Output program, output method, and information processing device

Abstract

To present a fair price for an item. [Solution] The information processing device predicts the first price, which is the future price of the first item, based on past purchase data and market data for the first item, which is the item to be predicted. The information processing device also predicts the second price, which is the future price of the second item, based on past purchase data and market data for the second item, which is determined based on the raw material classification to which the first item belongs and a market correlation index. The information processing device calculates the predicted future price of the first item based on the first price and the second price, and outputs the calculated prediction result.

Description

Technical Field

[0001] The present invention relates to an output program and the like.

Background Art

[0002] For example, when procuring raw materials for paints, the person in charge negotiates with suppliers and determines the purchase price for each item of the raw materials. At this time, it is important to predict the price for each item in view of future market trends.

[0003] For items for which sufficient price negotiation has been conducted by focusing on important items, it is possible to predict the price following market changes by a statistical method based on the correlation analysis of the purchase actual price and the market price.

[0004] In addition, as prior art, there is a method of calculating the future acceptable minimum price by including the futures trading price and the forward trading price of the market as raw material price data.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] However, for example, when the number of raw material items is large, for items for which only a small amount is procured or the procurement frequency is low and sufficient price negotiation has not been conducted, the correlation between the purchase actual price and the market price is low, and it is difficult to predict the price with high accuracy by statistical analysis.

[0007] When it is difficult to predict the price, the item is often purchased at a high price, and there is a risk of opportunity loss.

[0008] Therefore, it is necessary to present appropriate prices for items based on market trends.

[0009] In one aspect, the present invention aims to provide an output program, an output method, and an information processing device that can present an appropriate price for an item. [Means for solving the problem]

[0010] In the first plan, the computer performs the following processes: The computer predicts the future price of item 1, which is the item to be predicted, based on historical purchase data and market data for item 1. For item 2, which is determined based on the raw material classification to which item 1 belongs and a market correlation index, the computer predicts the future price of item 2, which is the item 2, based on historical purchase data and market data for item 2. The computer calculates the predicted future price of item 1 based on the first and second prices and outputs the calculated prediction result. [Effects of the Invention]

[0011] It is possible to present a fair price for an item. [Brief explanation of the drawing]

[0012] [Figure 1] Figure 1 is a diagram illustrating the processing content of the information processing device according to this embodiment. [Figure 2] Figure 2 shows an example of the data structure of external market data. [Figure 3] Figure 3 is a diagram illustrating the process of creating representative data. [Figure 4] Figure 4 shows an example of interpolated representative data. [Figure 5] Figure 5 shows an example of a raw material classification market data matrix. [Figure 6] Figure 6 illustrates an example of the process for determining market correlation indicators. [Figure 7]Figure 7 is a functional block diagram showing the configuration of the information processing device according to this embodiment. [Figure 8] Figure 8 is a flowchart showing the processing procedure of the information processing device according to this embodiment. [Figure 9] Figure 9 shows an example of a computer hardware configuration that achieves similar functions to the information processing device in this embodiment. [Modes for carrying out the invention]

[0013] The following describes in detail, with reference to the drawings, embodiments of the output program, output method, and information processing device disclosed in this application. However, this embodiment does not limit the present invention. [Examples]

[0014] The processing content of the information processing device according to this embodiment will now be explained. Figure 1 is a diagram illustrating the processing content of the information processing device according to this embodiment. The information processing device according to this embodiment will be referred to as "information processing device 100". As shown in Figure 1, the information processing device 100 has monthly aggregated purchase history data 141 and external market data 142. The monthly aggregated purchase history data 141 contains various information regarding the purchase history of each raw material. For example, the various information regarding the purchase history includes the source of purchase, supplier, purchase price, etc., which are set on a monthly basis.

[0015] External market data 142 contains exchange rates and market data for each raw material for each month and year. Figure 2 shows an example of the data structure of the external market data. As shown in Figure 2, this external market data 142 registers exchange rates and market data for each raw material for each month and year. The raw materials shown in Figure 2 include naphtha, low-sulfur heavy oil C, castor oil, zinc, steel, crude oil, MMA (methyl methacrylate), titanium dioxide, and epoxy resin, but are not limited to these. Furthermore, in addition to historical values, external market data 142 may also use future forecast values ​​published in the market or forecast values ​​set independently.

[0016] The information processing device 100 executes representative data creation processing based on the monthly aggregated purchase performance data 141 (step S10). The information processing device 100 executes optimal lead time calculation and market linkage index calculation processing based on the external market data 142 (step S11).

