Demand prediction device and demand prediction method

The demand forecasting device corrects sales records during supply shortages to accurately predict future demand, addressing the underestimation issue post-emergency events, ensuring timely and sufficient product supply and preventing lost sales opportunities.

WO2024241750A1PCT designated stage expired Publication Date: 2024-11-28HITACHI LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
PCT/JP2024/014916
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-23
Filing Date
2024-04-15
Publication Date
2024-11-28

AI Technical Summary

Technical Problem

Existing demand forecasting methods underestimate future demand after emergency events like pandemics, leading to supply shortages and missed sales opportunities due to incorrect predictions based on reduced sales data during such events.

Method used

A demand forecasting device that acquires actual supply and sales data, corrects sales records during supply shortages and out-of-stock periods, and predicts future demand based on the corrected sales performance to ensure accurate forecasting.

Benefits of technology

Enhances demand forecasting accuracy by correcting sales records during supply shortages, preventing underestimation of demand and ensuring sufficient inventory post-emergency events, thereby reducing lost sales opportunities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure JP2024014916_28112024_PF_FP_ABST
    Figure JP2024014916_28112024_PF_FP_ABST
Patent Text Reader

Abstract

The present invention performs demand prediction with higher accuracy. This demand prediction device comprises: an acquisition unit that acquires the actual supply quantity of a product during a past prescribed period and the actual sales quantity of the product during the prescribed period; a determination unit that determines whether the actual supply quantity is less than prescribed and the product is in short supply during the prescribed period; a correction unit that corrects the actual sales quantity if it is determined that the actual supply quantity is less than the prescribed supply quantity and the product is in short supply during the prescribed period; and a prediction unit that predicts a predicted demand quantity of the product for a period following the prescribed period on the basis of the corrected actual sales quantity.
Need to check novelty before this filing date? Find Prior Art

Description

Demand forecasting device and demand forecasting method

[0001] The present invention relates to a demand forecasting device and a demand forecasting method. The present invention claims priority to Japanese Patent Application No. 2023-084409, filed on May 23, 2023, and the contents of that application are incorporated by reference into this application in designated states where incorporation by reference of documents is permitted.

[0002] When a company supplies or sells products to customers, if there is a surplus or shortage relative to customer demand, excess inventory or missed demand (opportunity loss) can result in a decline in profits. To prevent this problem, companies, particularly those in the supply chain, need to produce the products and procure parts for production in advance so that they can supply the right amount of products to customers in a timely manner. To achieve this, it is preferable to predict future demand based on past sales performance, for example, and prepare to supply an amount of products to the supply chain, including customers, that matches that demand. For example, the technology disclosed in Patent Document 1 calculates a projected sales quantity as the future demand quantity based on the product's past production quantity, sales quantity, and inventory quantity.

[0003] Japanese Patent Application Laid-Open No. 2021-174452

[0004] However, when an emergency event such as a pandemic-related lockdown or a natural disaster occurs, parts procurement and production may not proceed as planned, resulting in a shortage of product supply and inventory, and product sales may fall below actual demand. If this reduced sales volume is used as past performance to predict future demand, demand forecasts may be made based on the supply shortages caused by the pandemic, even after the pandemic has subsided, which could result in underestimating demand. If parts procurement and production are carried out based on this underestimated demand, supply preparations to meet actual demand may not be made, and the quantity of products that would have been sold after the pandemic has subsided may not be sold, resulting in lost sales opportunities.

[0005] The present invention has been made in view of the above circumstances, and aims to perform demand forecasting with higher accuracy.

[0006] The present application includes a number of means for solving at least part of the above problems, examples of which are as follows.

[0007] In order to solve the above problem, a demand forecasting device according to one embodiment of the present invention has an acquisition unit that acquires the actual supply quantity of a product during a specified period in the past and the actual sales quantity of the product during the specified period; a determination unit that determines whether the actual supply quantity is less than the specified supply quantity and whether the product is out of stock during the specified period; a correction unit that corrects the actual sales quantity when it is determined that the actual supply quantity is less than the specified supply quantity and the product is out of stock during the specified period; and a prediction unit that predicts the forecast demand quantity of the product for the specified period and beyond based on the corrected actual sales quantity.

[0008] According to the present invention, demand forecasting can be performed with higher accuracy.

[0009] Problems, configurations, and effects other than those described above will become apparent from the following description of the embodiments.

