls each device connected to the system bus 212. The RAM 202 functions as a main memory, work area, temporary save area, and the like for the CPU 201. The ROM 203 stores a boot program.

  The input control unit 204 controls input from the input unit 205 realized by a pointing device such as a keyboard or a mouse. The display control unit 206 controls display by the display unit 207 realized by a CRT monitor, a liquid crystal monitor, or the like.

  The external memory control unit 208 controls access to the external memory 209. The external memory 209 stores an operation system (OS) that is a control program of the CPU 201, a program for realizing the functions of the demand prediction apparatus 100, the above-described various databases (DB), files, applications, master files, user files, and the like. To do.

  Further, the external memory 209 further stores a table or the like for realizing the function of the demand prediction device 100. Examples of the external memory 209 include a hard disk (HD), a flexible disk (FD), a compact flash (registered trademark) that can be connected to a PCMCIA card slot via an adapter, and smart media.

  Further, the processing or function in the present embodiment is realized by the CPU 201 loading a program recorded in the external memory 209 to the RAM 202 as necessary and executing it.

  The communication I / F control unit 210 executes communication control processing between the external device and the demand prediction device 100 via a network 211 such as a LAN or the Internet.

  FIG. 3 is a flowchart illustrating an example of processing of the demand prediction apparatus 100. In this process, the computer recorded in the external memory 209 or the like is stored in each functional unit (product information setting unit 110, similar product information setting unit 120, similar product seasonal index calculation execution unit 130, and prediction execution shown in FIG. The CPU 201 performs the control according to a program for causing the computer to function as the unit 140 and the prediction result confirmation unit 150).

  In step S301, the product information setting unit 110 sets information necessary for new product demand prediction. At this time, the product information setting unit 110 displays a product information setting screen as shown in FIG. 10 on the display unit 207 to prompt the user to input.

  And the merchandise information setting part 110 receives the information input via the merchandise information setting screen. Note that the user sets product information for which new product demand prediction is to be performed with respect to the setting of the product information item.

  In the product information, a product code, a product name, a release date, a variable period, a variable period seasonality flag, an initial sales quantity, and the like are set.

  The fluctuation period means a period from the release until the demand stabilizes. For example, in a monthly forecast, if the release date is November 1, 2012 and the fluctuation period is 4 (months), the demand will stabilize from March 2013.

  In this period, for example, it is assumed that the demand will be affected by the sales side's intended strategy to increase demand, so the period when demand stabilizes is defined as a specific period separately. .

  The initial sales volume means the planned sales volume for the entire cycle with a new product release date. For example, in the case of monthly prediction, the initial sales volume is the sales planned amount for the release month, and in the case of weekly prediction, it is the sales planned amount for the sales week.

  By setting the initial sales quantity, it is possible to grasp the demand scale of new products before release.

  The fluctuation period seasonality presence / absence flag means whether the new product to be predicted is affected by seasonality within the fluctuation period.

  Immediately after launch, there are some products whose demand changes greatly due to strategic factors rather than seasonality. For this reason, products that are entered as none in the variable period seasonality flag field do not reflect seasonality in the forecast value within the variable period, while products that enter yes have seasonality in the forecast value within the variable period. Is reflected. The product information input on the product information setting screen is registered in the product information DB 101.

  In step S302, the similar product information setting unit 120 sets a similar product for the new product. At this time, the similar product information setting unit 120 displays a similar product information setting screen as shown in FIG. 11 on the display unit 207 to prompt the user to input.

  Then, the similar product information setting unit 120 receives information input via the similar product information setting screen. Here, FIG. 11 shows an example of a similar product information setting screen.

  Note that the user can set similar product information for a new product, and a plurality of similar products can be set for one new product. The product code and the product name can be directly input, or can be selected from the product information set in step S301.

  In the similar product information, a similar product code, a similar product name, a representative seasonal index flag, a sale date, a sale date, a sale period, a weight, and the like are set.

  Here, the similar product code means an already-released product to be referred to when performing new product prediction, and the release date / sale date is input for the similar product.

  Further, the term “sale period” means a period of decline in demand before the sale, and “weight” means the degree of influence of similar products on the new product (for example, the degree of change in demand amount is similar). The total weight of similar products is set to 1 for new products. The weight will be described again in the execution of prediction in step S304.

  Note that a new product and a similar product can be selected from a list box for displaying a list of product codes and product names, as shown in the lower left of the similar product information setting screen. Information on this product is acquired from the product information DB 101.

  The similar product information input on the similar product information setting screen is registered in the similar product information DB 102.

  Returning to the description of FIG. 3 again, in step S303, in the similar product seasonal index calculation execution process, the seasonal index of the similar product is calculated from the past demand record of the similar product and registered in the similar product seasonal index information DB 103.

  FIG. 6 is a diagram illustrating an example of information registered in the similar product seasonal index information DB 103 after execution of the similar product seasonal index calculation process as described above.

  Here, the seasonal index is obtained by quantifying fluctuations in demand due to the season. The higher the numerical value, the greater the demand, and the lower the value, the smaller the demand.

  The DB registration items include a similar product code, a target cycle serial number, a seasonal index, and the like.

  The target cycle serial number means a serial number from the beginning of the year that indicates when the calculated seasonal index is. For example, when the prediction is monthly, if the target cycle serial number is 1, it means January, and if the target cycle serial number is 2, it means the seasonal index in February.

