JP2007286977A - Demand forecasting method and demand forecasting program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a demand forecasting method and a demand forecasting program that more accurately forecast demand for merchandise of no or few order results. <P>SOLUTION: A management computer 21 comprises a transition data storage part 22 and an order result data storage part 23. The transition data storage part 22 stores transition data on the transition of service parts. The order result data storage part 23 stores order result data on the service parts. The management computer 21 acquires the order result data on a forecasting object, and if the forecasting object service part is a redesigned service part, identifies the older generations of service parts by the transition data. The management computer 21 acquires the order result data on the identified service parts, calculates a demand forecasting function from the order result data and the previously acquired order result data on the forecasting object, and calculates forecast values of the forecasting object service part from the demand forecasting function. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a demand prediction method and a demand prediction program for predicting the demand for goods.

  When providing products to customers, it is necessary to manage product inventory appropriately. This product includes not only finished products but also consumables used for finished products, replacement parts due to failure, and the like. And by carrying out accurate inventory management, it is possible to suppress inventory loss due to surplus inventory and opportunity loss due to shortage of inventory.

  Accurate demand forecasting is necessary for accurate inventory management. In such demand prediction, for example, multiple regression analysis is used. In this multiple regression analysis, past results are analyzed to create a prediction formula. However, if the prediction formula once created is continuously used, the error between the predicted value and the actual value may increase. Therefore, it is conceivable to recreate the prediction formula based on the results for each short period, but the load for creating the prediction formula increases.

  For this reason, a sales prediction method for performing sales prediction for each product category in consideration of the variation factor has been proposed (for example, see Patent Document 1). In this sales prediction method, first, a moving average value is calculated by averaging sales results of a product whose sales volume is predicted for a predetermined period. Then, the fluctuation forecast quantity of the sales quantity of the product is calculated from the fluctuation factors that are considered to affect the sales quantity on the forecast sales date. Further, the sales forecast quantity is calculated by correcting the moving average value based on the fluctuation forecast quantity. Thereby, the sales performance in the predetermined period immediately before the sales prediction date can be reflected in the sales prediction, and the followability of the predicted value can be improved.

  However, the demand prediction based on the moving average value has a large prediction error due to a time lag, and it is necessary to prepare a large amount of safety stock. In addition, when calculating the sales forecast quantity by correcting the moving average value based on the forecasted fluctuation quantity, identify the fluctuation factors that may affect the sales quantity on the sales forecast date, and then change the sales quantity of the product. It is necessary to calculate the predicted quantity. However, fluctuations in the sales volume of merchandise are often derived from various factors, and it is not easy to identify them.

Therefore, a demand forecasting method and a demand forecasting program capable of more efficiently and accurately forecasting demand for goods have been studied (for example, see Patent Document 2). In the technique described in this document, the management computer applies a growth model to the cumulative amount transition of the order quantity, and calculates a trend function for the order result. Next, the management computer calculates the transition of the difference between the order record and the trend function. Then, the periodic intensity of the difference transition is calculated using the periodogram. When the periodicity is determined based on the tuning strength, a periodic function is calculated by applying a quadratic Sin model composed of a quadratic function and a trigonometric function to the difference transition between the order record and the trend curve. Then, the management computer 21 performs demand prediction using a function generated by synthesizing the trend function and the periodic function.
JP 2000-339543 A (first page) JP-A-2004-234471 (first page)

In the techniques described in the above cited references 1 and 2, demand prediction is performed based on past sales. For this reason, it is difficult to predict demand for products with a small sales record. For example, as shown in FIG. 8, a new product whose design has been changed in February may have a sales record for only four months from March to June. In the sales performance in these four months, the number of orders received decreases every month. For this reason, if the demand forecast for the next month (July) or the month after next (August) is performed from this sales record, the demand forecast is less than the recent (June) sales record.

However, the downward trend in sales performance over the past four months may be affected by seasonal fluctuations. Thus, if the sales performance period is short, the error may increase.
The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a demand prediction method and a demand prediction program for predicting demand more accurately with respect to a product with no or few orders received. It is in.