[0010] FIG. 1 is a schematic diagram showing an example of the functional configuration of a demand prediction device according to this embodiment. FIG. 2A is a schematic diagram showing an example of the data structure of sales plan information according to this embodiment. FIG. 2B is a schematic diagram showing an example of the data structure of sales performance information according to this embodiment. FIG. 3A is a schematic diagram showing an example of the data structure of inventory plan information according to this embodiment. FIG. 3B is a schematic diagram showing an example of the data structure of inventory performance information according to this embodiment. FIG. 4A is a schematic diagram showing an example of the data structure of supply plan information according to this embodiment. FIG. 4B is a schematic diagram showing an example of the data structure of supply performance information according to this embodiment. FIG. 5 is a schematic diagram showing an example of the data structure of stock-out information according to this embodiment. FIG. 6 is a schematic diagram showing an example of the data structure of demand estimation information according to this embodiment. FIG. 7A is a schematic diagram showing an example of the data structure of demand forecast information according to this embodiment calculated based on a "sales performance quantity" before correction. FIG. 7B is a schematic diagram showing an example of the data structure of demand forecast information according to this embodiment calculated based on a "sales performance quantity" after correction. FIG. 8 is a schematic diagram showing an example of processing performed by a processing unit according to this embodiment. FIG. 9 is a schematic diagram showing another example of processing performed by the processing unit according to this embodiment. FIG. 10 is a schematic diagram showing another example of a method for correcting the "sales volume" according to this embodiment. FIG. 11 is a schematic diagram showing an example of the average value of the "sales volume" over the past three years according to this embodiment. FIG. 12 is a schematic diagram showing an example of processing performed by the processing unit according to this embodiment when the unit of time is one day. FIG. 13 is a schematic diagram of an example of a screen display of the display unit according to this embodiment. FIG. 14 is a schematic diagram of an example of a screen display of the display unit according to this embodiment when the supply shortage impact button is pressed. FIG. 15 is a flowchart showing an example of processing performed by the demand prediction device according to this embodiment. FIG. 16 is a schematic diagram showing an example of the hardware configuration of the demand prediction device according to this embodiment.

[0011] An embodiment of the present invention will be described below with reference to the drawings. In all drawings used to describe the embodiment, the same components are generally designated by the same reference numerals, and repeated description thereof will be omitted where appropriate. It goes without saying that, in the following embodiments, the components (including element steps, etc.) are not necessarily essential unless otherwise specified or considered to be fundamentally essential. It goes without saying that the terms "consisting of A," "made of A," "having A," and "including A" do not exclude other elements, unless otherwise specified to include only the relevant element. Similarly, in the following embodiments, when referring to the shape, positional relationship, etc. of components, etc., this includes those that are substantially similar or similar to the shape, etc., unless otherwise specified or considered to be fundamentally essential.

[0012] 1 is a schematic diagram showing an example of the functional configuration of a demand prediction device according to this embodiment. The demand prediction device 10 is a computer such as a personal computer (PC) or a server that predicts future demand quantities of products. As an example, the demand prediction device 10 includes a processing unit 11, a storage unit 12, a communication unit 13, a display unit 14, and an input unit 15.

[0013] The communication unit 13 is an interface for connecting the demand prediction device 10 to a network 50 such as a LAN (Local Area Network) or the Internet.

[0014] The display unit 14 is a display device such as a liquid crystal display for displaying various information to the user.

[0015] The input unit 15 is an input device such as a keyboard or a mouse that allows a user to input various pieces of information to the demand prediction device 10. This embodiment is not limited to a mode in which a user directly operates the demand prediction device 10, and various pieces of information may be displayed on a terminal device (not shown) connected to the network 50, and the user may input various pieces of information to the demand prediction device 10 from the terminal device.

[0016] The memory unit 12 stores sales plan information 31, sales performance information 32, inventory plan information 33, inventory performance information 34, supply plan information 35, supply performance information 36, supply shortage information 37, stockout information 38, demand estimation information 39, and demand forecast information 40.

[0017] FIG. 2A is a schematic diagram showing an example of the data structure of the sales plan information 31, and FIG. 2B is a schematic diagram showing an example of the data structure of the sales performance information 32. As shown in FIG.

[0018] The sales plan information 31 is information that associates each of the following: "product name," "seller," "planned sales period," and "planned sales quantity."

[0019] "Product name" is the name of the product, and "seller" is the seller of the product. Note that "product name" is not necessarily limited to the actual name, and may be information that identifies the product, such as a product code. Similarly, "seller" is not necessarily limited to the actual seller name, and may be a seller code or the like. "Planned sales period" is the period during which the product is planned to be sold. In this example, the unit of period is one week, and the first day of the period is stored in "Planned sales period." This also applies to "sales period," "planned inventory period," "inventory period," "planned supply period," "supply period," "out-of-stock period," and "forecast period," which will be described later. "Planned sales quantity" is the sales quantity planned for the "planned sales period."

[0020] The sales performance information 32 is information that associates each of the "product name," "seller," "sales period," and "sales performance quantity."

[0021] The "sales period" is the period during which the product was actually sold, and the "sales volume" is the actual sales volume of the product during that period.

[0022] FIG. 3A is a schematic diagram showing an example of the data structure of the inventory plan information 33, and FIG. 3B is a schematic diagram showing an example of the data structure of the inventory performance information 34.

[0023] The inventory plan information 33 is information that associates each of the following: "product name," "location," "planned inventory period," and "planned inventory quantity."

[0024] "Location" is the inventory location for the product, "Planned inventory period" is the period during which inventory is planned at that inventory location, and "Planned inventory quantity" is the quantity of the product planned to be stocked at the inventory location during that period.

[0025] The inventory performance information 34 is information that associates each of the "product name," "location," "inventory period," and "inventory performance quantity."