  When the prediction is weekly, when the target cycle serial number is 1, it means the first week from the beginning of the year, and when the target cycle serial number is 2, it means the seasonal index of the second week from the beginning of the year.

  Details of the processing in step S303 will be described with reference to FIG.

  FIG. 12 is a flowchart for explaining the details of the similar commodity seasonal index calculation execution process in step S303.

  In step S1201, new product information is acquired from the product information DB 101. The new product information here refers to the products for which demand is to be forecasted. Specifically, it is acquired for products before release, products within the variable period, and products that are less than one year old after the variable period ends. To do.

  In step S1202, the similar product information of the new product acquired in step S1201 is acquired from the similar product information DB.

  In this step, the similar product information having the product code of the similar product information corresponding to the product code included in the new product information is acquired.

  In step S1203, the demand record of the similar product acquired in step S1202 is acquired from the demand record DB 104.

  In this step, a demand record corresponding to the product code in the demand record DB 104 corresponding to the product code included in the similar product information is acquired.

  Furthermore, in this step, a temporary increase in demand due to an event such as a sales strategy performed for similar products is corrected.

  The reason for this correction is that an increase in demand due to an event is considered an abnormal value, and if a seasonal index including the abnormal value is calculated, an incorrect seasonal index may be calculated.

  Therefore, there are various correction methods, but if the demand for similar products is almost constant in the same month every year, it is easiest to correct by the average of actual values in the same month in the year when no event is held. This is a correction method.

  If the amount of demand varies greatly from year to year, there is also a concept that a trend line is drawn from data before and after the start of event execution, and the value on the line is used as the correction value for the event execution period. In some cases, correction is performed by specifying a correction quantity or a correction rate.

  Note that the occurrence period of the event is recognized in advance by the user, and the demand amount in the corresponding period may be corrected by inputting the occurrence period (screen not shown).

  In step S1204, the demand results after event correction of similar products acquired in step S1203 are totaled in accordance with the prediction cycle. For example, in the case of monthly prediction, the actual demand is aggregated in monthly units, and in the case of weekly prediction, the actual demand is aggregated in weekly units.

  Here, the demand performance acquisition period of similar goods is demonstrated with reference to FIG. FIG. 13 shows a demand record acquisition period of similar products when predicted on a monthly basis.

  As for the demand record of the similar product whose product code (similar product code) is R0001 shown in FIG. 13, the release month is January, the period from January to 3 months indicates the variable period, and the end-of-sale period is May. The last sales month is July. The current date is September.

  Further, it is shown that an event has occurred in December, it is shown that the amount of demand has suddenly increased due to the event, and FIG. 13 shows that the correction has been performed. This correction method is as described above.

  The change period of the new product is considered to be the same for the similar product, and the demand result acquisition period of the similar product is acquired by excluding the change period of the new product acquired in step S1201.

  Further, if sold out, it is acquired by excluding the end-of-sale period acquired in step S1202.

  This may be greatly influenced by sales strategy for a certain period immediately after the sale or before the end of sale, and may not represent the original demand. For this reason, it is considered preferable to remove the period from the result acquisition period.

  That is, in consideration of the exclusion period, in FIG. 13, the period from April 2010 to May 2011 is the demand result acquisition period.

  In addition, the fluctuation period and the end-of-sale period of new products are excluded, but for similar products, there are products with demand record information for more than one year. Is also possible.

Returning to the description of FIG. 12 again, in step S1205, the seasonal index of the similar product is calculated from the actual demand for each cycle of the similar product tabulated in step S1204. There are various ways to calculate the seasonal index. For example, if the forecast is monthly,
[Formula 1]

Seasonal index = actual demand for the month / average actual demand per month x 100 (%)

Calculate the seasonal index for each month.

  If similar products do not have seasonality, the seasonal index may be set to 1. Here, the average demand record of similar products is calculated from an arbitrary period in the demand record acquisition period acquired in step S1203.

  It is preferable to calculate from a period of one year or more. The average value calculation period may be fixed in the calculation, or the calculation period may be input by the product information setting unit 110. The above-described seasonal index calculation method is merely an example, and may be calculated by other methods.

  In step S1206, it is determined whether the similar product seasonal index calculation processing has been completed for all new products. If the processing for all the products has not been completed, the process returns to step S1202 and the next product is processed.

  Returning to the description of FIG. 3 again, in step S304, in the prediction execution process, the demand forecast of the new product is performed by using the demand result data of the similar product and the seasonal index of the similar product. In step S304, a product registered in the product information DB 101 is a processing target.

  Details of the processing in step S304 will be described with reference to FIG.

  FIG. 14 is a flowchart for explaining details of the prediction execution processing in step S304.

  In step S1401, the new product to be predicted (as described above, the new product here indicates the product to be predicted for demand, specifically, the product before release, within the variable period The product of less than one year from the end of the change period is targeted.) (Hereinafter, forecast product) similar product information is acquired from the similar product information DB 102, and the actual demand for the acquired similar product is obtained in step S1203, step Similar to the processing described in S1204, the data is acquired from the demand result DB 104, and the results are totaled according to the prediction cycle.

  Further, the actual demand for the predicted product is also aggregated in accordance with the prediction cycle. However, regarding the demand result acquisition period of the predicted product and the similar product, since the entire period after the release date is acquired, the limitation on the acquisition period described in step S1204 is not considered.

  In step S1402, whether or not the predicted product is pre-release is determined using the date (system date or the like) on which this processing is performed and the release date. If the predicted product is before release, the process proceeds to step S1403. If the predicted product is released, the process proceeds to step S1404.