  In order to solve the above-mentioned problem, the invention described in claim 1 includes a transition data storage unit that stores a product identifier that specifies a product and a product identifier that specifies a subsequent product of the product, and a time axis. Using a management computer connected to performance data storage means that records data relating to the actual value of product provision corresponding to the product identifier, the management computer comprising: A prediction target data acquisition step for acquiring specific product identifier data for specifying a product for which demand prediction is performed; a transition data extraction step for extracting a product identifier associated with the specific product identifier from the transition data storage unit; The actual value of the product with the specific product identifier and the actual value of the product with the extracted product identifier are recorded using the product identifier. The result extraction stage extracted from the means, the function calculation stage for calculating the demand prediction function of the product using the extracted actual value, and the predicted value of the product provision of the product for which the demand prediction is performed using the calculated demand prediction function And executing the demand forecast output stage for outputting the forecast value.

  The invention according to claim 2 is the demand prediction method according to claim 1, wherein the transition data storage means stores a product identifier of a past product up to the subsequent product in association with each other, and The transition data extracting step is to extract all product identifiers associated with the specific product identifier.

  According to a third aspect of the present invention, in the demand prediction method according to the first aspect, the transition data storage means stores a product identifier of a subsequent product and a product identifier of a previous generation product of the subsequent product in association with each other. The transition data extraction step includes a step of extracting a product identifier associated with the specific product identifier, and a product identifier of a product associated with the product of the extracted product identifier as a subsequent product. And performing the previous generation product extraction process to be extracted, and this previous generation product extraction process is repeated until the product identifier data cannot be extracted.

  According to a fourth aspect of the present invention, in the demand prediction method according to any one of the first to third aspects, in the data relating to the actual value associated with the extracted product identifier by the management computer, If the actual value is included, the total actual value calculating step of calculating the total value of the actual value of the same period is further executed, and the function calculating step uses the calculated total value to calculate the demand prediction function. The gist is to calculate.

The invention according to claim 5 relates to a transition data storage means for storing a product identifier for specifying a product and a product identifier for specifying a subsequent product of the product in association with each other, and an actual value of product provision corresponding to a time axis. A program for predicting demand for the product using a management computer connected to a record data storage means that records data associated with the product identifier, the management computer specifying the product for which the demand is predicted Prediction target data acquisition means for acquiring specific product identifier data, transition data extraction means for extracting a product identifier associated with the specific product identifier from the transition data storage means, actual value of the product of the specific product identifier and extracted Extracting the actual value of the product with the product identifier from the actual data storage means using the product identifier The function calculation means for calculating the demand prediction function of the product using the extracted actual value, and the predicted value of the product provision of the product for which the demand is predicted is calculated using the calculated demand prediction function. It is made to function as a demand prediction output means which outputs.

(Function)
According to the present invention, the management computer acquires specific product identifier data for specifying a product for which demand prediction is performed, and extracts a product identifier associated with the product identifier of the specific product identifier from the transition data storage unit. The management computer extracts the actual value of the product with the specific product identifier and the actual value of the product with the extracted product identifier from the actual data storage means, uses these actual values to calculate a demand forecast function for the product, A predicted value of product provision of the product to be performed is calculated. In general, if the number of past performance values is large, more accurate demand prediction can be performed. For this reason, when the number of data of the actual value of the succeeding product is small, a demand prediction function is calculated using the actual value data of the product preceding the succeeding product, and the demand prediction is performed using this function. That is, more accurate demand prediction can be performed by performing demand prediction of the subsequent product using the actual value of product provision of the product before the change of the product for which demand prediction is performed. Therefore, even if there is no or a small number of data of past performance values of products for which demand prediction is performed, the demand prediction of products can be performed more accurately. It should be noted that, here, “the product with which the product with the specific product identifier corresponds to the succeeding product” is changed not only the product immediately before the product for which the demand prediction is made (one generation before) but also a plurality of changes. In some cases, it includes products sold after each change.

  According to the present invention, the transition data storage means stores product identifiers of past products until the subsequent product is reached. For this reason, the management computer makes a demand prediction not only using the product immediately before this product but also using data relating to the actual values of all the products before reaching this product. Therefore, since the number of data of the actual value used for the demand prediction is increased, the demand prediction can be performed more accurately even for, for example, a succeeding product that has been repeatedly changed in a short period of time.