[0026] The "inventory period" is the period during which the product was actually in stock at the "base." The "actual inventory quantity" is the actual inventory quantity during that period.

[0027] FIG. 4A is a schematic diagram showing an example of the data structure of the supply plan information 35, and FIG. 4B is a schematic diagram showing an example of the data structure of the supply performance information 36.

[0028] The supply plan information 35 is information that associates "product name," "supplier," "supply destination," "planned supply period," and "planned supply quantity" with each other.

[0029] "Supplier" is the name of the supplier that supplies the product, and "Destination" is the name of the destination to which the supplier delivers the product. "Planned supply period" is the period during which the product is scheduled to be supplied to the "Destination," and "Planned supply quantity" is the quantity of the product that is planned to be supplied during that period.

[0030] The supply record information 36 is information in which "product name," "supplier," "supply destination," "supply period," and "supply record quantity" are associated with each other.

[0031] The "supply period" is the period during which the product was actually supplied to the "supply destination." The "actual supply quantity" is the actual value of the quantity of the product actually supplied during that period.

[0032] 5 is a schematic diagram showing an example of the data structure of the out-of-stock information 38 according to this embodiment. The out-of-stock information 38 is information that associates "product name," "seller," "determination period," and "out-of-stock determination."

[0033] The "determination period" is the period during which it is determined whether or not a product is out of stock. The "out of stock determination" is information indicating whether or not a product was out of stock during the "determination period." If a product was out of stock, it will be "Yes," and if a product was not out of stock, it will be "No."

[0034] In this embodiment, the presence or absence of a stockout is determined based on the "actual inventory quantity" in the actual inventory information 34. For example, it is determined that a stockout occurred during a period in which the "actual inventory quantity" is zero.

[0035] FIG. 6 is a schematic diagram showing an example of the data structure of the demand estimation information 39. As shown in FIG.

[0036] The demand estimation information 39 is information that associates each of the "product name," "seller," "sales period," and "estimated demand quantity."

[0037] The "estimated demand quantity" is an estimate of the demand quantity of a product during the "sales period." The demand quantity is a quantity that indicates the demand for a product, and in this embodiment, the "sales record quantity" in the sales record information 32 is used as the "estimated demand quantity," which is the estimated value. However, during a period when there is a shortage of inventory or stockout, products cannot be sold even if they would like to be sold, and the sales record quantity is therefore smaller than when there is no shortage of inventory or stockout. Therefore, in this embodiment, the "sales record quantity" for a period when there is a shortage of inventory or stockout is corrected to obtain the "estimated demand quantity" for that period. The correction method will be described later.

[0038] 7A and 7B are schematic diagrams showing an example of the data structure of the demand forecast information 40. The demand forecast information 40 is information that associates each of the following: "product name," "location," "forecast period," and "demand forecast quantity."

[0039] The "demand forecast quantity" is a forecast value of the sales quantity of a product in a future "forecast period." In this embodiment, the demand forecast quantity is calculated based on the "sales record quantity" in the sales record information 32 (see FIG. 2B). However, if the "sales record quantity" for a period in which there is an inventory shortage or stockout is corrected, the demand forecast quantity is calculated based on the corrected "sales record quantity." FIG. 7A is a schematic diagram showing an example of the data structure of demand forecast information 40 calculated based on the "sales record quantity" before correction, and FIG. 7B is a schematic diagram showing an example of the data structure of demand forecast information 40 calculated based on the "sales record quantity" after correction. A method for calculating the "demand forecast quantity" in the demand forecast information 40 will be described later.

[0040] 1 again, the above-mentioned information 31 to 36 is stored in, for example, a storage (not shown) connected to the network 50, and the processing unit 11 acquires it from the storage (not shown) as needed and stores it in the memory unit 12.

[0041] The processing unit 11 includes a sales information acquisition unit 16, an inventory information acquisition unit 17, a supply information acquisition unit 18, a supply shortage calculation unit 19, a stockout detection unit 20, a judgment unit 21, a correction unit 22, a prediction unit 23, and a display control unit 24.

[0042] The sales information acquisition unit 16 acquires the sales plan information 31 and the sales performance information 32 via the network 50 and stores them in the storage unit 12 .

[0043] The inventory information acquisition unit 17 acquires the inventory plan information 33 and the inventory performance information 34 via the network 50 and stores them in the storage unit 12 .

[0044] The supply information acquisition unit 18 acquires the supply plan information 35 and the supply performance information 36 via the network 50 and stores them in the storage unit 12 .

[0045] The supply shortage calculation unit 19 calculates the supply shortage of a product for a certain period of time. The calculation method will be described with reference to FIG.

[0046] 8 is a schematic diagram showing an example of processing performed by the processing unit 11. In this example, a one-week period is used in accordance with the fact that the time periods of the information 31 to 36 are each one week. Furthermore, the periods in the inventory plan information 33 and the inventory performance information 34 may be interpreted as representing the end of the period resulting from calculation of receipts and payments of supply and demand for each period, and may not represent the inventory plan and performance values ​​for the entire period.