  In step S1403, at the time before the predicted product is released, the predicted demand after the release is calculated. FIG. 15 is a flowchart for explaining the details of the prediction execution process in step S1403 when the predicted product is not yet released.

  In step S1501, it is determined whether or not the predicted product has seasonality within the fluctuation period. If the predicted product has no seasonality within the variable period, the process proceeds to step S1502, and if the predicted product has seasonality within the variable period, the process proceeds to step S1503.

  Information about whether or not the predicted product has seasonality within the change period is registered as a change period seasonality presence / absence flag in the product information DB 101.

BD(i)：当該予測商品の時点iのベース需要量
RｊD(i)：類似商品jの時点iの需要量（イベント補正後需要量）
RｊW：類似商品jの重み
N：変動期間
M:類似商品設定数

ここでｉは、当該予測商品ならびに類似商品の発売サイクル（発売月、発売週等）を基点とした各時点を意味する。例えば、予測単位が月の場合、ｉが１の場合は、当該予測商品ならびに類似商品の発売月を意味する。 In step S1502, if the predicted product is not affected by seasonal variation within the variation period (when there is no seasonality), the base within the variation period of the predicted product from a plurality of similar products set in the predicted product. Calculate demand.
The base demand amount means the demand power possessed by the predicted product, and is calculated by the following formula from the demand amount of similar products and the weighted sum of the weights.
[Formula 2]

BD (i): Base demand at the time i of the forecasted product
RjD (i): Demand volume at time point i for similar product j (event-corrected demand volume)
RjW: Weight of similar product j
N: Change period
M: Number of similar product settings

Here, i means each time point based on the sales cycle (release month, release week, etc.) of the predicted product and similar products. For example, when the prediction unit is month and i is 1, it means the release month of the predicted product and similar products.

  When i is 2, it means the month following the release month of the predicted product and similar products. M represents the number of similar products associated with the predicted product, and the weight of each similar product is acquired from the information on the similar product registered in the similar product information DB 102.

  Further, the demand amount of the similar product is the demand amount after the event correction calculated in step S1401, and the fluctuation period is acquired from the product information DB 101.

BDｊ(i)＝RｊD(i)／RｊSB（i）

BD(i)：当該予測商品の時点iのベース需要量
BDｊ(i)：類似商品jの時点iのベース需要量
RｊSA（i）：当該予測商品の時点iに該当するサイクルの類似商品ｊの季節指数
RｊW：類似商品jの重み
RｊD(i)：類似商品jの時点iの需要量（イベント補正後需要量）
RｊSB（i）：類似商品jの時点iに該当するサイクルの類似商品ｊの季節指数
N：変動期間
M:類似商品設定数

ここで、類似商品ｊの時点ｉのベース需要量とは、ステップＳ１４０１で算出したイベント補正後の類似商品の需要量から類似商品の季節性を除去した需要量である。 In step S1503, when the predicted product is affected by seasonal variation within the variation period (when there is seasonality), the base within the variation period of the predicted product from a plurality of similar products set in the predicted product Calculate demand. The base demand amount is calculated by the following formula in consideration of seasonality according to the release time of the predicted product from the demand amount of similar products.
[Formula 3]

BDj (i) = RjD (i) / RjSB (i)

BD (i): Base demand at the time i of the forecasted product
BDj (i): Base demand at the time i of the similar product j
RjSA (i): Seasonal index of similar product j in the cycle corresponding to point i of the forecasted product
RjW: Weight of similar product j
RjD (i): Demand volume at time point i for similar product j (event-corrected demand volume)
RjSB (i): Seasonal index of similar product j in cycle corresponding to time point i of similar product j
N: Change period
M: Number of similar product settings

Here, the base demand amount at the time point i of the similar product j is a demand amount obtained by removing the seasonality of the similar product from the demand amount of the similar product after the event correction calculated in step S1401.

  In addition, the seasonal index of the similar product j in the cycle corresponding to the time point i of the predicted product is, for example, when the prediction unit is month, and when the cycle corresponding to the time point i is 5 in the predicted product is March, This means the March seasonal index of similar products j.

  On the other hand, the seasonal index of the similar product j in the cycle corresponding to the time point i of the similar product j is similar if, for example, the prediction unit is month, and the cycle corresponding to the time point i in the similar product is 6 is October. It means the October season index of product j. The seasonal index of similar products is acquired from the similar product seasonal index information DB 103.

  As described above, the base demand amount is calculated from the weighted sum of the demand amount obtained by adding seasonality to the base demand amount of the similar product according to the release cycle of the predicted product and the weight of the similar product.

  FIG. 16 is a diagram illustrating a base demand amount calculation process (in consideration of seasonality) within the fluctuation period of the predicted product when the prediction unit is a monthly unit. FIG. 16 is an example in which one similar product is set as the predicted product, and the variation period is 3 months.

  The upper graph of FIG. 16 shows the result of event correction performed on the actual demand of similar products whose product code (similar product code) shown in FIG. 13 is R0001. , The seasonal index of each similar product is shown.

  The graph on the left side of the lower part of FIG. 16 shows a case where seasonality is recognized within the fluctuation period with respect to the actual demand of the similar product whose product code (similar product code) shown in the upper part of FIG. 16 is R0001. Using the seasonal index of each month within the fluctuation period, the result of calculating the base demand for each month without seasonality is shown.