[0017] The information processing device 100 executes future price prediction processing for each item with the optimal lead time as an explanatory variable based on the representative data (step S12).

[0018] The information processing device 100 executes price index calculation processing (step S13). The information processing device 100 executes recommended prediction processing (step S14).

[0019] Subsequently, the processing shown in steps S10 to S14 of FIG. 1 will be described more specifically.

[0020] The representative data creation processing described in step S10 of FIG. 1 will be described. FIG. 3 is a diagram for explaining the representative data creation processing. In the example shown in FIG. 3, the case where the information processing device 100 creates representative data for the raw material (CI-NH25) will be described, but the raw material is not limited to this.

[0021] The information processing device 100 acquires information 20 indicating the purchase price for each month for the combination of purchase Gr (Group) and purchase destination Code related to the raw material (CI-NH25) from the monthly aggregated purchase performance data 141. The purchase Gr is a group corresponding to the base for procuring the raw material. The purchase Gr is information for identifying the supplier. The supplier code is information for identifying the supplier of the raw material.

[0022] For example, Information 20 shows that on "February 1, 2018," one transaction took place between purchasing group "Gr2" and supplier code "100475," and the price of the raw material "CI-NH25" purchased by purchasing group "Gr2" from supplier code "100475" was "93.5 (yen)." In each cell of Information 20, a blank cell indicates that no transaction took place between the purchasing group and supplier code on the corresponding date (number of transactions is 0).

[0023] The information processing device 100 determines one representative pair of purchasing group and supplier code based on the information 20. For example, from the pairs of purchasing group and supplier code in information 20, the information processing device 100 selects the pair of purchasing group "Gr5" and supplier code "100502" that has the most recent transaction count. In the example shown in Figure 3, the information processing device 100 identifies the date information for each year, month, and day for the pair of purchasing group "Gr5" and supplier code "100502" that has the most recent transaction count as representative data 21.

[0024] The information processing device 100 interpolates missing values ​​if they exist in the identified representative data. For example, in the representative data 21 shown in Figure 3, there are missing values ​​in the cells from February 1, 2018 to May 1, 2021.

[0025] Here, we will explain an example of the process by which the information processing device 100 interpolates missing values. The information processing device 100 compares the cells of the supplier codes "100475" and "100502" in the purchasing group "Gr5", identifies the region 21a where values ​​are set in each cell, and calculates the difference between the values ​​of each cell. The difference value of region 21a is "114 - 102.5 = 11.5".

[0026] The information processing device 100 interpolates missing values ​​by subtracting the difference value "11.5" from the cell values ​​(prices) of supplier code "100475" in purchasing group "Gr5" from February 1, 2018 to May 1, 2021.

[0027] Figure 4 shows an example of interpolated representative data. For example, the information processing device 100 calculates "93.5" by subtracting the difference value "11.5" from the price "105" for the pair of purchase group "Gr5" and supplier code "100475" on "February 1, 2018". The information processing device 100 sets "93.5" as the price for the pair of purchase group "Gr5" and supplier code "100502" on "February 1, 2018". The information processing device 100 interpolates the missing values ​​in the representative data 21 by performing the same process as above for other missing values.

[0028] In the example above, the information processing device 100 interpolated the missing values ​​using the difference value of one region 21a, but it may also interpolate the missing values ​​using the average value of the difference values ​​of multiple regions.

[0029] The information processing device 100 constructs a time series model using representative data 21 from which missing values ​​have been interpolated. For example, the time series model is one that, given prices for a certain period in the past as explanatory variables, predicts the price for a specified future month.

[0030] For example, time series models include random walks, multiple regression, Poisson regression, ARIMA, ARIMAX, dynamic linear models, exponentially smoothed state spaces, and neural networks. Any well-known technique may be used to construct the time series model.

[0031] The time series model constructed using the representative data 21 in Figure 4 predicts the price for a specified future month for the pair of purchase group "Gr5" and supplier code "100502". The information processing device 100 predicts the price for a specified future month for pairs other than purchase group "Gr5" and supplier code "100502" as follows.

[0032] For example, when the information processing device 100 predicts the price for a specified future month for the pair of purchasing group "Gr4" and supplier code "100502", it calculates the difference value of the most recent price. In the example shown in Figure 4, the difference value "11.5" is calculated by subtracting the price "118.5" for the pair of purchasing group "Gr4" and supplier code "100502" from the price "130" for the pair of purchasing group "Gr5" and supplier code "100502" on "December 1, 2021".

