Information processing method, information processing device, and information processing program

Abstract

An information processing device according to the present invention calculates a predicted sales amount, a predicted sales upper limit amount, and a predicted sales lower limit amount for a prescribed product for every future unit period of time, calculates a predicted inventory amount for the prescribed product for every future unit period of time on the basis of an actual inventory amount, an actual remaining order amount, a maximum inventory amount, a maximum remaining order amount, the predicted sales amount, the predicted sales upper limit amount, and the predicted sales lower limit amount to satisfy the constraint conditions that a predicted remaining order amount not exceed the maximum remaining order amount when the predicted sales amount is at the predicted sales upper limit amount and the predicted inventory amount not exceed the maximum inventory amount when the predicted sales amount is at the predicted sales lower limit amount, calculates a predicted production amount for the prescribed product for every future unit period of time on the basis of the actual inventory amount, the predicted sales amount, and the predicted inventory amount, and outputs the predicted production amount for every unit period of time.

Description

Information processing method, information processing device, and information processing program

[0001] The present disclosure relates to a technology for calculating a predicted production volume of a predetermined product for each unit period and outputting the predicted production volume for each unit period.

[0002] Traditionally, companies use PSI (Production, Sales, Inventory) tools for ordering. The PSI tool predicts future sales volumes of products, determines appropriate inventory levels, and determines production or order volumes to achieve the predicted sales volumes and determined inventory levels.

[0003] For example, the ordering support device disclosed in Patent Document 1 has a demand forecasting processing unit that performs demand forecasting from PSI (production / sales / inventory) data, an inventory diagnosis processing unit that determines whether an item is out of stock or in excess of stock from the PSI data, and a visualization processing unit that calculates future inventory data based on the PSI data and the demand forecast results from the demand forecasting processing unit, and displays the future inventory status together with future production / sales data on a graph defined by the time axis and inventory amount.

[0004] Furthermore, for example, the inventory replenishment system shown in Patent Document 2 includes a storage device that stores information on the production capacity for producing multiple items, information on the minimum lot size, unit lot size, and maximum lot size of the inventory replenishment amount for each item, information on the standard inventory amount, maximum inventory amount, and cumulative replenishment amount for each item, and information on the current available inventory amount for each item, and a processing device that calculates the inventory replenishment amount for each item based on the various information stored in the storage device.

[0005] Furthermore, for example, the production status visualization system disclosed in Patent Document 3 creates production process model information that makes it possible to know, for each of a plurality of processes, the required element items, the number of required element items, and the destination of the finished item, and for each of a plurality of processes, based on the number of element items, sets the planned production number as the smaller of the producible number or the planned production number, which indicates the maximum number of finished items that can be produced in that process, and specifies the sum of the planned production number and the inventory number as the output number.

[0006] However, the above-mentioned conventional technology does not determine the optimal inventory level using the forecasted sales volume while taking into account constraints at the production site, and does not automatically determine the forecasted production volume using the forecasted inventory level, so further improvement is needed.

[0007] Japanese Patent Publication No. 2021-174452 Japanese Patent No. 4624191 Japanese Patent Publication No. 2022-113032

[0008] The present disclosure has been made to solve the above problems, and aims to provide a technology that can determine an optimal predicted inventory quantity using predicted sales volume while taking into account constraints at the production site, and that can automatically determine a predicted production quantity using the predicted inventory quantity.

[0009] The information processing method according to the present disclosure is an information processing method executed by a computer, which acquires a past actual production volume, actual sales volume, actual inventory volume, and actual backlog order volume for a predetermined product for each unit period, acquires a maximum inventory volume and a maximum backlog order volume for the predetermined product for each unit period, and calculates, based on the past actual sales volume, a predicted sales volume for the predetermined product for each unit period in the future, a predicted upper limit sales volume that is acceptable when the sales volume exceeds the predicted sales volume, and a predicted lower limit sales volume that is acceptable when the sales volume falls below the predicted sales volume, and calculates the past actual inventory volume, the backlog order volume, the maximum inventory volume, the past actual inventory volume, the maximum backlog order volume, the future ... The method includes calculating a predicted inventory quantity for the specified product for each future unit period based on the maximum backlog, the forecast sales quantity, the forecast upper sales limit, and the forecast lower sales limit, which satisfies the constraints that the forecast backlog will not exceed the maximum backlog when the forecast sales quantity reaches the forecast upper sales limit, and that the forecast inventory quantity will not exceed the maximum inventory when the forecast sales quantity reaches the forecast lower sales limit; calculating a predicted production quantity for the specified product for each future unit period based on the actual inventory quantity, the forecast sales quantity, and the forecast inventory quantity; and outputting the predicted production quantity for each unit period.

[0010] According to the present disclosure, it is possible to determine an optimal forecast inventory quantity using forecast sales volume while taking into account constraints at the production site, and it is possible to automatically determine a forecast production quantity using the forecast inventory quantity.