  According to the present invention, the transition data storage means stores the product identifier of the subsequent product and the product identifier of the previous generation product of the subsequent product in association with each other. The management computer performs the process of extracting the product identifier associated with the specific product identifier, and then performs the previous generation product extraction process of extracting the product identifier of the product associated with the product of the extracted product identifier as the subsequent product. Repeat until product identifier data cannot be extracted. For this reason, even if the product identifier of the past product until the subsequent product is not stored, for example, only the product before and after the change is stored, the past product until the product for which the demand is predicted is specified. can do.

  According to the present invention, the management computer, when the data regarding the actual value associated with the extracted product identifier includes the actual value of the same period, the total value of the actual values of these different product identifiers. The demand prediction function is calculated using this total value. For this reason, even if there is a time when different products before and after the change are spread simultaneously, more accurate demand prediction can be performed.

  ADVANTAGE OF THE INVENTION According to this invention, a demand can be predicted more correctly regarding goods with no order receipt results or few.

  Hereinafter, an embodiment of a demand prediction process embodying the present invention will be described with reference to FIGS. The present embodiment will be described as a demand forecasting method and a demand forecasting program that are used when forecasting demand for service parts based on the actual order of service parts of products provided to customers. Here, the service parts mean parts that are exchanged with service in the event of wear or failure. The service part is a minimum unit for maintaining the function of the product, and the service part includes not only a part but also a unit in which the parts are combined.

  In the present embodiment, as shown in FIG. 1, an order record is input using an order receiving system 10. Further, an order is instructed based on the demand forecast output to the order receiving system 10. The order receiving system 10 is installed in a service parts management department that manages service parts, and receives orders received at sales bases, service bases, etc., and outputs orders for production departments, purchasing departments, and the like.

  The order receiving system 10 is a computer terminal having a function of transmitting data, a function of displaying received data, and the like. The order receiving system 10 includes a CPU, RAM, and ROM (not shown), input means such as a keyboard and mouse, output means such as a display, communication means such as a communication interface, and the like.

  The order receiving system 10 is connected to the demand prediction system 20 via the network N. The demand prediction system 20 is a computer system that performs various data processing relating to demand prediction. The demand prediction system 20 includes a management computer 21.

  The management computer 21 performs data transmission / reception with the order receiving system 10, management processing of various data for executing demand prediction, and the like. This management computer 21 has a CPU, RAM, ROM, etc. (not shown), and will be described later (prediction target data acquisition stage, transition data extraction stage, performance extraction stage, function calculation stage, demand forecast output stage, and total performance calculation stage. Etc.). And the management computer 21 functions as a prediction object data acquisition means, a transition data extraction means, a performance extraction means, a function calculation means, a demand prediction output means, a total performance calculation means, etc. by executing the demand prediction program for this. To do.

  Here, the prediction target data acquisition means acquires specific product identifier data for specifying a product for which demand prediction is performed. The transition data extraction unit extracts a product identifier associated with the specific product identifier from the transition data storage unit. Further, the transition data extraction unit extracts all product identifiers associated with the specific product identifier. The result extracting means extracts the result value of the product of the specific product identifier and the result value of the extracted product of the product identifier from the result data storage means using the product identifier. When the management computer includes data on the actual value associated with the extracted product identifier and the actual value of the same period is included, the total actual value calculating means calculates the total value of the actual value of the same period. calculate. The function calculating means calculates a demand prediction function of the product using the extracted actual value. The function calculation means calculates the demand prediction function using the calculated total value. The demand prediction output means calculates a predicted value of product provision of a product for which demand prediction is performed using the calculated demand prediction function, and outputs the predicted value.

  Further, the demand prediction system 20 includes a transition data storage unit 22 as a transition data storage unit and an order record data storage unit 23 as a result data storage unit. The management computer 21 includes these data storage units (22, 23). )It is connected to the.

As shown in FIG. 2, the transition data storage unit 22 stores transition data 220 for specifying the transition of service parts. The transition data 220 is set when the design of the service part is first changed. Specifically, data relating to identifiers of service parts before and after the design change (that is, the first and second generation service parts) is recorded. Then, each time a design change is made thereafter, data relating to the identifier of the changed service part is sequentially added to the transition data 220. For this reason, since service part identifiers corresponding to the number of design changes are stored, if n service part identifiers are stored, there are n-1 design changes. The transition data 220 includes, for each service part system, data related to a classification name, a first service part identifier to an nth service part identifier.