[0033] The information processing device 100 predicts the price for a specified future month for the pair of purchase group "Gr4" and supplier code "100502" by subtracting the difference value "11.5" from the predicted value of the time series model constructed using representative data 21. The information processing device 100 similarly calculates the difference value for other pairs of purchase groups and supplier codes, and predicts the price for a specified future month for other pairs of purchase groups and supplier codes by subtracting (or adding) the difference value from the predicted value of the time series model constructed using representative data 21.

[0034] The information processing device 100 may construct a separate time series model for each raw material, or it may generate a time series model that integrates the time series models of multiple raw materials. In addition to raw materials, the information processing device 100 may also generate a time series model for naphtha prices using a similar method.

[0035] The above explains the process for creating representative data.

[0036] Next, we will explain the process for calculating the optimal leading period and the market correlation index, as described in step S11 of Figure 1. First, we will explain the process for calculating the optimal leading period, and then we will explain the process for calculating the market correlation index.

[0037] The optimal lead period is the period during which the correlation between the lag number of months for each external market data item and the price is highest, for each target item and each external market data item.

[0038] Let's explain the lag period. The lag period is the difference in the periods being compared. For example, when comparing the price fluctuations of a certain item A over a certain period with the price fluctuations of a certain raw material X over a certain period, lag periods of 0, 1, and 2 are as follows. The certain period is assumed to be 12 months. Lag period 0: Compare the price fluctuations of item A over a certain period (May of this year to June of next year) with the price fluctuations of raw material X over a certain period (May of this year to June of next year). Lag period 1: Compare the price fluctuations of item A over a certain period (May of this year to June of next year) with the price fluctuations of raw material X over a certain period (April of this year to May of next year). Lag period 2: Compare the price fluctuations of item A over a certain period (May of this year to June of next year) with the price fluctuations of raw material X over a certain period (March of this year to April of next year).

[0039] The information processing device 100 determines external market data items based on the raw material classification to which the target item belongs and the raw material classification market data matrix. For example, there are approximately 9,000 types of items, and each item is classified into approximately 30 types of raw material classifications. Here, we will refer to the target item as "Item A" and the raw material classification of Item A as "Additives" for this explanation.

[0040] The raw material classification market data matrix is ​​information that defines the relationship between the raw material classification of an item and the raw materials used to create that item. Figure 5 shows an example of the raw material classification market data matrix. Each row 143a of the raw material classification market data matrix 143a is set for each raw material classification, and each column 143b is set for each raw material.

[0041] In each cell of the raw material classification market data matrix 143, a "×" is set for raw materials that do not correspond to a raw material classification. For example, in the example shown in Figure 5, the information processing device 100 identifies the raw materials corresponding to the raw material classification "additives" of item A as naphtha, castor oil, and MMA, and uses each of the identified raw materials as an external market data item. In other words, the external market data items for the raw material classification "additives" of item A are naphtha, castor oil, and MMA.

[0042] The information processing device 100 obtains information on price fluctuations of item A over a certain period and information on price fluctuations of external market data items (for example, naphtha, castor oil, MMA) over a certain period from external market data 142.

[0043] The information processing device 100 searches for the lag period that maximizes the market correlation index between the price of item A and the price of an external market data item, while changing the lag period. The market correlation index is an indicator that shows whether an item is being traded at an optimal price linked to the external market price. The market correlation index can take values ​​from 0 to 1, with values ​​closer to 1 indicating higher market correlation. For example, the range of lag periods to search is set to 1 to 8 months.

[0044] As an example, we will describe a case where the information processing device 100 compares the price fluctuations of item A over a certain period with the price fluctuations of external market data items (e.g., naphtha, castor oil, MMA) over a certain period to find the lag period and identify a market correlation index. The information processing device 100 obtains information on each price fluctuation from external market data 142.

[0045] Figure 6 illustrates an example of the process for determining market correlation indicators. In Figure 6, the item is designated as Item A, and the raw material is castor oil. In Figure 6, "Item Price ↑" indicates that the price of the item (for example, Item A) has risen compared to the previous month. "Item Price ↓" indicates that the price of the item has fallen compared to the previous month. "Raw Material Price ↑" indicates that the price of the raw material (for example, castor oil) has risen compared to the previous month. "Raw Material Price ↓" indicates that the price of the raw material has fallen compared to the previous month.