[0011] 1 is a diagram showing a configuration of a production support system according to an embodiment. FIG. 2 is a diagram showing an example of PSI data according to the embodiment. FIG. 3 is a diagram showing an example of setting data according to the embodiment. FIG. 4 is a flowchart for explaining prediction processing by an information processing device according to an embodiment of the present disclosure. FIG. 5 is a schematic diagram for explaining in more detail the calculation of a predicted inventory amount by a predicted inventory amount calculation unit according to the embodiment. FIG. 6 is a diagram showing an example of a predicted sales amount, a predicted upper sales amount, a predicted lower sales amount, a predicted inventory amount, and a predicted production amount for each unit period, which are displayed on a display unit according to the embodiment. FIG. 7 is a diagram showing an example of a user interface screen displayed on a display unit according to the embodiment. FIG. 8 is a diagram showing an example of a user interface screen including a predicted sales amount, a predicted inventory amount, and a predicted production amount for each unit period, which are displayed on a display unit according to the embodiment.

[0012] (Knowledge forming the basis of the present disclosure) The above-mentioned conventional technologies each have their own problems, and do not automate the work processes at the production site while simultaneously considering all factors.

[0013] The above-mentioned Patent Document 1 predicts future production volumes, sales volumes, and inventory volumes for up to several months into the future, diagnoses future inventory from the predicted future production volumes, sales volumes, and inventory volumes, and automatically determines order dates and order volumes from the diagnosis results. However, Patent Document 1 does not take into account constraints on the production site, such as maximum inventory volumes and maximum backlog orders in a unit period.

[0014] Furthermore, in the above-mentioned Patent Document 2, since the inventory replenishment amount is not calculated using the predicted sales amount, it is not clear whether the inventory amount can be further reduced from the current inventory standard amount.

[0015] Furthermore, in the above-mentioned Patent Document 3, when the producible quantity is smaller than the planned production quantity, the producible quantity rectangular area is simply highlighted, and the planned production quantity is not automatically calculated in accordance with the constraints.

[0016] In order to solve the above problems, the following techniques are disclosed.

[0017] (1) An information processing method according to one aspect of the present disclosure is an information processing method executed by a computer, the information processing method including: acquiring a past actual production volume, actual sales volume, actual inventory volume, and actual backlog order volume for a predetermined product for each unit period; acquiring a maximum inventory volume and a maximum backlog order volume for the predetermined product for each unit period; calculating, based on the past actual sales volume, a forecasted sales volume for the predetermined product for each unit period in the future; a forecasted upper limit sales volume that is acceptable when the sales volume exceeds the forecasted sales volume; and a forecasted lower limit sales volume that is acceptable when the sales volume falls below the forecasted sales volume; calculating a predicted inventory quantity for the specified product for each future unit period based on the actual inventory quantity, the maximum backlog, the forecasted sales quantity, the upper forecast sales quantity, and the lower forecast sales quantity, such that the predicted backlog will not exceed the maximum backlog when the forecasted sales quantity reaches the upper forecast sales quantity, and the predicted inventory quantity will not exceed the maximum inventory when the forecasted sales quantity reaches the lower forecast sales quantity; calculating a predicted production quantity for the specified product for each future unit period based on the actual inventory quantity, the forecasted sales quantity, and the forecasted inventory quantity; and outputting the predicted production quantity for each unit period.

[0018] According to this configuration, a forecasted sales volume for a future unit period of a specified product, a forecasted upper sales volume that is acceptable when the sales volume exceeds the forecasted sales volume, and a forecasted lower sales volume that is acceptable when the sales volume falls below the forecasted sales volume are calculated based on the actual sales volume for each unit period of the specified product in the past. Then, a forecasted inventory volume for a future unit period of the specified product is calculated, satisfying the constraints that the forecasted backlog does not exceed the maximum backlog when the forecasted sales volume reaches the forecasted sales upper limit, and the forecasted inventory does not exceed the maximum inventory when the forecasted sales volume reaches the forecasted sales lower limit. Then, a forecasted production volume for a future unit period of the specified product is calculated based on the actual inventory volume, the forecasted sales volume, and the forecasted inventory volume.

[0019] Therefore, the optimum predicted inventory quantity can be determined using the predicted sales quantity while taking into consideration constraints at the production site, and the predicted production quantity can be automatically determined using the predicted inventory quantity.

[0020] (2) In the information processing method described in (1) above, the output may include outputting the forecast sales volume, the forecast upper sales volume, the forecast lower sales volume, the forecast inventory volume, and the forecast production volume for each unit period.

[0021] According to this configuration, the forecast sales volume, forecast upper limit sales volume, forecast lower limit sales volume, forecast inventory volume, and forecast production volume for each unit period are output to the display unit, so that the forecast sales volume, forecast upper limit sales volume, forecast lower limit sales volume, forecast inventory volume, and forecast production volume for each unit period can be presented to the user.