[0047] As shown in Table 51 in Fig. 8 , the supply shortage calculation unit 19 compares the "planned supply quantity" in the supply plan information 35 with the "actual supply quantity" in the actual supply information 36. The supply shortage calculation unit 19 then uses the "planned supply quantity" as a predetermined supply quantity that serves as a criterion for determining whether or not there is a supply shortage, and calculates the "supply difference amount" obtained by subtracting the "actual supply amount" from the "planned supply amount" for the same period as the supply shortage amount for that period. In periods when the supply shortage amount is negative, a shortage of supply of the product has occurred, and in periods when the supply shortage amount is 0 or greater, no shortage of supply of the product has occurred.

[0048] In the example of Figure 8, the supply difference amount is negative for the week starting on 10 / 17, 10 / 24, and 10 / 31, resulting in a product supply shortage. On the other hand, the supply difference amount is equal to or greater than 0 for the other periods, resulting in no product supply shortage.

[0049] Referring again to FIG. 1 , the stockout detection unit 20 detects whether a product is out of stock during each period. For example, as shown in Table 52 of FIG. 8 , the stockout detection unit 20 determines whether a product is out of stock based on the "actual inventory quantity" in the actual inventory information 34. For example, the stockout detection unit 20 detects that a stockout occurred during a period in which the "actual inventory quantity" is 0, and determines that a stockout did not occur during a period in which the "actual inventory quantity" is greater than 0.

[0050] In the example of FIG. 8, stockouts occur during the week beginning on 10 / 24, 10 / 31, 11 / 7, and 11 / 14, but do not occur during other periods.

[0051] Furthermore, the out-of-stock detection unit 20 generates out-of-stock information 38 (see FIG. 5) indicating the result of the out-of-stock detection, and stores the out-of-stock information 38 in the storage unit 12 .

[0052] Referring again to Fig. 1, the determination unit 21 determines whether the "actual supply quantity" is less than the predetermined supply quantity during a certain period and whether a stockout has occurred during that period.

[0053] 8, the "supply difference amount" is negative and there is a shortage in the week beginning on October 24 and October 31. Therefore, the determination unit 21 determines that the "actual supply quantity" is less than the predetermined supply quantity and there is a shortage in the week beginning on October 24 and October 31.

[0054] Referring again to Figure 1, when the determination unit 21 determines that the "actual supply quantity" for a certain period is less than the predetermined supply quantity and that a stockout has occurred during that period, the correction unit 22 corrects the "actual sales quantity" for that period. This correction method will be described again with reference to Figure 8.

[0055] As shown in Table 53 in Figure 8, the correction unit 22 corrects the "actual sales quantity" for the one week period starting on 10 / 24 and 10 / 31, and sets the corrected value as the "estimated demand quantity." As an example, the correction unit 22 corrects the "actual sales quantity" by increasing the "actual sales quantity" for these periods. Since the sales performance is pushed down during periods when there is a shortage of supply and stockout of a product, by increasing the "actual sales quantity" in this way, it is possible to obtain the original performance value without a shortage of supply or stockout.

[0056] In this case, the correction unit 22 corrects the "actual sales quantity" based on at least one of the "demand forecast quantity" predicted before the period to be corrected and the "sales plan quantity" set before the period to be corrected. In the example of FIG. 8, the "sales plan quantity" for the period to be corrected is used as the corrected "estimated demand quantity." For example, for the week of October 24th, the "sales plan quantity" of "900" becomes the corrected "estimated demand quantity." Note that the "demand forecast quantity" for the period to be corrected may also be used as the corrected "estimated demand quantity." The "demand forecast quantity" and the "sales plan quantity" are quantities set in advance assuming that there will be no product shortages or out-of-stocks. Therefore, by performing correction based on at least one of these quantities, a value close to the "actual sales quantity" in the absence of supply shortages or out-of-stocks can be obtained.

[0057] In the example of Figure 8, the "supply plan quantity" is used as the predetermined supply quantity that serves as the criterion for determining whether there is a shortage of goods, and stockouts are detected depending on whether the "actual inventory quantity" is 0 or not, but this embodiment is not limited to this, and processing such as that shown in Figure 9 may also be performed.

[0058] Fig. 9 is a schematic diagram showing another example of processing performed by the processing unit 11. As shown in Table 51 in Fig. 9, in this example, the "supply plan quantity" minus the threshold value of 400 is set to a "surplus supply quantity" greater than 0. The "surplus supply quantity" is the lower limit of the quantity of goods that can be supplied with a margin. In this example, the same threshold value of "400" is used for all periods, but the threshold value may be changed for each period.

[0059] The determination unit 21 then uses the "surplus supply quantity" as the predetermined supply quantity, and calculates the "shortage of supply" which is the difference between the "actual supply quantity" and the "surplus supply quantity". If the "shortage of supply" is negative, the determination unit 21 determines that the "actual supply quantity" is insufficient compared to the predetermined supply quantity, and if the "shortage of supply" is equal to or greater than 0, the determination unit 21 determines that the "actual supply quantity" is not insufficient compared to the predetermined supply quantity. By determining whether there is a supply shortage based on the "surplus supply quantity", it is possible to determine that a supply shortage has occurred when there is no longer any surplus in the supply quantity.