[0046] In Figure 6, for a lag of "0" months, there are "3" months in the 12-month period in which "item price ↑" and "raw material price ↑", or "item price ↓" and "raw material price ↓". In this case, the information processing device 100 calculates the market correlation index between item A and castor oil for a lag of "0" as "3 / 12 = 0.25".

[0047] A lag month of "1" means that in a 12-month period, there is one month in which both the "item price ↑" and the "raw material price ↑", or both the "item price ↓" and the "raw material price ↓". In this case, the information processing device 100 calculates the market correlation index between item A and castor oil for a lag month of "1" as "1 / 12 = 0.08".

[0048] The information processing device 100 similarly calculates the market correlation index between item A and castor oil for lag months 2 to 8. The information processing device 100 identifies the lag month in which the market correlation index is maximized among lag months 1 to 8 as the optimal leading period for castor oil relative to item A.

[0049] Similarly, the information processing device 100 identifies the lag number of months at which the market correlation index for naphtha relative to item A is at its maximum value as the optimal leading period for naphtha relative to item A. The information processing device 100 identifies the lag number of months at which the market correlation index for MMA relative to item A is at its maximum value as the optimal leading period for MMA relative to item A.

[0050] The information processing device 100 identifies the largest value among the maximum values ​​of the market correlation index for naphtha for item A, the maximum value of the market correlation index for castor oil for item A, and the maximum value of the market correlation index for MMA for item A as the market correlation index for item A.

[0051] The above explains the process for calculating the optimal leading period and the market correlation index.

[0052] Next, we will explain the future price forecasting process described in step S12 of Figure 1. Here, we will explain the case of forecasting the future price of item A. The external market data items for item A will be "naphtha, castor oil, MMA". The optimal lead period for the external market data item "naphtha" obtained in step S11 of Figure 1 will be "3 months". The optimal lead period for the external market data item "castor oil" will be "4 months". The optimal lead period for the external market data item "MMA" will be "2 months". The information processing device 100 obtains the price for each external market data item with the optimal lead period from the monthly aggregated purchase history data 141.

[0053] The information processing device 100 calculates the predicted value (future price) of item A by inputting the price of each external market data item for the optimal leading period into the time series data generated in step S10. For example, the information processing device 100 inputs the price of naphtha three months prior to the base month, the price of castor oil four months prior to the base month, and the price of MMA two months prior to the base month into the time series data and calculates the predicted value of item A. The information processing device 100 calculates the predicted values ​​on a monthly basis for one to six months ahead from the base month. For example, the base month is the month to be predicted.

[0054] The information processing device 100 calculates predicted values ​​for other items besides item A in the same manner.

[0055] This concludes our explanation of the future price prediction process.

[0056] Next, we will explain the price index calculation process described in step S13 of Figure 1. The information processing device 100 calculates the average price (weighted average price) for each raw material classification as a price index. Below, we will explain the case of calculating the price index for the raw material classification "additives".

[0057] The information processing device 100 extracts items from among multiple items included in the raw material classification whose market correlation index is above a threshold. The threshold is set in advance. For example, let's consider the case where the market correlation index of items A, B, and C, among multiple items included in the raw material classification "additives," is above the threshold. Let the market correlation index of item A be a, the market correlation index of item B be b, and the market correlation index of item C be c. The predicted values ​​(future prices) of items A, B, and C have already been calculated by the process described in step S12.

[0058] The weights used by the information processing device 100 when calculating the weighted average of the extracted items A, B, and C are as follows: Weight of item A: wa = a / (a ​​+ b + c) Weight of item B: wb = b / (a ​​+ b + c) Weight of item C: wc = c / (a ​​+ b + c)

[0059] The information processing device 100 calculates the price index for the raw material classification "additives" as follows: Price index for raw material classification "additives" = wa × (predicted value for item A) + wb × (predicted value for item B) + wc × (predicted value for item C)

[0060] The information processing device 100 calculates the price index for each month of the raw material classification "additives" using the respective predicted values ​​if there are multiple predicted values ​​for each month. Similarly, the information processing device 100 also calculates the price index for each month of the raw material classification other than additives.

[0061] The above explains the price index calculation process.

[0062] Next, we will explain the recommendation prediction process described in step S14 of Figure 1. For items whose market correlation index is below the threshold, the predicted values ​​are expected to have low accuracy. For items whose market correlation index is below the threshold, the information processing device 100 uses the price index of the raw material classification to which the item belongs as a basis and presents the "price at which it could have been purchased in light of the external market environment (recommended price)."