[0022] (3) The information processing method described in (1) or (2) above may further include obtaining a maximum production volume for the specified product in a unit period, and, if the predicted sales volume in a first unit period exceeds the maximum production volume, increasing the predicted inventory volume in a second unit period that precedes the first unit period.

[0023] With this configuration, even if the predicted sales volume in the first unit period exceeds the maximum production volume, the inventory that was increased in advance in the second unit period before the first unit period can be used for sales.

[0024] (4) In the information processing method described in any one of (1) to (3) above, the calculation of the forecast sales volume, the forecast upper limit sales volume, and the forecast lower limit sales volume may include calculating the forecast sales volume, the forecast upper limit sales volume, and the forecast lower limit sales volume by inputting the actual sales volume into a forecast model created by pre-learning and obtaining the forecast sales volume, the forecast upper limit sales volume, and the forecast lower limit sales volume output from the forecast model.

[0025] According to this configuration, the predicted sales volume, the predicted upper limit sales volume, and the predicted lower limit sales volume can be easily calculated using a prediction model created by pre-learning.

[0026] (5) The information processing method described in any one of (1) to (4) above may further include acquiring calendar information regarding past and future days of the week and holidays, weather information regarding past and future weather, and past actual sales volumes of similar products to the specified product, and calculating the forecast sales volume, the forecast upper sales volume, and the forecast lower sales volume may include calculating the forecast sales volume, the forecast upper sales volume, and the forecast lower sales volume based on the actual sales volume of the specified product, the calendar information, the weather information, and the actual sales volumes of the similar products.

[0027] According to this configuration, the forecast sales volume, forecast upper limit sales volume, and forecast lower limit sales volume are calculated using not only the actual sales volume of a specified product, but also calendar information regarding past and future days of the week and holidays, weather information regarding past and future weather, and past actual sales volumes of products similar to the specified product, so the forecast sales volume, forecast upper limit sales volume, and forecast lower limit sales volume can be calculated with greater accuracy.

[0028] (6) In the information processing method described in any one of (1) to (5) above, calculating the predicted inventory amount may include calculating the predicted inventory amount and the predicted backorder amount for a unit period t using the actual inventory amount and the actual backorder amount for a unit period t-1, the predicted sales amount for the unit period t, and a variable production amount for the unit period t; calculating multiple combinations of the predicted inventory amount and the predicted backorder amount for the unit period t by changing the production amount; selecting a combination from the multiple combinations that satisfies the constraint conditions and has the smallest sum of the predicted inventory amount and the predicted backorder amount; and calculating the predicted inventory amount from the selected combination as the predicted inventory amount for the unit period t.

[0029] Once the forecasted sales volume and production volume for unit period t and the actual inventory volume and actual backlog order volume for unit period t-1 are obtained, the forecasted inventory volume can be calculated. Here, the production volume for unit period t is a variable, and by changing the production volume, multiple combinations of the forecasted inventory volume and the forecasted backlog order volume for unit period t are calculated. Then, from the multiple combinations, a combination that satisfies the constraint conditions and has the smallest total of the forecasted inventory volume and the forecasted backlog order volume is selected. Then, the forecasted inventory volume of the selected combination is calculated as the forecasted inventory volume for unit period t.

[0030] Therefore, the predicted inventory amount can be optimized because the constraint conditions are satisfied and the smallest predicted inventory amount is calculated.

[0031] (7) In the information processing method described in any one of (1) to (6) above, calculating the forecast production volume may include calculating the forecast production volume by adding the forecast inventory volume to the forecast sales volume and subtracting the actual inventory volume from the added volume.

[0032] According to this configuration, the predicted production volume can be calculated by adding the predicted inventory amount to the predicted sales volume and subtracting the actual inventory amount from the added amount.

[0033] (8) In the information processing method described in any one of (1) to (7) above, the calculation of the predicted inventory amount may include calculating the predicted inventory amount for each of a plurality of warehouses when the specified product passes through a plurality of warehouses from the time the specified product is produced until the time the specified product is sold and the specified product moves through the plurality of warehouses for each predicted unit period.

[0034] According to this configuration, even if there is a lead time between production and sales, the predicted inventory amount and predicted production amount can be calculated taking the lead time into consideration.

[0035] Furthermore, the present disclosure can be realized not only as an information processing method that executes the characteristic processes described above, but also as an information processing device having a characteristic configuration corresponding to the characteristic processes executed by the information processing method. Furthermore, the present disclosure can also be realized as a computer program that causes a computer to execute the characteristic processes included in such an information processing method. Therefore, the same effects as those of the above information processing method can also be achieved in the following other aspects.