[0060] Furthermore, as shown in Table 52, the out-of-stock detection unit 20 detects that a stockout has occurred during a certain period if the "actual inventory quantity" during that period is lower than a surplus inventory quantity greater than 0. The surplus inventory quantity is a quantity greater than 0 that allows a product to be stocked with a margin. By determining whether or not there is a stockout based on the surplus inventory quantity, it is possible to determine that a stockout has occurred when there is no more surplus inventory.

[0061] Then, as shown in Table 53, the determination unit 21 determines for each period whether the "actual supply quantity" is less than the predetermined supply quantity and whether a stockout has occurred. Depending on the determination result, the correction unit 22 corrects the "actual sales quantity." Note that, in this example, processing is performed for both Table 51 and Table 52 in FIG. 9, but the processing of FIG. 8 described above may be performed for either Table 51 or Table 52.

[0062] In the example of FIG. 9, the "sales plan quantity" is set as the corrected "sales performance quantity" as in FIG. 8, but the correction method is not limited to this.

[0063] Fig. 10 is a schematic diagram showing another example of a method for correcting the "actual sales quantity." In the example of Fig. 10, the correction unit 22 determines the corrected "actual sales quantity" as α × "forecasted demand quantity" + (1 - α) × "planned sales quantity," and determines the corrected "actual sales quantity" as the "estimated demand quantity." Here, α is a correction coefficient between 0 and 1.

[0064] "Estimated demand quantity (1)", "Estimated demand quantity (2)", and "Estimated demand quantity (3)" in Figure 10 are the "estimated demand quantities" when α = 0, α = 1, and α = 0.6, respectively. Note that if the "actual sales quantity" after correction is smaller than the "actual sales quantity" before correction, the correction unit 22 does not correct the "actual sales quantity", but sets the "actual sales quantity" before correction to the "estimated demand quantity". The white cells in Figure 10 are cells in which the "actual sales quantity" before correction is set to the "estimated demand quantity" in this way.

[0065] Furthermore, in the examples of Figures 8 to 10, the "sales volume" is corrected to obtain the "estimated demand volume", but the average value of past "sales volume" may also be corrected to obtain the "estimated demand volume".

[0066] 11 is a schematic diagram showing an example of the average value of the "sales volume" for the past three years. In this example, the "sales volume (average)" is the average value of the "sales volume" for the past three years during the same period.

[0067] In this way, the correction unit 22 may use the average value of the "sales volume" over the past N years (N is a natural number) in place of the "sales volume" and correct this average value as described above. Furthermore, if the demand cycle of the product is one year, the average value over the past N years may be used, but if the demand cycle is one month, the average value of the "sales volume" over the past N months may be used. Similarly, if the demand cycle is one week, the average value over the past N weeks may be calculated.

[0068] Furthermore, although the examples of FIGS. 8 to 11 use a unit of one week as the period, the unit of one day may also be used.

[0069] 12 is a schematic diagram showing an example of processing performed by the processing unit 11 when the unit of time is one day. When the unit of time in each piece of information 31 to 36 is one day, the unit of time in each of tables 51 to 53 in FIG. 12 is also one day, and the correction unit 22 can correct the "sales volume" in the same way as when the unit of time is one week.

[0070] On the other hand, if the unit period (unit period) of each piece of information 31 to 36 is one week, the supply shortage calculation unit 19 prorates each of the "supply plan quantity" and "actual supply quantity" for one week to the quantity for each day (predetermined period) included in that one week. Then, the supply shortage calculation unit 19 calculates the supply difference amount for each day based on the prorated "supply plan quantity" and "actual supply quantity."

[0071] Furthermore, the stockout detection unit 20 proportionally divides each of the "sales plan quantity" and "actual sales quantity" for one week into the quantities for each day included in that week, calculates receipts and payments for the supply plan quantity and actual supply quantity prorated for each day, and calculates the "inventory plan quantity" and "actual inventory quantity" for each day.The stockout detection unit 20 then detects stockouts for each day based on the calculated "inventory plan quantity" and "actual inventory quantity."

[0072] The determining unit 21 determines for each day whether the "actual supply quantity" is less than the "planned supply quantity" and whether a stockout has occurred.

[0073] The correction unit 22 then allocates each of the "demand forecast quantity," "sales plan quantity," "actual sales quantity," and "estimated demand quantity" for one week proportionally to the quantities included in that one week. The correction unit 22 then corrects the "actual sales quantity" on a daily basis based on these quantities and the determination result of the determination unit 21.

[0074] By dividing the quantity for one week into the quantity for each day in this way, the "sales volume" can be corrected for each day even if the unit of time for each piece of information 31 to 36 is one week.

[0075] The allocation ratio may be determined in advance by the user based on the supply and demand trends for each day of the week. For example, if it is expected that the supply of a product will be greater on weekdays than on weekends, the "actual supply quantity" may be allocated at an allocation ratio that reflects this expectation. For example, the allocation ratio may be set to an expected value of the ratio between the "actual supply quantity" on weekdays and the "actual supply quantity" on weekends.