[0063] For example, suppose the market correlation index for item D, which is included in the raw material classification "additives," is below a threshold. The information processing device 100 calculates the recommended price for item D one month from now as follows: The information processing device 100 compares the current month's price index for the raw material classification "additives" with the price index one month from now to calculate the rate of change in the price index. The information processing device 100 calculates the recommended price by multiplying the current month's price of item D by the rate of change in the price index.

[0064] For example, if the current price of item D for the current month is "100 yen" and the rate of change is "10%", then the recommended price of item D one month later will be "110 yen".

[0065] This concludes the explanation of the recommendation and prediction process.

[0066] As described above, the information processing device 100 predicts the future price of an item based on monthly aggregated purchasing data 141 and external market data 142. For items where the market correlation index is above a threshold, the information processing device 100 presents the predicted future price, as these items are those for which price negotiations have been sufficiently conducted. On the other hand, for items where the market correlation index is below a threshold, the information processing device 100 presents a recommended price based on the price index of the raw material classification to which the item belongs. This makes it possible to present an appropriate price for the item.

[0067] Furthermore, the information processing device 100 described above can predict highly accurate raw material prices by considering past purchase prices and market trends, enabling optimal purchase timing and price negotiations. For items that are being purchased at inflated prices due to insufficient negotiation, the device can recommend the price at which they could actually be purchased (recommended price), thereby reducing lost opportunities. By using data (monthly aggregated purchase history data 141, external market data 142) and evidence calculated by the information processing device 100 in price negotiations, a certain degree of standardization can be achieved, rather than relying on subjective judgments. The costs required for price negotiations can be reduced.

[0068] Next, an example of the configuration of the information processing device 100 in this embodiment will be described. Figure 7 is a functional block diagram showing the configuration of the information processing device according to this embodiment. As shown in Figure 7, the information processing device 100 has a communication unit 110, an input unit 120, a display unit 130, a storage unit 140, and a control unit 150.

[0069] The communication unit 110 performs data communication with external devices via the network. The communication unit 110 is a NIC (Network Interface Card), etc. For example, the communication unit 110 may acquire monthly aggregated purchase history data 141, external market data 142, etc. from external devices.

[0070] The input unit 120 is an input device that inputs various types of information to the control unit 150 of the information processing device 100. For example, the input unit 120 can be a keyboard, mouse, touch panel, etc. The user may operate the input unit 120 to specify the items to be included in the future price prediction.

[0071] The display unit 130 is a display device that displays information output from the control unit 150.

[0072] The storage unit 140 contains monthly aggregated purchase history data 141, external market data 142, and a raw material classification market data matrix 143. The storage unit 140 is a memory, etc.

[0073] The monthly aggregated purchasing data 141 contains various information regarding the purchasing history of each raw material. Further explanations regarding the monthly aggregated purchasing data 141 are the same as those provided in Figure 1.

[0074] External market data 142 contains exchange rates and market data for each raw material for each month and year. Further explanation of external market data 142 is the same as the explanation given in Figure 2.

[0075] The raw material classification market data matrix 143 defines the relationship between the raw material classification of an item and the raw materials used to create that item. Further explanation of the raw material classification market data matrix 143 is the same as the explanation given in Figure 5.

[0076] Next, we will move on to the explanation of the control unit 150. The control unit 150 includes a creation unit 151, a correlation index calculation unit 152, a price forecasting unit 153, a price index calculation unit 154, and a recommendation forecasting unit 155. The control unit 150 consists of a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), etc.

[0077] The creation unit 151 extracts representative data for each raw material from the monthly aggregated purchase history data 141, and constructs a time series model to predict the future monthly prices of the raw materials based on this representative data. Further explanation of the creation unit 151 is the same as the explanation of the representative data creation process described in step S10 of Figure 1.

[0078] The correlation index calculation unit 152 searches for the lag period (optimal leading period) for the external market data item corresponding to the item. The correlation index calculation unit 152 identifies the maximum value of the market correlation index for the lag period of the external market data item as the market correlation index for the item. Further explanation of the correlation index calculation unit 152 is the same as the explanation of the optimal leading period calculation and market correlation index calculation process described in step S11 of Figure 1.

[0079] The price forecasting unit 153 uses the time series model constructed by the creation unit 151 to predict the future price (predicted value) of the target item. The price forecasting unit 153 predicts the future price of items among multiple items belonging to the same raw material classification as the target item, where the market correlation index is above a threshold. The process by which the price forecasting unit 153 calculates the predicted value of an item is the same as the explanation of the future price forecasting process described in step S12 of Figure 1.