In the classification name data area, data related to the name for classifying the system of service parts is stored.
In the first service part identifier data area, data relating to an identifier for identifying the first service part in this transition is stored. For example, a part number or the like is used as the service part identifier.

  The nth service part identifier data area stores data relating to an identifier for identifying the nth generation service part in this transition. The latest service part identifier after the design change can be specified by the service part identifier stored in the nth service part identifier data area.

  As shown in FIG. 3, the order record data storage unit 23 stores order record data 230 regarding each service part. In this order record data 230, data on the order quantity (number of orders) of service parts is stored together with data on the service part identifier and order month. The order record data 230 is set after the service parts are opened (expanded) to each service base, and is additionally stored when an order record confirmed from the order system 10 is received. Therefore, the service parts differ in the number of orders received depending on the month of release, and the number of data corresponding to that is stored. Specifically, the order record data 230 stores, for each service part, data related to the service part identifier, the order date, and the number of orders received.

In the service part identifier data area, data relating to an identifier for specifying a service part with a received order record is stored.
In the order date data area, data related to the date when the service parts were ordered is stored.

In the order quantity data area, data related to the order quantity of the service parts is stored in the corresponding order date. This order quantity is stored in units of months as a time-axis variable.
In the system configured as described above, a processing procedure in the case where a demand forecast for service parts is performed will be described with reference to FIGS. 4 to 6 and 8.

  First, the demand prediction system 20 acquires prediction target specifying data as specific product identifier data for specifying a service part that performs demand prediction (step S1-1). Specifically, the management computer 21 of the demand prediction system 20 acquires the service part identifier of the service part for which the demand prediction is performed from the order receiving system 10 via the network N.

  Next, the demand prediction system 20 acquires the order result data to be predicted (step S1-2). Specifically, the management computer 21 of the demand prediction system 20 extracts the order record data 230 having the service part identifier acquired from the order order system 10 from the order record data storage unit 23.

  Next, the demand prediction system 20 determines whether or not the service part to be predicted is a service part after the design change (step S1-3). Specifically, the management computer 21 compares the service part identifier acquired from the order receiving system 10 with the service part identifier of the transition data 220 stored in the transition data storage unit 22.

  As a result of the comparison, when the transition data 220 including the service part identifier to be predicted is recorded (in the case of “YES” in step S1-3), the management computer 21 predicts from the transition data 220. A service part of a generation before the target service part is specified (step S1-4). That is, the management computer 21 specifies a service part identifier of a service part having the service part to be predicted as a subsequent product. Specifically, the management computer 21 acquires from the transition data 220 all service part identifiers stored in association with the service part to be predicted.

  Next, the management computer 21 acquires the order record data of the specified service parts (step S1-5). Specifically, the management computer 21 extracts all the order record data 230 having the service part identifier specified in step S1-4 from the order record data storage unit 23.

  Next, the management computer 21 determines whether there is a track record of different service parts in the same year and month (step S1-6). Specifically, the management computer 21 makes a determination by comparing the order date and time of the order record data 230 acquired in steps S1-2 and S1-5.

  When the order record data 230 of different service part identifiers includes the order record data 230 of the common order date (when “YES” in step S1-6), the management computer 21 combines Result data is generated (step S1-7). Specifically, the management computer 21 adds up the number of orders received in the order record data 230 of different service part identifiers in the same order date, and calculates a combined result of the order date. If there is no track record of different service parts of the same year and month (in the case of “NO” in step S1-6), the composite track record data is not calculated.

  Next, the management computer 21 performs a demand prediction function calculation process by applying the model using the acquired order record (step S1-8). In the calculation process of the demand prediction function, the management computer 21 uses a technique similar to the technique described in Patent Document 2 described above. For example, after calculating a trend curve, if there is a periodic fluctuation, a periodic fluctuation curve is obtained, and a demand prediction curve is calculated based on the trend curve and the periodic fluctuation curve. That is, when there is periodicity, a function generated by combining the trend function and the periodic function may be selected as the demand prediction model.