[0076] Also, if it is expected that there will be more demand for a product on weekends than on weekdays, the allocation ratio can be set to the estimated ratio of the "actual sales quantity" on weekends to the "actual sales quantity" on weekdays. By setting the estimated ratio of the "actual supply quantity" and "actual sales quantity" for each day as the allocation ratio in this way, it is possible to reflect the demand trend for each day in the allocated quantity.

[0077] Referring again to Figure 1, the prediction unit 23 predicts the "demand forecast quantity" of a product for a certain predetermined period onward, based on the "sales volume" for that period, generates demand forecast information 40 indicating the prediction results, and stores it in the storage unit 12. If the "sales volume" has been corrected, the prediction unit 23 predicts the "demand forecast quantity" based on the corrected "sales volume."

[0078] Methods for predicting demand forecast quantities include the moving average method, exponential smoothing method, regression analysis method, and weighted moving average method. Demand forecast quantities may also be predicted using time series analysis methods such as ARIMA (Autoregressive Integrated Moving Average) and multiple regression. Furthermore, demand forecast quantities may also be predicted using machine learning.

[0079] The display control unit 24 displays the "demand forecast quantity" of the product predicted based on the uncorrected "sales actual quantity" and the "demand forecast quantity" based on the corrected "sales actual quantity" so that they can be compared on the display unit 14. The display control unit 24 may also display these "demand forecast quantities" on the display unit of a terminal device (not shown) connected to the network 50.

[0080] Fig. 13 is a schematic diagram of an example of a screen display on the display unit 14. As shown in Fig. 13, a forecast window 60 is displayed on the display unit 14. The forecast window 60 is a window that shows the time transition of the "demand forecast quantity," and includes a demand forecast button 61 and a supply shortage impact button 62.

[0081] When the user presses the demand forecast button 61, the display control unit 24 displays a trend graph 63 showing the "demand forecast quantity" of the product predicted based on the uncorrected "sales actual quantity" in the forecast window 60. If the user wants to check the "demand forecast quantity" based on the corrected "sales actual quantity," the user presses the supply shortage impact button 62.

[0082] 14 is a schematic diagram of an example of a screen display on the display unit 14 when the Supply Shortage Impact button 62 is pressed. As shown in FIG. 14 , when the Supply Shortage Impact button 62 is pressed, the display control unit 24 displays a transition graph 63 based on the uncorrected "actual sales quantity" along with a transition graph 64 showing the "forecasted demand quantity" based on the corrected "actual sales quantity." The display control unit 24 aligns the time axes of the transition graphs 63, 64 to display the transition graphs 63, 64 in a manner that allows them to be compared. This allows the user to understand the impact of product supply shortages and stockouts on the future "forecasted demand quantity."

[0083] FIG. 15 is a flowchart showing an example of processing performed by the demand prediction device 10 according to this embodiment.

[0084] First, the prediction unit 23 determines whether there is a demand forecast calculation request (step S10). For example, the prediction unit 23 determines that there is a calculation request (YES) when the demand forecast button 61 in the forecast window 60 is pressed, and determines that there is no calculation request (NO) when the demand forecast button 61 is not pressed. If the determination here is NO, the processing ends.

[0085] On the other hand, if the determination is YES, the process proceeds to step S11. In step S11, the processing unit 11 acquires the information 31 to 36. For example, the sales information acquisition unit 16 acquires the sales plan information 31 and the sales performance information 32, and the inventory information acquisition unit 17 acquires the inventory plan information 33 and the inventory performance information 34. Furthermore, the supply information acquisition unit 18 acquires the supply plan information 35 and the supply performance information 36.

[0086] The acquired information 31 to 36 each includes a common past period, which will be referred to as the evaluation period hereinafter. In the examples of Figures 2A to 4A, the first day of the evaluation period is October 3, 2022, and the last day of the evaluation period is the last day of the week that begins November 7, 2022.

[0087] Next, the prediction unit 23 predicts the "demand forecast quantity" based on the "sales performance quantity" before correction (step S12). For example, the prediction unit 23 predicts the future "demand forecast quantity" after the evaluation period based on the "sales performance quantity" in the sales performance information 32 for the evaluation period. Then, the prediction unit 23 generates demand forecast information 40 (see FIG. 7A ) indicating the prediction result and stores it in the storage unit 12.

[0088] Next, the display control unit 24 displays the prediction result of the "demand forecast quantity" based on the "sales performance quantity" before correction (step S13). For example, the display control unit 24 displays the prediction result in a transition graph 63 as shown in FIG.

[0089] Next, the correction unit 22 determines whether there is a correction request for the "sales volume" (step S14). For example, the correction unit 22 determines that there is a correction request (YES) when the supply shortage impact button 62 in the forecast window 60 is pressed, and determines that there is no correction request (NO) when the supply shortage impact button 62 is not pressed.

[0090] If the answer here is YES, the process proceeds to step S15, where the supply shortage calculation unit 19 calculates the supply shortage of the product for each period (week) included in the evaluation period.