[0080] The price index calculation unit 154 calculates a price index for the raw material classification based on the market correlation index of the items included in the raw material classification. Further explanation of the price index calculation unit 154 is the same as the explanation of the price index calculation process described in step S13 of Figure 1.

[0081] The recommendation forecasting unit 155 performs the following processing for each item. If the market correlation index of the target item is above a threshold, the recommendation forecasting unit 155 uses the future price of the target item predicted by the price forecasting unit 153 as the recommended price for the item, and outputs this recommended price to the display unit 130 for display.

[0082] On the other hand, if the market correlation index of the target item is below a threshold, the recommendation forecasting unit 155 calculates a recommended price for the target item, sets the recommended price as the recommended price for the item, and outputs this recommended price to the display unit 130 for display. Further explanation of the recommendation forecasting unit 155 is the same as the explanation of the recommendation forecasting process described in step S14 of Figure 1.

[0083] Next, an example of the processing procedure of the information processing device 100 according to this embodiment will be described. Figure 8 is a flowchart of the processing procedure of the information processing device according to this embodiment. As shown in Figure 8, the creation unit 151 of the information processing device 100 extracts representative data from the monthly aggregated purchase history data 141 and interpolates the representative data (step S101). The creation unit 151 constructs a time series model based on the representative data (step S102).

[0084] The correlation index calculation unit 152 of the information processing device 100 identifies external market data items for the raw material classification of an item based on the raw material classification of the item and the raw material classification market data matrix 143 (step S103). The correlation index calculation unit 152 searches for the lag number of months (optimal leading period) of the external market data items (step S104).

[0085] The correlation index calculation unit 152 identifies the maximum value of the market correlation index for the lag month of the external market data item as the market correlation index for the item (step S105). The price forecasting unit 153 of the information processing device 100 obtains the price for the optimal leading period of the external market data item for the raw material classification of the item from the monthly aggregated purchase history data 141 (step S106).

[0086] The price forecasting unit 153 predicts the future price of an item by inputting the price of an external market data item for the optimal leading period into a time series model (step S107). The price index calculation unit 154 of the information processing device 100 calculates the average price (weighted average price) for each raw material classification as a price index (step S108).

[0087] If the market correlation index of the target item is above a threshold (step S109, Yes), the recommendation prediction unit 155 of the information processing device 100 proceeds to step S110. The recommendation prediction unit 155 outputs the future price of the target item predicted by the price prediction unit 153 as the recommended price for the item (step S110).

[0088] On the other hand, if the market correlation index of the target item is not above a threshold (step S109, No), the recommendation forecasting unit 155 proceeds to step S111. The recommendation forecasting unit 155 calculates the recommended price (step S111). The recommendation forecasting unit 155 outputs the recommended price as the recommended price for the target item (step S112).

[0089] Next, the effects of the information processing device 100 according to this embodiment will be explained. The information processing device 100 predicts the future price of an item based on monthly aggregated purchase history data 141 and external market data 142. For items where the market correlation index is above a threshold, the information processing device 100 presents the predicted future price, as this indicates that price negotiations have been sufficiently conducted for that item. On the other hand, for items where the market correlation index is below a threshold, the information processing device 100 presents a recommended price based on the price index of the raw material classification to which the item belongs. This makes it possible to present an appropriate price for the item.

[0090] According to the information processing device 100, highly accurate raw material prices can be predicted considering past purchase prices and market trends, enabling optimal purchase timing and price negotiations. For items that are being purchased at inflated prices due to insufficient negotiation, opportunity losses can be reduced by predicting the price at which they could actually be purchased (recommended price). By using data (monthly aggregated purchase history data 141, external market data 142) and evidence calculated by the information processing device 100 in price negotiations, a certain degree of standardization can be achieved, rather than relying on subjective judgments. The costs required for price negotiations can be reduced.

[0091] The information processing device 100 uses a time-series model constructed based on monthly aggregated purchase history data 141 and external market data 142 to calculate future forecast values ​​for each item, and calculates a market correlation index for each item based on the external market data 142. The information processing device 100 calculates a price index based on the forecast values ​​of other items belonging to the raw material classification of the target item, whose market correlation index is above a threshold, and the forecast value of the target item, and calculates and outputs a recommended price for the target item based on the price index. This makes it possible to present an appropriate price for the item.