  For example, the actual demand in January without removing the seasonality is 171 from the upper graph of FIG. 16, and the actual demand is calculated using the January seasonal index 150% shown in the upper right of FIG. As a result of the correction, the January base demand 114 shown on the left side of the lower part of FIG. 16 is obtained.

  And the graph which showed the result of having calculated | required the base demand amount of the prediction object goods using the base demand amount calculated | required by removing an event and seasonality is shown by the lower right side of FIG.

  The forecast target product will be released in October and the release date of similar products is January, so the release date will be different, and in order to determine the base demand of the forecast target product, special consideration must be given to seasonality Also occurs.

  Therefore, with respect to the base demand volume in January, which is the release date of similar products (shown on the left side of the lower part of FIG. 16), the seasonal index of the similar products in October, which is the release date of the forecast target product (upper part of FIG. 16). (Shown on the right side of FIG. 16) is used to obtain the base demand amount 80 in October (the graph on the right side of the lower part of FIG. 16), which is the release date of the forecast target product.

In step S1504, the base demand amount calculated in steps S1502 and S1503 is used to perform prediction calculation of demand within the fluctuation period of the predicted product. The predicted quantity at time i within the variable period is calculated using the following formula based on the actual demand for similar products.
[Formula 4]

Y (i) = α × BD (i) (1 ≦ i ≦ N)
α = SY / BD (1)

Y (i): Expected quantity at time i
BD (i): Base demand at the time i of the forecasted product
SY: Initial forecast quantity of the forecast product
N: Change period α: Demand scale adjustment factor

Here, BD (1) means the base demand amount of the predicted product calculated from a plurality of similar products with the point in time i being 1 (release cycle), and α is the initial predicted quantity of the predicted product and the base of the release cycle. It is a coefficient that adjusts the demand scale in demand. For example, there is a possibility that the base demand amount of the sales cycle will increase due to an event such as a sales strategy performed at the time of sales for the predicted product. The coefficient that adjusts the increase is the demand scale adjustment coefficient.

  By using the demand scale adjustment coefficient, it is possible to calculate a predicted quantity corresponding to an increase in demand due to an event while capturing a wave of demand for similar products. Note that the initial predicted quantity of the predicted product is registered in the product information DB 101 (initial sales quantity). Thus, the predicted quantity within the fluctuation period is calculated by capturing the wave of demand during the fluctuation period of the similar product (immediately after the sale of the similar product). The predicted quantity calculated at this time is registered in the prediction result DB 106.

  FIG. 17 is a diagram illustrating a predicted quantity calculation process within a variable period when the prediction unit is a monthly unit. FIG. 17 shows an example in which the similar product set for the predicted product is one product and the variable period is three months.

  The graph shown on the left side of FIG. 17 is the graph shown on the right side of the lower part of FIG. 16, and shows the base demand amount of the prediction target product within the fluctuation period.

  However, since the initial forecast quantity for releasing the forecast target product is required to be 80, and the base demand amount in the subsequent fluctuation period is also determined based on this initial forecast quantity, When the initial release quantity is different from 80, for example, 100, the initial forecast quantity is corrected, and the scale of the initial forecast quantity after this correction is changed to the base of each month within the subsequent fluctuation period. It is necessary to correct the demand.

  Therefore, the October base demand shown in the graph on the left side of FIG. 17 is 80. Originally, since the initial forecast quantity is 100, the base demand amount for October shown in the graph on the right side of FIG. 100 is shown in FIG.

  Then, for each month, November, December,... In the subsequent fluctuation period, the graph on the left side of FIG. 17 is used to match the base demand in October shown in the graph on the right side of FIG. The value obtained by multiplying the base demand amount for each month by the initial forecast quantity (100, which is the base demand amount for October shown in the graph on the right side of FIG. 17), is shown in the graph on the left side of FIG. The base demand amount is obtained by dividing 80, which is the base demand amount in October.

Y（i）:時点iの予測数量
BD：当該予測商品のベース需要量
BDj：類似商品jのベース需要量
RｊD (i)：類似商品jの時点iの需要量（イベント補正後需要量）
RｊW：類似商品jの重み
RxSA（i）：当該予測商品の時点iに該当するサイクルの類似商品xの季節指数
RｊSB（i）：類似商品jの時点iに該当するサイクルの類似商品ｊの季節指数
N：変動期間
M:類似商品設定数
L:類似商品jのベース需要量算出期間
SY：当該予測商品の初回予測数量
α：需要スケール調整係数

ここで、類似商品のベース需要量は、各時点の類似商品の需要量から季節性を除外して一定期間の平均値を算出して求める。平均をとる期間Ｌは、固定値でもよいし、商品情報設定部１１０の入力項目として設定することも可能である。 In step S1505, prediction calculation after the fluctuation period of the predicted product is performed. The predicted quantity at the time point i after the fluctuation period is calculated by the following calculation formula based on the actual demand for similar products.
[Formula 5]
Y (i) = α x BD x RxSA (i) (N <i)

Y (i): Expected quantity at time i
BD: Base demand for the forecast product
BDj: Base demand for similar products j
RjD (i): Demand volume at time point i for similar product j (event-corrected demand volume)
RjW: Weight of similar product j
RxSA (i): Seasonal index of similar product x in the cycle corresponding to the time point i of the forecast product
RjSB (i): Seasonal index of similar product j in cycle corresponding to time point i of similar product j
N: Change period
M: Number of similar product settings
L: Base demand calculation period for similar products j
SY: Initial forecast quantity of the forecast product α: Demand scale adjustment factor

Here, the base demand amount of the similar product is obtained by calculating an average value for a certain period by excluding seasonality from the demand amount of the similar product at each time point. The average period L may be a fixed value, or may be set as an input item of the product information setting unit 110.