[0091] Next, the stockout detection unit 20 detects whether the product was out of stock during each period (week) included in the evaluation period (step S16).The stockout detection unit 20 then generates stockout information 38 (see FIG. 5) indicating the stockout detection result and stores it in the storage unit 12.

[0092] Next, the determining unit 21 determines whether the predetermined period (week) in which the supply shortage and stockout occurred is within the evaluation period (step S17). If the determination here is YES, the process proceeds to step S18.

[0093] In step S18, the correction unit 22 corrects the "actual sales quantity" for the specified period (week) in which the supply shortage and stockout occurred by increasing the "actual sales quantity" that was depressed during that specified period (week).

[0094] Next, the prediction unit 23 predicts the future "demand forecast quantity" after the predetermined period (week) in which the supply shortage and stockout occurred based on the corrected "sales actual quantity" (step S19). The prediction unit 23 then generates demand forecast information 40 (see FIG. 7B ) indicating the prediction result and stores it in the storage unit 12.

[0095] Next, the display control unit 24 displays the prediction result of the "demand forecast quantity" based on the corrected "sales performance quantity" (step S20). For example, the display control unit 24 displays the prediction result in a transition graph 64 as shown in FIG.

[0096] This completes the basic processing performed by the demand forecasting device 10 according to this embodiment. Note that the processing also ends in the same way when a NO determination is made in any of steps S10, S14, and S17.

[0097] According to the present embodiment, the sales volume for a period in which a product supply shortage or stockout occurs is corrected, so that the demand for that period is not underestimated. Therefore, demand forecasting can be performed with higher accuracy based on the corrected sales volume.

[0098] <Hardware Configuration> Next, the hardware configuration of the demand prediction device 10 according to this embodiment will be described.

[0099] FIG. 16 is a schematic diagram showing an example of the hardware configuration of the demand prediction device 10 according to this embodiment.

[0100] 16, the demand prediction device 10 includes a storage device 10a, a memory 10b, a processor 10c, an input device 10d, a display device 10e, a communication interface 10f, and a media reader 10g. These components are interconnected by a bus 10i.

[0101] The storage device 10a is a non-volatile storage device such as a hard disk drive (HDD) or a solid state drive (SSD), and stores a demand forecasting program 100 according to this embodiment.

[0102] The demand forecasting program 100 may be recorded on a computer-readable recording medium 10h, and the processor 10c may read the demand forecasting program 100 from the recording medium 10h.

[0103] The recording medium 10h may be a physically portable recording medium such as a CD-ROM (Compact Disc - Read Only Memory), a DVD (Digital Versatile Disc), or a USB (Universal Serial Bus) memory. A semiconductor memory such as a flash memory or a hard disk drive may also be used as the recording medium 10h.

[0104] The demand forecasting program 100 may be stored in a device connected to a public line, the Internet, a LAN (Local Area Network), etc. In this case, the processor 10c may read and execute the demand forecasting program 100.

[0105] The memory 10b is hardware that temporarily stores data, such as a dynamic random access memory (DRAM), and the demand forecasting program 100 is deployed on it.

[0106] The processor 10c is a CPU (Central Processing Unit) or a GPU (Graphical Processing Unit) that controls each part of the demand forecasting device 10. The processor 10c executes the demand forecasting program 100 in cooperation with the memory 10b, thereby realizing the processing unit 11 in FIG.

[0107] The storage unit 12 in FIG. 1 is realized by a storage device 10a and a memory 10b.

[0108] The input device 10d is an input device such as a keyboard or a mouse for realizing the input unit 15 in FIG.

[0109] The display device 10e is a display device such as a liquid crystal display for realizing the display unit 14 in FIG.

[0110] The communication interface 10f is hardware such as a network interface card (NIC) for connecting the demand prediction device 10 to the network 50. The communication interface 10f realizes the communication unit 13 in FIG.

[0111] The medium reading device 10g is hardware such as a USB reader for reading data recorded on the recording medium 10h.

[0112] The effects described in this specification are merely examples and are not limiting, and other effects may also be present.

[0113] The present invention is not limited to the above-described embodiments and includes various modifications. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those including all of the described components. Furthermore, it is possible to replace part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. Furthermore, it is possible to add, delete, or replace part of the configuration of each embodiment with other configurations.

[0114] Furthermore, the above-described configurations, functions, processing units, processing means, etc. may be partially or entirely implemented in hardware, for example, by designing them as integrated circuits. Furthermore, the above-described configurations, functions, etc. may be implemented in software by a processor interpreting and executing a program that implements each function. Information such as the program, decision table, and files that implement each function can be stored in memory, a storage device such as an HDD or SSD, or a recording medium such as an IC (Integrated Circuit) card, an SD (Secure Digital) card, or a DVD (Digital Versatile Disc). Furthermore, the control lines and information lines shown are those considered necessary for explanation, and do not necessarily represent all control lines and information lines in the product. In reality, it can be assumed that almost all components are interconnected.