[0036] (9) An information processing device according to another aspect of the present disclosure includes a first acquisition unit that acquires a past actual production volume, an actual sales volume, an actual inventory volume, and an actual backlog order volume of a predetermined product for each unit period; a second acquisition unit that acquires a maximum inventory volume and a maximum backlog order volume of the predetermined product for each unit period; a first calculation unit that calculates, based on the past actual sales volume, a predicted sales volume of the predetermined product for each unit period in the future, a predicted upper limit sales volume that is acceptable when the sales volume exceeds the predicted sales volume, and a predicted lower limit sales volume that is acceptable when the sales volume is lower than the predicted sales volume; and a calculation unit that calculates the past actual inventory volume, the backlog order volume, the maximum inventory volume, the maximum backlog order volume, The system includes a second calculation unit that calculates a predicted inventory quantity for the specified product for each future unit period based on the predicted sales quantity, the predicted upper sales quantity, and the predicted lower sales quantity, satisfying the constraints that the predicted backlog quantity will not exceed the maximum backlog quantity when the predicted sales quantity becomes the predicted upper sales quantity, and that the predicted inventory quantity will not exceed the maximum inventory quantity when the predicted sales quantity becomes the predicted lower sales quantity; a third calculation unit that calculates a predicted production quantity for the specified product for each future unit period based on the actual inventory quantity, the predicted sales quantity, and the predicted inventory quantity; and an output unit that outputs the predicted production quantity for each unit period.

[0037] (10) An information processing program according to another aspect of the present disclosure acquires a past actual production volume, actual sales volume, actual inventory volume, and actual backlog order volume for a predetermined product for each unit period, acquires a maximum inventory volume and a maximum backlog order volume for the predetermined product for each unit period, and calculates, based on the past actual sales volume, a forecast sales volume for the predetermined product for each unit period in the future, a forecast upper limit sales volume that is acceptable when the sales volume exceeds the forecast sales volume, and a forecast lower limit sales volume that is acceptable when the sales volume falls below the forecast sales volume, and calculates the actual inventory volume, the backlog order volume, the maximum inventory volume, the maximum backlog order volume, and the forecast sales volume. The computer is caused to function as follows: based on the sales volume, the predicted upper sales volume, and the predicted lower sales volume, a predicted inventory volume for the specified product per future unit period is calculated, which satisfies the constraints that the predicted backlog of orders will not exceed the maximum backlog of orders when the predicted sales volume reaches the predicted upper sales volume, and that the predicted inventory volume will not exceed the maximum inventory volume when the predicted sales volume reaches the predicted lower sales volume; based on the actual inventory volume, the predicted sales volume, and the predicted inventory volume, a predicted production volume for the specified product per future unit period is calculated, and the predicted production volume for each unit period is output.

[0038] (11) A non-transitory computer-readable recording medium according to another aspect of the present disclosure records an information processing program, the information processing program acquiring a past actual production volume, actual sales volume, actual inventory volume, and actual backlog order volume for a predetermined product for each unit period, acquiring a maximum inventory volume and a maximum backlog order volume for the predetermined product for each unit period, and calculating, based on the past actual sales volume, a predicted sales volume for the predetermined product for each unit period in the future, a predicted upper limit sales volume that is acceptable when the sales volume exceeds the predicted sales volume, and a predicted lower limit sales volume that is acceptable when the sales volume falls below the predicted sales volume, and calculating the past actual inventory volume, the backlog order volume, and the maximum inventory volume and the maximum backlog order volume for the predetermined product for each unit period in the future. The computer is caused to function as follows: based on the maximum inventory quantity, the maximum remaining order quantity, the forecasted sales quantity, the upper forecast sales quantity, and the lower forecast sales quantity, a forecasted inventory quantity for each future unit period of the specified product is calculated, which satisfies the constraints that the forecasted remaining order quantity will not exceed the maximum remaining order quantity when the forecasted sales quantity reaches the upper forecast sales quantity, and that the forecasted inventory quantity will not exceed the maximum inventory quantity when the forecasted sales quantity reaches the lower forecast sales quantity; based on the actual inventory quantity, the forecasted sales quantity, and the forecasted inventory quantity, a forecast production quantity for each future unit period of the specified product is calculated; and outputting the forecast production quantity for each unit period.

[0039] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. Note that each of the embodiments described below represents a specific example of the present disclosure. The numerical values, shapes, components, steps, and order of steps shown in the following embodiments are merely examples and are not intended to limit the present disclosure. Furthermore, among the components in the following embodiments, components that are not described in the independent claims that represent the highest concept are described as optional components. Furthermore, in all embodiments, the respective contents can be combined.

[0040] (Embodiment) FIG. 1 is a diagram showing the configuration of a production support system according to this embodiment.

[0041] The production support system shown in FIG. 1 includes an information processing device 1, an input unit 2, and a display unit 3.

[0042] The input unit 2 is, for example, a keyboard, a mouse, or a touch panel, and accepts information input by a user. The input unit 2 is connected to the information processing device 1 via a wired or wireless connection so that they can communicate with each other. Note that the input unit 2 may also be connected to the information processing device 1 via a network so that they can communicate with each other. The network may be a local area network or a wide area network.

[0043] The information processing device 1 includes a processor 11, a memory 12, and a communication unit 13. The information processing device 1 is, for example, a personal computer, a tablet computer, or a server.