  Here, when the service part to be predicted is after the design change (in the case of “YES” in step S1-3), the order result data acquired in steps S1-2 and S1-5 is used. Here, when the composite performance data is calculated (step S1-7), the calculated order quantity for the year and month is used. Further, when the service part to be predicted has not been changed in design (in the case of “NO” in step S1-3), the demand prediction function is calculated using the order record acquired in step S1-2.

  Next, the management computer 21 calculates a predicted value (step S1-9). Specifically, for example, the number of orders received in the next month or the month after next is calculated using the demand prediction function calculated in step S1-8.

  Then, the management computer 21 outputs the calculated predicted value result to the order receiving system 10 via the network N (step S1-10). In the order receiving system 10, the result of the predicted value is output to output means such as a display. By using this demand prediction, the service parts management department can issue an ordering instruction according to the demand.

(Example)
Next, service part demand prediction using specific numerical values will be described with reference to FIGS. Here, as shown in FIG. 5, the identifier of the service part for which the demand prediction is performed is “A3”. It is assumed that the service part has been redesigned twice, and the service part “A2” that has been redesigned from the original service part “A1” is further redesigned. Therefore, the transition data storage unit 22 stores transition data 220 in which “A3” is stored as the third service part identifier. The transition data 220 stores “A1” as the first (first generation) service part identifier and “A2” as the second service part identifier.

  Further, as shown in FIG. 5, the order results of service parts for which demand prediction is made are from March 2005 to June 2005. Correspondingly, the order result data 230 is stored in the order result data storage unit 23. Remembered. Accordingly, the order record data 230 having “A3” as the service part identifier includes data in which the order date is from March to June 2005. Further, it is assumed that the order record data 230 having “A2” as the service part identifier records data with the order date from July 2004 to February 2005. Furthermore, it is assumed that the order record data 230 having “A1” as a service part identifier records data with the order date from July 2002 to June 2004.

  When the demand prediction is performed for the service part with the service part identifier “A3”, the demand prediction system 20 acquires the service part identifier “A3” from the order receiving system 10 as the prediction target specifying data (step S1-1). . The management computer 21 of the demand prediction system 20 extracts the order record data 230 having the service part identifier “A3” from the order record data storage unit 23 (step S1-2). As a result, data related to the number of orders received from March to June 2005 can be extracted.

  Next, the management computer 21 extracts the transition data 220 having the service part identifier “A3” from the transition data storage unit 22 (“YES” in step S1-3). Then, the management computer 21 specifies the service parts (A2, A1) of the previous generation of the service part having the identifier “A3” from the extracted transition data 220 (step S1-4). Then, the management computer 21 acquires the received order data of the extracted service part identifiers “A2, A1” (step S1-5).

  In the present embodiment described here, the extracted order record data 230 assumes that there is no order for the same month for different service parts (in the case of “NO” in step S1-6). Therefore, the management computer 21 calculates a demand prediction function using the order record data 230 acquired in steps S1-2 and S1-5 (step S1-8). A curve based on the demand prediction function calculated here is shown by a solid line in FIG. In FIG. 6, the trend curve used to calculate the demand prediction curve is also indicated by a dotted line.

  Then, from this demand prediction function, the order quantity (predicted value) of the next month (July 2005) and the next month (August 2005) of the service parts is calculated (step S1-9), and the result is output to the order receiving system 10 (Step S1-10).

According to this embodiment, the following effects can be obtained.
In the present embodiment, the management computer 21 is connected to the transition data storage unit 22 and the order record data storage unit 23. The transition data storage unit 22 stores transition data 220 relating to the transition of service parts. The order record data storage unit 23 stores order record data 230 relating to service parts. The management computer 21 of the demand prediction system 20 acquires the order result data 230 to be predicted (step S1-2), and when the service part to be predicted is a service part after the design change, the transition data 220. The service parts of the previous generation are identified, and the order result data 230 of the service parts is acquired (step S1-5). The management computer 21 performs a demand prediction function calculation process using the order record data 230 acquired in steps S1-2 and S1-5 (step S1-8), and calculates a predicted value using the demand prediction function. (Step S1-9), the result of the predicted value is output (Step S1-10). In general, if the number of past actual values is large, more accurate demand prediction can be performed. For this reason, more accurate demand prediction can be performed by calculating also using the service part order record before the design change of the service part for which demand prediction is performed. Therefore, even if the number of past performance values of service parts for which demand prediction is performed is small, demand prediction for service parts can be performed more accurately.