[0115] 10...demand forecasting device, 10a...storage device, 10b...memory, 10c...processor, 10d...input device, 10e...display device, 10f...communication interface, 10g...medium reading device, 10h...recording medium, 10i...bus, 11...processing unit, 12...storage unit, 13...communication unit, 14...display unit, 15...input unit, 16...sales information acquisition unit, 17...inventory information acquisition unit, 18...supply information acquisition unit, 19...supply shortage calculation unit, 20...out-of-stock detection unit, 21...determination unit, 22... Correction unit, 23...prediction unit, 24...display control unit, 31...sales plan information, 32...sales performance information, 33...inventory plan information, 34...inventory performance information, 35...supply plan information, 36...supply performance information, 37...supply shortage information, 38...stock-out information, 39...demand estimation information, 40...demand forecast information, 50...network, 51-53...table, 60...forecast window, 61...demand forecast button, 62...supply shortage impact button, 63, 64...transition graph, 100...demand forecast program.

Claims

1. A demand forecasting device having: an acquisition unit that acquires an actual supply quantity of a product during a specified period in the past and an actual sales quantity of the product during the specified period; a determination unit that determines whether the actual supply quantity is shorter than the specified supply quantity and the product is out of stock during the specified period; a correction unit that corrects the actual sales quantity when it is determined that the actual supply quantity is shorter than the specified supply quantity and the product is out of stock during the specified period; and a prediction unit that predicts a demand forecast quantity for the product for the specified period and beyond based on the corrected actual sales quantity.

2. A demand forecasting device as described in claim 1, further comprising a display control unit that causes a display unit to display a demand forecast quantity of the product predicted based on the actual sales quantity without the correction and a demand forecast quantity based on the actual sales quantity after the correction in a comparable manner.

3. A demand forecasting device according to claim 1, wherein the correction unit corrects the actual sales quantity by increasing the actual sales quantity.

4. A demand forecasting device as described in claim 3, wherein the correction unit corrects the actual sales quantity for the specified period based on at least one of a demand forecast quantity predicted before the specified period and a sales plan quantity set before the specified period.

5. A demand forecasting device as described in claim 1, wherein the determination unit determines that the product is out of stock during the specified period if the actual inventory quantity of the product during the specified period is below a surplus inventory quantity that is greater than 0.

6. A demand forecasting device according to claim 1, wherein the predetermined supply quantity is a surplus supply quantity greater than zero.

7. A demand forecasting device as described in claim 1, wherein the acquisition unit acquires actual supply information indicating the actual supply quantity for each unit period including a plurality of the specified periods and actual sales information indicating the actual sales quantity for each unit period, the determination unit makes the determination based on the actual supply quantity and the actual sales quantity in the unit period apportioned among the plurality of specified periods, and the correction unit corrects the apportioned actual sales quantity.

8. A demand forecasting device as described in claim 7, wherein the apportioned actual supply quantity is a quantity obtained by apportioning the actual supply quantity in the unit period to a plurality of the specified periods using an estimated ratio of the actual supply quantities in each of the plurality of specified periods as a proration ratio, and the apportioned actual sales quantity is a quantity obtained by apportioning the actual sales quantity in the unit period to a plurality of the specified periods using an estimated ratio of the actual sales quantities in each of the plurality of specified periods as a proration ratio.

9. A demand forecasting method in which a computer executes the steps of: acquiring the actual supply quantity of a product during a specified period in the past and the actual sales quantity of the product during the specified period; determining whether the actual supply quantity is shorter than the specified supply quantity and the product is out of stock during the specified period; correcting the actual sales quantity when it is determined that the actual supply quantity is shorter than the specified supply quantity and the product is out of stock during the specified period; and predicting the forecast demand quantity of the product for the specified period and beyond based on the corrected actual sales quantity.

10. A demand forecasting method as described in claim 9, further comprising the step of causing the computer to display on a display unit a demand forecast quantity of the product predicted based on the actual sales quantity without the correction and the demand forecast quantity based on the actual sales quantity after the correction in a manner that allows them to be compared.

11. The demand forecasting method according to claim 9, wherein the computer corrects the actual sales quantity by increasing the actual sales quantity.

12. A demand forecasting method according to claim 11, wherein the computer corrects the actual sales quantity for the specified period based on at least one of a demand forecast quantity predicted before the specified period and a sales plan quantity set before the specified period.

13. A demand forecasting method as described in claim 9, wherein the computer determines that the product is out of stock during the specified period if the actual inventory quantity of the product during the specified period is below a surplus inventory quantity that is greater than 0.

14. The demand forecasting method according to claim 9, wherein the predetermined supply quantity is a surplus supply quantity greater than zero.

15. A demand forecasting method as set forth in claim 9, wherein the computer acquires actual supply information indicating the actual supply quantity for each unit period including a plurality of the specified periods and actual sales information indicating the actual sales quantity for each of the unit periods, prorates each of the actual supply quantity and the actual sales quantity in the unit period to quantities for a plurality of the specified periods, makes the determination based on said quantities, and corrects the prorated sales quantity.

Citation Information

Patent Citations

  • Method for analyzing and predicting merchandise sales quantity, method for deciding merchandise order quantity and storage medium with program stored therein

    JP2001084239A

  • Demand prediction device, demand prediction method, and program

    JP2021103374A