[0092] The information processing device 100 identifies multiple raw materials corresponding to the raw material classification of an item, and calculates a market correlation index for the item based on the price fluctuations of the item over a certain period in the past, stored in external market data 142, and the price fluctuations of the multiple raw materials over a certain period in the past. This quantifies the correlation between the price fluctuations of raw materials and the price fluctuations of the item, thereby improving the accuracy of recommended prices.

[0093] The information processing device 100 calculates the average price (weighted average price) for each raw material category as a price index, and calculates a recommended price based on the rate of change of the price index. This makes it possible to identify the price index for each raw material category and calculate a recommended price.

[0094] Next, an example of a computer hardware configuration that achieves the same functions as the information processing device 100 described above will be explained. Figure 9 shows an example of a computer hardware configuration that achieves the same functions as the information processing device in this embodiment.

[0095] As shown in Figure 9, the computer 200 includes a CPU 201 that performs various calculations, an input device 202 that receives data input from the user, and a display 203. The computer 200 also includes a communication device 204 and an interface device 205 that exchange data with external devices via a wired or wireless network. Furthermore, the computer 200 includes a RAM 206 for temporarily storing various information and a hard disk drive 207. Each of these devices 201 to 207 is connected to a bus 208.

[0096] The hard disk drive 207 contains a creation program 207a, a correlation index calculation program 207b, a price prediction program 207c, a price index calculation program 207d, and a recommendation prediction program 207e. The CPU 201 reads each of the programs 207a to 207e and loads them into the RAM 206.

[0097] The creation program 207a functions as the creation process 206a. The correlation indicator calculation program 207b functions as the correlation indicator calculation process 206b. The price prediction program 207c functions as the price prediction process 206c. The price index calculation program 207d functions as the price index calculation process 206d. The recommendation prediction program 207e functions as the recommendation prediction process 206e.

[0098] The processing in creation process 206a corresponds to the processing in creation unit 151. The processing in correlation indicator calculation process 206b corresponds to the processing in correlation indicator calculation unit 152. The processing in price forecasting process 206c corresponds to the processing in price forecasting unit 153. The processing in price index calculation process 206d corresponds to the processing in price index calculation unit 154. The processing in recommendation forecasting process 206e corresponds to the processing in recommendation forecasting unit 155.

[0099] Furthermore, programs 207a to 207e do not necessarily have to be stored on the hard disk drive 207 from the beginning. For example, each program could be stored on a "portable physical medium" such as a flexible disk (FD), CD-ROM, DVD, magneto-optical disk, or IC card inserted into the computer 200. Then, the computer 200 could read and execute each program 207a to 207e.

[0100] With regard to embodiments including each of the above examples, the following additional information is disclosed.

[0101] (Note 1) Based on past purchase data and market data for item 1, which is the item to be predicted, we predict the future price of item 1, which is item 1. Based on the raw material classification to which the first item belongs and the market correlation index, the second item is determined, and based on past purchase data and market data for the second item, the second price, which is the future price of the second item, is predicted. Based on the first price and the second price, the system calculates a forecast of the future price of the first item and outputs the calculated forecast result. An output program characterized by having a computer perform the processing.

[0102] (Note 2) The output program according to Note 1, characterized in that for each of the multiple items including the first item and the second item, a computer further performs a process to identify a plurality of raw materials corresponding to the raw material classification of the item, and calculate a market correlation index for the item based on the price fluctuations of the item over a certain period in the past stored in the market data and the price fluctuations of the plurality of raw materials over a certain period in the past.

[0103] (Note 3) The price index is calculated based on the weights obtained from the first price and the second price, and the market correlation index of the first item and the market correlation index of the second item. The output program according to Appendix 2, characterized in that it causes a computer to further perform a process of calculating a prediction result of the future price of the first item using the price index.

[0104] (Note 4) The output program according to Note 3, characterized in that the computer further performs a process to calculate the future price of the first item based on the rate of change of the price index.

[0105] (Note 5) Based on past purchase data of item 1, which is the item to be predicted, and market data, the first price, which is the future price of item 1, is predicted. Based on the raw material classification to which the first item belongs and the market correlation index, the second item is determined, and based on past purchase data and market data for the second item, the second price, which is the future price of the second item, is predicted. Based on the first price and the second price, the system calculates a forecast of the future price of the first item and outputs the calculated forecast result. An output method characterized by the processing being performed by a computer.