[0044] The processor 11 is, for example, a central processing unit (CPU). The processor 11 implements a PSI data acquisition unit 111, a setting data acquisition unit 112, a forecast sales volume calculation unit 113, a forecast inventory calculation unit 114, a forecast production volume calculation unit 115, and an output unit 116.

[0045] The memory 12 is a storage device capable of storing various types of information, such as a random access memory (RAM), a hard disk drive (HDD), a solid state drive (SSD), or a flash memory. The memory 12 stores various types of information.

[0046] The memory 12 stores PSI data and setting data.

[0047] FIG. 2 is a diagram showing an example of PSI data in this embodiment, and FIG. 3 is a diagram showing an example of setting data in this embodiment.

[0048] The PSI data includes the actual production volume, actual sales volume, actual inventory volume, and actual backorder volume for a specific product for each past unit period. As shown in FIG. 2 , the memory 12 stores PSI data that associates a product model number for identifying the product, a date (year and month), the actual production volume, the actual sales volume, the actual inventory volume, and the actual backorder volume. For example, FIG. 2 shows the actual production volume, the actual sales volume, the actual inventory volume, and the actual backorder volume for a product with model number "a01" for June 2023 and July 2023. Note that, although the unit period in this embodiment is one month, the present disclosure is not particularly limited thereto and may be one day, one week, or any other period.

[0049] Actual production volume represents the amount of a specified product produced within a unit period. If the product is produced at another company's factory, the production volume is also called the order volume. Actual sales volume represents the amount of a specified product sold within a unit period. Actual inventory volume represents the amount of a specified product stored in inventory at the end of a unit period. Actual backlog order volume represents the amount of a specified product ordered within a unit period that has not been delivered at the end of the unit period. Note that backlog orders may occur even if there is inventory, because a company wants to maintain a certain amount of inventory.

[0050] The memory 12 may also store PSI data that associates a product number for identifying the product, information for identifying the customer, a date, actual production volume, actual sales volume, actual inventory volume, and actual backlog order volume.

[0051] The setting data includes the maximum production volume, maximum inventory volume, and maximum backorder volume for a given product in a unit period. As shown in Fig. 3, the memory 12 stores setting data that associates a product number for identifying the product with the maximum production volume in a unit period, the maximum inventory volume in a unit period, and the maximum backorder volume in a unit period. For example, Fig. 3 shows the maximum production volume, maximum inventory volume, and maximum backorder volume for a product with product number "a01" and a product with product number "a02" in a unit period. The unit period is the same period as the PSI data, e.g., one month.

[0052] Maximum production volume represents the maximum amount of a specified product that can be produced within a unit period. Maximum inventory volume represents the maximum amount of a specified product that can be stored as inventory within a unit period. Maximum backlog volume represents the maximum amount of orders for a specified product that have been received within a unit period that can be tolerated even if they have not been delivered by the end of the unit period.

[0053] The input unit 2 accepts PSI data input by a user. The processor 11 of the information processing device 1 stores the PSI data input by the input unit 2 in the memory 12. The input unit 2 also accepts setting data input by a user. The processor 11 of the information processing device 1 stores the setting data input by the input unit 2 in the memory 12.

[0054] The information processing device 1 may receive PSI data or setting data from an external device such as a server, and store the received PSI data or setting data in the memory 12 .

[0055] The input unit 2 also accepts a user's selection of a product for which production volume, sales volume, inventory volume, and backorder volume per unit period are to be predicted. The input unit 2 may also accept a user's selection of a unit period for prediction, such as one day, one week, or one month. The input unit 2 may also accept a user's selection of a period for prediction, such as one week, one month, or six months.

[0056] The PSI data acquisition unit 111 acquires the actual production volume, actual sales volume, actual inventory volume, and actual backorder volume of a specified product for each unit period in the past. The PSI data acquisition unit 111 reads PSI data including the actual production volume, actual sales volume, actual inventory volume, and actual backorder volume of a specified product for each unit period in the past from the memory 12. The PSI data acquisition unit 111 may read, for example, monthly PSI data for the past year from the memory 12. Furthermore, if the memory 12 stores daily PSI data and monthly production volume, sales volume, inventory volume, and backorder volume are predicted, the PSI data acquisition unit 111 may aggregate the daily PSI data to create monthly PSI data.

[0057] The PSI data acquiring unit 111 may also acquire the past actual sales volume of a similar product to the predetermined product. The PSI data acquiring unit 111 may read the past actual sales volume of the similar product from the memory 12.

[0058] The setting data acquisition unit 112 acquires the maximum production volume, maximum inventory volume, and maximum backorder volume of a predetermined product in a unit period. The setting data acquisition unit 112 reads setting data including the maximum production volume, maximum inventory volume, and maximum backorder volume of a predetermined product in a unit period from the memory 12.

[0059] The communication unit 13 acquires calendar information related to past and future days of the week and holidays, and weather information related to past and future weather. The communication unit 13 receives calendar information and weather information transmitted by an external server. The weather information includes, for example, weather information for the past year and weather information for the future week. The weather information may be used when calculating predicted sales volume on a daily or weekly basis.