  In this embodiment, the management computer 21 calculates composite performance data when there is a track record of different service parts in the same year (in the case of “YES” in step S1-6) (step S1-7). ). Specifically, the management computer 21 adds up the number of orders received for different service part identifiers in the same order date, and calculates the combined result of the order date. For this reason, even if there is a time when different products are spread simultaneously, more accurate demand prediction can be performed.

  In the present embodiment, when the service part to be predicted is a service part after the design change, the management computer 21 acquires all the service part identifiers stored in association with the service part to be predicted. . Then, the management computer 21 extracts all the order record data 230 having the acquired service part identifiers from the order record data storage unit 23, and uses these order record data 230 to calculate the demand prediction function. Therefore, when the service part for which the demand prediction is performed has undergone a plurality of design changes, the demand prediction can be performed using the order result data 230 for the service parts before all these design changes are made. For this reason, since the number of data of the actual value used for the demand prediction of service parts increases, a more accurate demand prediction can be performed.

  In the present embodiment, the transition data storage unit 22 to which the management computer 21 is connected stores the transition data 220 including data on the first service part identifier to the nth service part identifier for each service part system. Is done. For this reason, the management computer 21 can efficiently identify the transition of all service parts whose design has been changed with respect to this service part by extracting the transition data 220 in which the identifier of the service part to be predicted is stored. .

Moreover, you may change the said embodiment as follows.
In the above embodiment, when the demand prediction function calculation process is performed (step S1-8), the same technique as that described in Patent Document 2 is used, but the method for calculating the demand prediction function is limited to this. It is not something. For example, when a simpler demand forecast is desired, the management computer 21 may perform an order forecast using only a trend curve.

In the above embodiment, the transition data 220 is data in which service part identifiers in which products for which demand prediction is made are associated in order from the first generation. Instead of this, as shown in FIG. 7, transition data 220 in which service part identifiers before and after the design change are associated may be used. In other words, any data may be used as long as it is transition data that can associate service part identifiers before and after the design change.

  Even when the previous generation service part is specified using the transition data 220 shown in FIG. 7, the management computer 21 executes the transition data extraction stage. However, in the transition data extraction stage, a process of extracting a product identifier associated with the specific product identifier and a product identifier of a product associated with the product of the extracted product identifier as a subsequent product are extracted. Performing a previous generation product extraction process. The management computer 21 repeats the previous generation product extraction process until the product identifier data cannot be extracted. Specifically, the management computer 21 extracts the transition data 220 having the service part identifier to be predicted as the service part identifier after the design change. And the transition data 220 which makes the extracted service part identifier the service part identifier after a design change is extracted. The transition data 220 is extracted using the service part identifier extracted in this way as the service part identifier after the design change until the corresponding transition data 220 cannot be extracted. And the service part identifier before the design change of all the extracted transition data 220 is acquired. As a result, all the service parts that are followed by the service part for which the demand prediction is performed can be specified only by associating the service part identifiers before and after the design change.

  ○ In the above-mentioned embodiment, after acquiring the order result data to be predicted, the service part after the design change is specified and the order result data of this service part is obtained. However, the order of obtaining the order result data is It is not limited to. After obtaining the forecast target identification data, identify the service part of the previous generation of the service part to be forecasted, and acquire the order receipt result data of the forecast target and the order result data of the previous generation service part together Good.

  In the above embodiment, the order record data 230 of the service parts to be predicted is acquired from the order record data storage unit 23, and the demand forecast function is calculated using the order record data 230. Alternatively, if there is no order data for predicted service parts, the demand prediction function may be calculated using the order results for service parts of the previous generation of this service part. By using the service part order history before the design change, even if the service part to be predicted is sold for the first time after the design change, even if there is no track record of the service part to be predicted, the service part after the design change Can be predicted more accurately.