  Further, the base demand amount of the predicted product is calculated from the weighted sum of the calculated base demand amount of the similar product and its weight.

  The seasonal index of the similar product X means a seasonal index of the similar product to be given to the predicted product.

  A similar product with the representative seasonal index flag turned on is selected from the similar products information DB 102 among the similar products for the predicted product, and the seasonal index of the selected similar product is acquired from the similar product seasonal index information DB 103.

  If there are multiple similar products for which the representative seasonal index flag is on, the seasonal index is the average value.

  The seasonal index of the similar product X in the cycle corresponding to the time point i of the predicted product means the seasonal index of the cycle corresponding to the time point i of the predicted product. For example, when the prediction unit is a month, i is 5, and when it corresponds to March, the March seasonal index is applied. The demand scale adjustment coefficient is a coefficient calculated in step S1504.

  Thus, the prediction calculation after the fluctuation period can be obtained by adjusting the demand scale by adding the seasonality according to the release cycle of the predicted product to the base demand amount of the predicted product. . The predicted quantity calculated at this time is registered in the prediction result DB 106.

  FIG. 18 is a diagram illustrating a predicted quantity calculation process after the fluctuation period when the prediction unit is monthly. FIG. 18 shows an example in which the similar product set as the predicted product is one product, and the similar product base demand calculation period L is three months.

  The graph shown on the left side of the upper part of FIG. 18 shows the demand amount obtained by excluding seasonality with respect to the actual demand for similar products shown in the upper part of FIG. The correction is made for the period, but here, the correction is made for three months after the fluctuation period.

  After obtaining the demand amount after the event correction for the April demand result 50 in the upper graph of FIG. 16, the seasonality is removed by using the April seasonal index 110 shown on the right side of the upper portion of FIG. 45 (the April demand shown in the left graph in the upper part of FIG. 18), which is the base demand, is obtained.

  Then, after obtaining the demand amounts in May and June in the same manner, the base demand amount after the fluctuation period is obtained using the demand amounts from April to June (1 in the graph on the right side of the upper part of FIG. 18). Shows demand 47 after the month).

  When the base demand after the fluctuation period is obtained, the base demand 80 in the release month (see FIG. 18) is obtained for each month by multiplying the base demand after the fluctuation period by the initial predicted quantity 100. The value obtained by dividing by the demand amount in October of the graph shown on the upper right side of the upper stage is multiplied by the seasonal index of similar products in the month (showing the seasonal index on the right side of the upper stage in FIG. 16).

  Returning to the description of FIG. 14 again, in step S1404, it is determined whether or not the predicted product is released and the current time is within the fluctuation period of the predicted product, and if the current time is within the fluctuation period of the predicted product. The process proceeds to step S1405, and if the predicted product has been released and the fluctuation period has passed, the process proceeds to step S1406. Here, the fluctuation period of the predicted product is registered in the product information DB 101.

  In step S1405, the predicted quantity of the predicted product after the current time is calculated. FIG. 19 is a flowchart for explaining the details of the prediction execution processing in step S1405. The processing flow is the same as in FIG.

  In step S1901, it is determined whether or not the predicted product has seasonality within the variation period. If the predicted product does not have seasonality within the variation period, the process proceeds to step S1902, and the predicted product varies. If there is seasonality within the period, the process proceeds to step S1903.

  Information about whether or not the predicted product has seasonality within the change period is registered as a change period seasonality presence / absence flag in the product information DB 101.

BD(i)：当該予測商品の時点iのベース需要量
RｊD(i)：類似商品jの時点iの需要量（イベント補正後需要量）
RｊW：類似商品jの重み
N：変動期間
M:類似商品設定数
S:現時点（当該予測商品が発売されてSサイクル経過した時点）

ステップＳ１９０３において、当該予測商品が変動期間内に季節変動による影響を受ける場合（季節性がある場合）、当該予測商品に設定されている複数の類似商品から当該予測商品の変動期間内のベース需要量を算出する。 In step S1902, if the predicted product is not affected by seasonal variation within the variation period (when there is no seasonality), the base within the variation period of the predicted product from a plurality of similar products set in the predicted product. Calculate demand. Similar to step S1502, the base demand amount is calculated from the demand amount of similar products (event corrected demand amount) and the weighted sum of the weights using the following equation.

[Formula 6]

BD (i): Base demand at the time i of the forecasted product
RjD (i): Demand volume at time point i for similar product j (event-corrected demand volume)
RjW: Weight of similar product j
N: Change period
M: Number of similar product settings
S: Current time (when the predicted product is released and S cycles have passed)

In step S1903, if the predicted product is affected by seasonal variation within the variation period (when there is seasonality), the base demand within the variation period of the predicted product from a plurality of similar products set in the predicted product Calculate the amount.