[0060] The predicted sales volume calculation unit 113 calculates the predicted sales volume of the specified product for each future unit period, the predicted upper limit sales volume that is acceptable if the sales volume exceeds the predicted sales volume, and the predicted lower limit sales volume that is acceptable if the sales volume falls below the predicted sales volume, based on the actual sales volume of the specified product, calendar information, weather information, and actual sales volumes of similar products. For example, the predicted sales volume calculation unit 113 calculates the predicted sales volume, the predicted upper limit sales volume, and the predicted lower limit sales volume of the specified product for each month over the next six months.

[0061] The forecast sales volume calculation unit 113 inputs the actual sales volume of a specified product, calendar information, weather information, and actual sales volumes of similar products into a forecast model created by pre-learning, and calculates the forecast sales volume, forecast upper limit sales volume, and forecast lower limit sales volume by obtaining the forecast sales volume, forecast upper limit sales volume, and forecast lower limit sales volume output from the forecast model.

[0062] The forecast sales volume calculation unit 113 estimates the forecast sales volume, forecast upper sales volume, and forecast lower sales volume by inputting the acquired actual sales volume, calendar information, weather information, and actual sales volume of similar products into a trained prediction model obtained by machine learning the relationship between the actual sales volume of a specified product, calendar information, weather information, and actual sales volume of similar products and the sales volume, upper sales volume, and lower sales volume.

[0063] The predictive model is created by machine learning. Examples of machine learning include supervised learning, which learns the relationship between input and output using training data in which labels (output information) are assigned to input information; unsupervised learning, which builds a data structure only from unlabeled input; semi-supervised learning, which handles both labeled and unlabeled data; and reinforcement learning, which learns behaviors that maximize rewards through trial and error. Specific machine learning techniques include neural networks (including deep learning using multi-layer neural networks), genetic programming, decision trees, Bayesian networks, and support vector machines (SVMs). Any of the specific examples listed above may be used in the machine learning of the present disclosure.

[0064] The machine learning of the prediction model may be performed using the actual sales volume of a specific product for the past three months, calendar information, weather information, and actual sales volumes of similar products as input values, and the sales volume, upper sales limit, and lower sales limit of the specific product for the current month as output values. The prediction model may also be trained using forecasted sales volumes calculated in the past. In this case, the difference between the actual sales volume and the forecasted sales volume may be used as training data for the upper sales limit and the lower sales limit.

[0065] Furthermore, the forecast sales volume calculation unit 113 may calculate the forecast sales volume, forecast upper sales volume, and forecast lower sales volume for each future unit period of a predetermined product based only on the actual sales volume of the predetermined product. The forecast sales volume calculation unit 113 may calculate the forecast sales volume, forecast upper sales volume, and forecast lower sales volume by inputting the actual sales volume of the predetermined product into a prediction model created by pre-learning and acquiring the forecast sales volume, forecast upper sales volume, and forecast lower sales volume output from the prediction model.

[0066] Furthermore, the predicted sales volume calculation unit 113 may calculate the predicted sales volume, predicted upper sales volume, and predicted lower sales volume for each future unit period of the predetermined product based on the actual sales volume of the predetermined product and at least one of calendar information, weather information, and actual sales volumes of similar products. The predicted sales volume calculation unit 113 may calculate the predicted sales volume, predicted upper sales volume, and predicted lower sales volume by inputting the actual sales volume of the predetermined product and at least one of calendar information, weather information, and actual sales volumes of similar products into a prediction model created by pre-learning, and acquiring the predicted sales volume, predicted upper sales volume, and predicted lower sales volume output from the prediction model.

[0067] The predicted inventory quantity calculation unit 114 calculates a predicted inventory quantity for each future unit period of a specified product that satisfies the constraints that the predicted remaining order quantity will not exceed the maximum remaining order quantity when the predicted sales quantity reaches the predicted sales upper limit, and that the predicted inventory quantity will not exceed the maximum inventory quantity when the predicted sales quantity reaches the predicted sales lower limit, based on the actual inventory quantity, the actual backlog order quantity, the maximum inventory quantity, the maximum backlog order quantity, the predicted sales quantity, the upper forecast sales quantity, and the lower forecast sales limit. For example, the predicted inventory quantity calculation unit 114 calculates the predicted inventory quantity for each month over the next six months of the specified product.

[0068] More specifically, the forecasted inventory quantity calculation unit 114 calculates the forecasted inventory quantity and forecasted backorder quantity for unit period t using the actual inventory quantity and actual backorder quantity for unit period t-1, the forecasted sales quantity for unit period t, and the variable production quantity for unit period t. Next, the forecasted inventory quantity calculation unit 114 calculates multiple combinations of the forecasted inventory quantity and forecasted backorder quantity for unit period t by changing the production quantity. Next, the forecasted inventory quantity calculation unit 114 selects, from the multiple combinations, a combination that satisfies the constraint conditions and minimizes the sum of the forecasted inventory quantity and the forecasted backorder quantity. Next, the forecasted inventory quantity calculation unit 114 calculates the forecasted inventory quantity for unit period t from the selected combination. Then, the forecasted inventory quantity calculation unit 114 sequentially calculates the forecasted inventory quantities for unit periods t+1, t+2, t+3, t+4, and t+5.