  In the above-described embodiment, the demand prediction function is calculated using all orders received for service parts that are the previous generation of service parts to be predicted. Instead of this, the demand prediction function may be calculated using only the order receipt results for a predetermined period (for example, the latest 60 months) from the date of recent demand results. For example, if the demand forecast is not accurate due to consideration of the old order results, the demand forecast function is calculated by excluding the old order results (order results before a predetermined period based on the latest). May be. Thereby, a more accurate predicted value of demand can be calculated.

  In the above embodiment, the demand for service parts is predicted. The target of the demand forecast is not limited to this, and any product that can be associated with a subsequent product of a past product due to a partial change of the product, for example, a demand forecast for software that is upgraded, etc. It may be.

1 is a schematic diagram of a demand prediction system in an embodiment. Explanatory drawing of the data memorize | stored in the transition data memory | storage part. Explanatory drawing of the data memorize | stored in the order record data storage part. The flowchart for demonstrating the process sequence of the demand prediction process in embodiment. The figure which shows the sales performance of the service parts of an Example. Explanatory drawing for demonstrating the demand forecast of the service parts of an Example. Explanatory drawing of the data memorize | stored in the transition data memory | storage part in the example of a change. Explanatory drawing which shows the demand forecast of the service parts after the design change in a prior art.

Explanation of symbols

DESCRIPTION OF SYMBOLS 20 ... Demand prediction system, 21 ... Management computer, 22 ... Transition data storage part as transition data storage means, 23 ... Order received result data storage part as performance data storage means.

Claims (5)

Transition data storage means for storing a product identifier for specifying a product and a product identifier for specifying a subsequent product of the product in association with each other, and data relating to the actual value of product provision corresponding to the time axis is recorded in association with the product identifier. Using a management computer connected to the actual result data storage means, and a method for forecasting demand for the product,
The management computer is
A prediction target data acquisition stage for acquiring specific product identifier data for specifying a product for which demand prediction is performed;
A transition data extraction stage for extracting a product identifier associated with the specific product identifier from the transition data storage means;
A result extraction stage of extracting the actual value of the product of the specific product identifier and the actual value of the product of the extracted product identifier from the actual data storage means using the product identifier;
A function calculation stage for calculating a demand forecast function of the product using the extracted actual value,
A demand prediction method comprising: calculating a predicted value of product provision of a product for which demand prediction is performed using the calculated demand prediction function, and executing a demand prediction output stage for outputting the predicted value. In the transition data storage means, product identifiers of past products until the subsequent product is stored in association with each other,
The demand prediction method according to claim 1, wherein the transition data extraction step extracts all product identifiers associated with the specific product identifier. In the transition data storage means, the product identifier of the subsequent product and the product identifier of the previous generation product of the subsequent product are stored in association with each other,
The transition data extraction step includes:
Performing a process of extracting a product identifier associated with the specific product identifier;
Performing a previous generation product extraction process of extracting a product identifier of a product associated with the product of the extracted product identifier as a subsequent product,
The demand prediction method according to claim 1, wherein the previous generation product extraction process is repeated until data of the product identifier cannot be extracted. In the data related to the actual value associated with the extracted product identifier, the management computer includes a total actual value calculating step for calculating the total value of the actual values for the same period And execute
The demand prediction method according to claim 1, wherein the function calculation step calculates the demand prediction function using the calculated total value. A transition data storage unit that stores a product identifier that specifies a product and a product identifier that specifies a subsequent product of the product in association with each other, and records data related to the actual value of product provision corresponding to the time axis in association with the product identifier. Using a management computer connected to the performance data storage means, a program for forecasting demand for the product,
The management computer;
Prediction target data acquisition means for acquiring specific product identifier data for specifying a product for which demand prediction is performed;
Transition data extraction means for extracting a product identifier associated with the specific product identifier from the transition data storage means,
A result extracting means for extracting a result value of the product of the specific product identifier and a result value of the extracted product of the product identifier from the result data storage means using a product identifier
Using the extracted actual value, the function calculation means for calculating the demand forecast function of the product, and using the calculated demand prediction function, the forecast value of the product provision of the product for which the demand prediction is performed is calculated, and this predicted value is output. A demand forecasting program that functions as a demand forecasting output means.

Images