BDｊ(i)＝RｊD(i)／RｊSB（i）

BD(i)：当該予測商品の時点iのベース需要量
BDｊ(i)：類似商品jの時点iのベース需要量
RｊSA（i）：当該予測商品の時点iに該当するサイクルの類似商品ｊの季節指数
RｊW：類似商品jの重み
RｊD(i)：類似商品jの時点iの需要量（イベント補正後需要量）
RｊSB（i）：類似商品jの時点iに該当するサイクルの類似商品ｊの季節指数
N：変動期間
M:類似商品設定数
S:現時点（当該予測商品が発売されてSサイクル経過した時点）

ステップＳ１９０４において、ステップＳ１９０２、ステップＳ１９０３で算出したベース需要量を用いて当該予測商品の変動期間内の予測計算を行う。ステップＳ１５０４と同様に変動期間内のｉ時点の予測数量を類似商品の需要実績を基に以下の計算式で算出する。

［式８］

Y(i)＝α×BD（i） (S≦i≦N)

Y（i）:時点iの予測数量
BD(i)：当該予測商品の時点iのベース需要量
BD(l)：当該予測商品の時点lのベース需要量
D（l）：当該予測商品の時点ｌの需要量（イベント補正後需要量）
N：変動期間
S:現時点（当該予測商品が発売されてSサイクル経過した時点）
α：需要スケール調整係数

ここで、ステップＳ１５０４との違いは、需要スケール調整係数の算出方法である。 Similar to step S1903, the base demand is calculated from the demand of similar products (demand after event correction) in consideration of seasonality according to the release time of the predicted product by the following formula.
[Formula 7]

BDj (i) = RjD (i) / RjSB (i)

BD (i): Base demand at the time i of the forecasted product
BDj (i): Base demand at the time i of the similar product j
RjSA (i): Seasonal index of similar product j in the cycle corresponding to point i of the forecasted product
RjW: Weight of similar product j
RjD (i): Demand volume at time point i for similar product j (event-corrected demand volume)
RjSB (i): Seasonal index of similar product j in cycle corresponding to time point i of similar product j
N: Change period
M: Number of similar product settings
S: Current time (when the predicted product is released and S cycles have passed)

In step S1904, using the base demand calculated in steps S1902 and S1903, prediction calculation within the fluctuation period of the predicted product is performed. Similar to step S1504, the predicted quantity at time i within the variable period is calculated using the following formula based on the actual demand for similar products.

[Formula 8]

Y (i) = α × BD (i) (S ≦ i ≦ N)

Y (i): Expected quantity at time i
BD (i): Base demand at the time i of the forecasted product
BD (l): Base demand at the time l of the forecasted product
D (l): Demand volume at the time l of the forecasted product (event-corrected demand volume)
N: Change period
S: Current time (when the predicted product is released and S cycles have passed)
α: Demand scale adjustment factor

Here, the difference from step S1504 is the calculation method of the demand scale adjustment coefficient.

  Since the predicted product has already been released, the demand amount of the predicted product is accumulated in the demand result DB 104. The demand scale adjustment coefficient is calculated from the ratio of the demand amount (event corrected demand amount) of the predicted product already existing and the base demand amount of the predicted product calculated from the similar product. The predicted quantity calculated at this time is registered in the prediction result DB 106.

Y（i）:時点iの予測数量
BD：当該予測商品のベース需要量
BDj：類似商品jのベース需要量
RｊD(i)：類似商品jの時点iの需要量（イベント補正後需要量）
RｊW：類似商品jの重み
RxSA（i）：当該予測商品の時点iに該当するサイクルの類似商品xの季節指数
RｊSB（i）：類似商品jの時点iに該当するサイクルの類似商品ｊの季節指数
N：変動期間
M:類似商品設定数
L:類似商品jのベース需要量算出期間
S:現時点（当該予測商品が発売されてSサイクル経過した時点）
BD(l)：当該予測商品の時点lのベース需要量
α：需要スケール調整係数

ここで、ステップＳ１５０５との違いは、需要スケール調整係数の算出方法である。 In step S1905, prediction calculation after the fluctuation period of the predicted product is performed. Similar to step S1505, the predicted quantity at the time point i after the fluctuation period is calculated by the following calculation formula based on the actual demand for similar products.

[Formula 9]

Y (i) = α x BD x RxSA (i) (N <i)

Y (i): Expected quantity at time i
BD: Base demand for the forecast product
BDj: Base demand for similar products j
RjD (i): Demand volume at time point i for similar product j (event-corrected demand volume)
RjW: Weight of similar product j
RxSA (i): Seasonal index of similar product x in the cycle corresponding to the time point i of the forecast product
RjSB (i): Seasonal index of similar product j in cycle corresponding to time point i of similar product j
N: Change period
M: Number of similar product settings
L: Base demand calculation period for similar products j
S: Current time (when the predicted product is released and S cycles have passed)
BD (l): Base demand amount at the time l of the forecast product α: Demand scale adjustment factor

Here, the difference from step S1505 is the method for calculating the demand scale adjustment coefficient.

  The demand scale adjustment coefficient is a coefficient calculated in step S1904.

  Thus, the prediction calculation after the fluctuation period is obtained by adjusting the demand scale by adding the seasonality according to the release cycle of the predicted product to the base demand amount of the predicted product in the same manner as in step S1505. It becomes possible. The predicted quantity calculated at this time is registered in the prediction result DB 106.

  Returning to the description of FIG. 14 again, in step S1406, it is determined whether or not the predicted product has been released and the current time is within one year from the end of the fluctuation period of the predicted product. If it is within one year from the end, the process proceeds to step S1407, and if the predicted product has passed one year from the end of the variable period, the process proceeds to step S1408. Here, the fluctuation period of the predicted product is registered in the product information DB 101.

  In step S1407, the predicted quantity of the predicted product after the current time is calculated. FIG. 20 is a flowchart for explaining the details of the prediction execution processing in step S1407.