[0069] The forecast production calculation unit 115 calculates a forecast production volume for a predetermined product for each future unit period based on the actual inventory volume, the forecast sales volume, and the forecast inventory volume. More specifically, the forecast production calculation unit 115 calculates the forecast production volume by adding the forecast inventory volume to the forecast sales volume and subtracting the actual inventory volume from the added amount.

[0070] The output unit 116 outputs the forecast sales volume, forecast upper sales volume, forecast lower sales volume, forecast inventory volume, and forecast production volume for each unit period to the display unit 3. Note that the output unit 116 may output only the forecast production volume for each unit period to the display unit 3.

[0071] The display unit 3 is, for example, a liquid crystal display device, and displays the information output by the output unit 116. The display unit 3 is connected to the information processing device 1 via a wired or wireless connection so that they can communicate with each other. Note that the display unit 3 may also be connected to the information processing device 1 via a network so that they can communicate with each other. The network may be a local area network or a wide area network.

[0072] The display unit 3 displays the forecast sales volume, forecast upper sales volume, forecast lower sales volume, forecast inventory volume, and forecast production volume for each unit period. Note that the display unit 3 may display only the forecast production volume for each unit period.

[0073] Next, the prediction process performed by the information processing device 1 according to the embodiment of the present disclosure will be described.

[0074] FIG. 4 is a flowchart illustrating the prediction process performed by the information processing device 1 according to the embodiment of the present disclosure.

[0075] First, in step S1, the PSI data acquisition unit 111 acquires PSI data including the actual production volume, actual sales volume, actual inventory volume, and actual backlog order volume for a specific product for each past unit period. Note that the input unit 2 may accept in advance input by a user of information for identifying a product and information for identifying a unit period.

[0076] Next, in step S2, the setting data acquisition unit 112 acquires setting data including the maximum production volume, maximum inventory volume, and maximum backorder volume for a predetermined product in a unit period. Note that the input unit 2 may accept the maximum production volume, maximum inventory volume, and maximum backorder volume input by the user in advance.

[0077] Next, in step S3, the communication unit 13 acquires calendar information regarding past and future days of the week and holidays, and weather information regarding past and future weather, and the PSI data acquisition unit 111 acquires the actual sales volume for each unit period in the past of products similar to the specified product.

[0078] Next, in step S4, the forecast sales calculation unit 113 calculates the forecast sales volume, forecast upper limit sales volume, and forecast lower limit sales volume for the specified product for each future unit period based on the actual sales volume of the specified product, calendar information, weather information, and actual sales volumes of similar products. The forecast sales calculation unit 113 inputs the actual sales volume of the specified product, calendar information, weather information, and actual sales volumes of similar products into the prediction model created by pre-learning, and obtains the forecast sales volume, forecast upper limit sales volume, and forecast lower limit sales volume output from the prediction model.

[0079] Next, in step S5, the predicted inventory quantity calculation unit 114 calculates the predicted inventory quantity for each future unit period of a specified product based on the actual inventory quantity, actual remaining order quantity, maximum inventory quantity, maximum remaining order quantity, predicted sales quantity, predicted upper sales limit quantity, and predicted lower sales limit quantity, so as to satisfy the constraints that when the predicted sales quantity becomes the predicted upper sales limit quantity, the predicted remaining order quantity does not exceed the maximum remaining order quantity, and when the predicted sales quantity becomes the predicted lower sales limit quantity, the predicted inventory quantity does not exceed the maximum inventory quantity.

[0080] Here, the calculation of the optimal predicted inventory amount will be described in more detail.

[0081] FIG. 5 is a schematic diagram for explaining in more detail the calculation of the predicted inventory amount by the predicted inventory amount calculation unit 114 according to this embodiment.

[0082] In FIG. 5, the unit period is one month. The predicted stock amount calculation unit 114 calculates the predicted stock amount for the current month and the following months on the first day of the current month. ｔrepresents the predicted sales volume for this month t, and p ｔ＋１ represents the forecast sales volume for the next month t+1, and s ｔ represents the forecast backlog of orders for this month t, and s ｔ＋１ represents the forecast backlog of orders for the next month t+1, and s ｔ－１ represents the backlog of orders for the previous month t-1, and z ｔ represents the forecast inventory amount for this month t, and z ｔ＋１ represents the predicted inventory amount for the next month t+1, and z ｔ－１ represents the actual inventory amount for the previous month t-1, and o ｔ represents the production volume of this month t, and o ｔ＋１ represents the production volume for the next month t+1.