[0106] (Appendix 6) The output method according to Appendix 5, characterized in that for each of the multiple items including the first item and the second item, the computer further performs a process to identify multiple raw materials corresponding to the raw material classification of the item, and calculate a market correlation index for the item based on the price fluctuations of the item over a certain period in the past stored in the market data and the price fluctuations of the multiple raw materials over a certain period in the past.

[0107] (Note 7) The price index is calculated based on the weights obtained from the first price and the second price, and the market correlation index of the first item and the market correlation index of the second item. The output method according to Appendix 6, characterized in that a computer further performs a process of calculating a prediction result of the future price of the first item using the price index.

[0108] (Note 8) The output method according to Note 7, characterized in that the computer further performs a process to calculate the future price of the first item based on the rate of change of the price index.

[0109] (Note 9) Based on past purchase data of item 1, which is the item to be predicted, and market data, the first price, which is the future price of item 1, is predicted. Based on the raw material classification to which the first item belongs and the market correlation index, the second item is determined, and based on past purchase data and market data for the second item, the second price, which is the future price of the second item, is predicted. Based on the first price and the second price, the system calculates a forecast of the future price of the first item and outputs the calculated forecast result. An information processing device having a control unit that performs processing.

[0110] (Note 10) The information processing apparatus according to Note 9, wherein the control unit further performs a process to calculate a market correlation index for each of the multiple items, including the first item and the second item, by identifying a plurality of raw materials corresponding to the raw material classification of the item, and based on the price fluctuations of the item over a certain past period stored in the market data and the price fluctuations of the plurality of raw materials over a certain past period.

[0111] (Note 11) The information processing apparatus according to Note 10, characterized in that the control unit further performs a process of calculating a price index based on the first price and the second price and weights obtained from the market correlation index of the first item and the market correlation index of the second item, and calculating a prediction result of the future price of the first item using the price index.

[0112] (Note 12) The information processing apparatus according to Note 11, characterized in that the control unit further performs a process to calculate the future price of the first item based on the rate of change of the price index. [Explanation of Symbols]

[0113] 100 Information Processing Devices 110 Communications Department 120 Input section 130 Display section 140 Storage section 141 Monthly aggregated data on purchasing performance 142 External Market Data 143 Raw Material Classification Market Data Matrix 150 Control Unit 151 Creation Department 152 Interlocking index calculation section 153 Price Forecasting Section 154 Price Index Calculation Section 155 Recommended Forecast Section

Claims

1. Based on past purchase data and market data for item 1, which is the item to be predicted, we predict the future price of item 1, which is item 1. Based on the raw material classification to which the first item belongs and the market correlation index, the second item is determined, and based on past purchase data and market data for the second item, the second price, which is the future price of the second item, is predicted. Based on the first price and the second price, the system calculates a forecast of the future price of the first item and outputs the calculated forecast result. An output program characterized by having a computer perform the processing.

2. The output program according to claim 1, characterized in that, for each of the multiple items including the first item and the second item, a computer further performs a process to identify a plurality of raw materials corresponding to the raw material classification of the item, and calculate a market correlation index for the item based on the price fluctuations of the item over a certain period in the past stored in the market data and the price fluctuations of the plurality of raw materials over a certain period in the past.

3. A price index is calculated based on the first price and the second price, and the weights obtained from the market correlation index of the first item and the market correlation index of the second item. The output program according to claim 2, characterized in that it causes a computer to further perform a process of calculating a prediction result of the future price of the first item using the price index.

4. The output program according to claim 3, characterized in that it causes a computer to further perform a process of calculating the future price of the first item based on the rate of change of the price index.

5. Based on past purchase data and market data for item 1, which is the item to be predicted, we predict the future price of item 1, which is item 1. Based on the raw material classification to which the first item belongs and the market correlation index, the second item is determined, and based on past purchase data and market data for the second item, the second price, which is the future price of the second item, is predicted. Based on the first price and the second price, the system calculates a forecast of the future price of the first item and outputs the calculated forecast result. An output method characterized by the processing being performed by a computer.

6. Based on past purchase data and market data for item 1, which is the item to be predicted, we predict the future price of item 1, which is item 1. Based on the raw material classification to which the first item belongs and the market correlation index, the second item is determined, and based on past purchase data and market data for the second item, the second price, which is the future price of the second item, is predicted. Based on the first price and the second price, the system calculates a forecast of the future price of the first item and outputs the calculated forecast result. An information processing device having a control unit that performs processing.

Images