In step S2001, the base demand amount calculation calculation after the fluctuation period of the predicted product is performed. The base demand amount at the time i after the fluctuation period is calculated by the following formula.
[Formula 10]

BD (i) = D (i) / RxSA (i)

BD (i): Base demand at the time i of the forecasted product
D (i): Demand volume at the time i of the forecasted product (event-corrected demand volume)
RxSA (i): Seasonal index of similar product x in the cycle corresponding to the time point i of the forecast product

Here, the seasonal index of the similar product X means a seasonal index of the similar product to be given to the predicted product. A similar product with the representative seasonal index flag turned on is selected from the similar products information DB 102 among the similar products for the predicted product, and the seasonal index of the selected similar product is acquired from the similar product seasonal index information DB 103. If there are multiple similar products for which the representative seasonal index flag is on, the seasonal index is the average value.

  In step S2002, the base predicted quantity for each cycle after the fluctuation period of the predicted product is calculated from the base demand amount of the predicted product calculated in step S2001.

  Here, by cycle means by month or week. Further, when the base predicted quantity calculation is performed, the base predicted quantity is calculated using a mathematical prediction model registered in the prediction model information DB 105.

In step S2003, a predicted quantity calculation for each cycle after the fluctuation period of the predicted product is calculated from the base predicted quantity of the predicted product calculated in step S2002 by the following formula.
[Formula 11]

Y (i) = BY (i) x RxSA (i)

Y (i): Expected quantity at time i
BY (i): Base forecast quantity at the time i of the forecasted product
RxSA (i): Seasonal index of similar product x in the cycle corresponding to the time point i of the forecast product

FIG. 21 is a diagram illustrating a predicted quantity calculation process after the present time when the prediction unit is a monthly unit. FIG. 21 shows an example in which one similar product is set as the predicted product and the variation period is three months. The predicted quantity calculated at this time is registered in the prediction result DB 106.

  In FIG. 21, the current time indicates April 2012. Since the predicted product has already been released in October 2010, the demand amount from October 2010 to March 2011 is used as it is.

  However, the predicted quantity after April 2012 is shown in the upper right of FIG. 21 in order to obtain the base demand without the seasonality with respect to the demand in January, February and March 2012. The base demand is calculated using the seasonal index. Then, the predicted quantity after the present time is obtained using the obtained base demand.

  Returning to the description of FIG. 14 again, in step S1408, since the predicted product has been released and the current time has passed since the end of the fluctuation period of the predicted product, the demand amount of the predicted product (demand after event correction) Calculate the predicted quantity from (quantity).

  Here, the fluctuation period of the predicted product is registered in the product information DB 101. The processing flow is the same as in FIG. 20, and the difference from step S2001 to step S2003 is that the seasonal index of the similar product uses the seasonal index of the predicted product. The predicted quantity calculated at this time is registered in the prediction result DB 106.

  Returning to the description of FIG. 14 again, in step S1409, when the prediction calculation of all the products is completed, the process ends. When there are uncalculated products, the process returns to step S1401.

  Returning to the explanation of FIG. 3 again, in step S305, the predicted value of the new product calculated in step S304 from the similar product information DB 102 and the prediction model information DB 105 and the demand result of the similar product are displayed on the screen, and the validity of the calculated value is displayed. Check sex.

  FIG. 22 shows an example of a new product demand output screen displayed in step S305. From the screen, it is possible to correct the value calculated in each step.

  In this new product demand output screen, the demand amount of a new product and the demand amount of a product similar to the new product are illustrated for each release cycle. In FIG. 22, from January to December 2013, Indicates demand.

  The new product demand output screen displays a similar product demand amount indicating the demand amount of a product similar to the new product, a new product demand amount indicating the demand amount if the new product has already been released, and a new product. Items such as a new product forecast quantity indicating the forecast quantity and a forecast correction value capable of correcting the forecast quantity.

  As for new product demand, in FIG. 22, it is shown that there was demand from January to March 2013, and the forecast correction value shows the new product forecast quantity at the initial output. This is an item that allows the user to input a correction value as appropriate. As described above, according to the present invention, by using a demand record of a product similar to a product for which demand prediction is performed, it is possible to accurately perform demand prediction from the start to the end of the product to be predicted.

  Although the embodiment has been described in detail above, the present invention can take an embodiment as, for example, a method, a program, a storage medium, or the like. Specifically, the present invention is applied to a system composed of a plurality of devices. The present invention may be applied, or may be applied to an apparatus composed of one device.

  The program according to the present invention is a program that allows a computer to execute (read) each processing method, and the storage medium according to the present invention stores a program that allows the computer to execute each processing method.

  The program in the present invention may be a program for each processing method of each device.

  As described above, a recording medium that records a program that implements the functions of the above-described embodiments is supplied to a system or apparatus, and a computer (or CPU or MPU) of the system or apparatus stores the program stored in the recording medium. It goes without saying that the object of the present invention can also be achieved by performing reading.

  In this case, the program itself read from the recording medium realizes the novel function of the present invention, and the recording medium storing the program constitutes the present invention.

  As a recording medium for supplying the program, for example, a flexible disk, hard disk, optical disk, magneto-optical disk, CD-ROM, CD-R, DVD-ROM, magnetic tape, nonvolatile memory card, ROM, EEPROM, silicon A disk or the like can be used.

  Further, by executing the program read by the computer, not only the functions of the above-described embodiments are realized, but also an OS or the like operating on the computer based on an instruction of the program is a part of the actual processing or It goes without saying that the case where the functions of the above-described embodiments are realized by performing all of the above processing is also included.

  Furthermore, after the program read from the recording medium is written to the memory provided in the function expansion board inserted into the computer or the function expansion unit connected to the computer, the function expansion board is based on the instructions of the program code. It goes without saying that the case where the CPU or the like provided in the function expansion unit performs part or all of the actual processing and the functions of the above-described embodiments are realized by the processing.

  Further, the present invention may be applied to a system composed of a plurality of devices or an apparatus composed of a single device.

  Needless to say, the present invention can be applied to a case where the present invention is achieved by supplying a program to a system or apparatus. In this case, by reading a recording medium storing a program for achieving the present invention into the system or apparatus, the system or apparatus can enjoy the effects of the present invention.

  Furthermore, by downloading and reading a program for achieving the present invention from a server, database, etc. on a network using a communication program, the system or apparatus can enjoy the effects of the present invention. In addition, all the structures which combined each embodiment mentioned above and its modification are also included in this invention.

100 demand prediction device 201 CPU
202 RAM
203 ROM
204 Input control unit 205 Input unit 206 Display control unit 207 Display unit 208 External memory control unit 209 External memory 210 Communication I / F control unit 211 Network 212 System bus

Claims (11)

A demand forecasting device for forecasting demand for new products,
A demand record storage means for storing a demand record including a demand amount of a product and a day when there was a demand;
Similar product information setting means for selecting a product similar to the new product;
Calculating means for calculating a base demand amount of the new product from a demand amount of the similar product selected by the similar product information setting means;
Using the base demand amount of the new product calculated by the calculating means, a prediction execution means for obtaining a demand amount after the release start date of the new product;
A demand forecasting device characterized by comprising:   The prediction execution means obtains the demand amount of the new product within the fluctuation period using the base demand amount of the new product, while the fluctuation period indicating a period from the release to the stabilization of the demand, 2. The demand forecasting apparatus according to claim 1, wherein after the period, the demand amount of the new product is obtained using the base demand amount of the new product for a certain period.   The demand prediction apparatus according to claim 1, wherein the calculation unit calculates the base demand amount by excluding a period from the release to the time when the demand is stabilized.   The demand prediction device according to claim 1, wherein the calculation unit calculates a base demand amount excluding a drop period of demand before the sale.   The prediction execution means uses the base demand of the new product in consideration of the seasonality in the fluctuation period indicating the period from the release to the stabilization of the demand, or on the other hand, the new product without considering the seasonality. The demand prediction apparatus according to any one of claims 1 to 4, wherein a demand amount is obtained by selecting whether to use a base demand amount. For multiple similar products for a new product, we ’ll assign a weight to indicate the impact of the similar product on the new product,
6. The demand prediction according to claim 1, wherein the calculation unit calculates a base demand amount of the new product from a demand amount determined using a weight set for the similar product. apparatus. A demand prediction system in which a client terminal and an information processing device are connected via a network,
The client terminal is
Similar product information setting means for selecting products similar to the new product,
With
The information processing apparatus includes:
A demand record storage means for storing a demand record including a demand amount of a product and a day when there was a demand;
Calculating means for calculating a base demand amount of the new product from a demand amount of the similar product selected by the similar product information setting means;
Using the base demand amount of the new product calculated by the calculating means, a prediction execution means for obtaining a demand amount after the release start date of the new product;
A demand forecasting system characterized by comprising: A control method for a demand prediction device for predicting the demand for new products,
The demand prediction device
Similar product information setting means for selecting a product similar to the new product;
The base demand amount of the new product is calculated from the demand amount stored in the demand result storage means for storing the demand result of the similar product selected by the similar product information setting means, including the demand amount of the product and the date when there was a demand. A calculation means;
Using the base demand amount of the new product calculated by the calculating means, a prediction execution means for obtaining a demand amount after the release start date of the new product;
The control method of the demand prediction apparatus characterized by performing these. A program that can be read and executed by a demand prediction device that predicts the demand for new products,
The demand forecasting device,
Similar product information setting means for selecting a product similar to the new product;
The base demand amount of the new product is calculated from the demand amount stored in the demand record storage unit that stores the demand record of the similar item selected by the similar item information setting unit and the demand record including the demand date. Calculating means for
Using the base demand amount of the new product calculated by the calculating means, a prediction execution means for obtaining a demand amount after the release start date of the new product;
A program characterized by making it function. A control method of a demand prediction system in which a client terminal and an information processing device are connected via a network,
The client terminal is
Similar product information setting means for selecting products similar to the new product,
Run
The information processing apparatus includes:
The base demand amount of the new product is calculated from the demand amount stored in the demand record storage unit that stores the demand record of the similar item selected by the similar item information setting unit and the demand record including the demand date. Calculating means for
Using the base demand amount of the new product calculated by the calculating means, a prediction execution means for obtaining a demand amount after the release start date of the new product;
The control method of the demand prediction system characterized by performing this. There is a program that can be read and executed by a demand prediction system in which a client terminal and an information processing device are connected via a network,
The client terminal is
Similar product information setting means for selecting products similar to the new product,
Function as
The information processing apparatus;
The base demand amount of the new product is calculated from the demand amount stored in the demand record storage unit that stores the demand record of the similar item selected by the similar item information setting unit and the demand record including the demand date. Calculating means for
Using the base demand amount of the new product calculated by the calculating means, a prediction execution means for obtaining a demand amount after the release start date of the new product;
A program characterized by making it function.

Images