[0083] The order volume for this month t is (p ｔ +s ｔ－１ ) and the spot amount for this month t is (o ｔ +z ｔ－１ ) The delivery amount for this month t is c ｔ is constrained by the smaller of the order volume and the actual volume for this month t. ｔ is min(p ｔ +s ｔ－１ , o ｔ +z ｔ－１ ) As a result, the forecast backlog of orders at the end of this month s ｔ is (p ｔ +s ｔ－１ )-c ｔ The predicted inventory amount at the end of this month is z ｔ is (o ｔ +z ｔ－１ )-c ｔ This becomes:

[0084] The combination of the actual backorder amount and the actual inventory amount of the previous month t-1 (s ｔ－１ , z ｔ－１ ) and the predicted sales volume p ｔ If the production volume is obtained, ｔ Once this is determined, the combination of the forecast backorder amount and forecast inventory amount at the end of this month (s ｔ , z ｔ ) is automatically determined.

[0085] If a combination of the backlog of orders and the inventory amount is considered as one state, the state will transition according to the production volume for each month.

[0086] The predicted inventory amount calculation unit 114 calculates a plurality of production amounts o that are equal to or less than the maximum production amount. １ ｔ , o ２ ｔ , ... and select multiple production quantities o １ ｔ , o ２ ｔ , ...Transition of the state of this month t (combination of forecast backorder amount and forecast inventory amount) for each (s １ ｔ , z １ ｔ ), (s ２ ｔ , z ２ ｔ ), .... The forecasted inventory quantity calculation unit 114 then calculates the transition of the state (combination of forecasted backorder quantity and forecasted inventory quantity) for the next month and thereafter for each of the multiple production quantities. This allows the forecasted inventory quantity calculation unit 114 to calculate the transition of the state (combination of forecasted backorder quantity and forecasted inventory quantity) in monthly increments from the previous month to the forecast period (e.g., six months ahead).

[0087] The predicted inventory amount calculation unit 114 calculates the cost y ｉ ｔ The estimated inventory quantity calculation unit 114 calculates the cost y of each state based on the following formula (1), and selects the state transition that minimizes the total cost of the state for each month. ｉ ｔ Calculate.

[0088] y ｉ ｔ = f (o ｉ ｔ |p ｔ +σ ＋ ｔ , s ｔ－１ , z ｔ－１ , o ｔ－１ ;s ０ , z ０ ) + f(o ｉ ｔ |p ｔ -σ －ｔ , s ｔ－１ , z ｔ－１ , o ｔ－１ ;s ０ , z ０ ) (1) In the above formula (1), o ｉ ｔ is a variable, and p ｔ +σ ＋ ｔ , s ｔ－１ , z ｔ－１ , and o ｔ－１ is a given variable, and s ０ and z ０ is a constant. The cost of each state y ｉ ｔ For both state i and time t, ｉ ｔ is considered a variable. A given variable is a variable with respect to time t but does not depend on state i. A constant is a fixed value that is independent of both state i and time t.

[0089] Here, p ｔ +σ ＋ ｔ represents the predicted sales upper limit, and p ｔ -σ － ｔ represents the predicted sales lower limit. ｔ When the predicted sales volume p ｔ is the predicted sales upper limit p ｔ +σ ＋ ｔ Even if the forecast backlog s ｔ is the maximum remaining order quantity s ０ In addition, the predicted inventory quantity calculation unit 114 selects a cost that does not exceed a certain production quantity o ｔ When the predicted sales volume p ｔ is the predicted sales lower limit p ｔ -σ － ｔ Even if the forecast inventory amount z ｔ is the maximum inventory amount z ０ Choose a cost that does not exceed

[0090] Furthermore, the functions of the first and second terms in the above formula (1) are expressed by the abstract function shown in the following formula (2).

[0091] f(o|p,s,z,o';s ０ , z ０ )=α[p+s−c−s ０ ] ＋ +β[o+z-c-z ０ ] ＋ +γ(p+s−c)+ω(o+z−c)+η(o−o′) (2) In equation (2), c is min(p+s, o+z), and the function [x] ＋ means max(0, x). In addition, in formula (2), α, β, γ, ω, and η are coefficients. α, β, γ, and ω satisfy the relationship α, β >> γ, ω.

[0092] The forecasted inventory quantity calculation unit 114 lists all states (combinations of forecasted backorder quantities and forecasted inventory quantities) for a predetermined period (for example, six months) ahead, and selects a transition of a state that minimizes the total cost value from among the multiple states. As shown in equation (2), the forecasted backorder quantity s ｔ is the maximum remaining order quantity s ０ The cost when the forecast backlog is larger than s ｔ is the maximum remaining order quantity s ０ Since the cost is larger than the cost when the forecast backlog order quantity s ｔ is the maximum remaining order quantity s ０ Similarly, the state where the forecast stock amount z is larger than ｔ is the maximum inventory amount z ０ If the cost is greater than the forecast inventory amount z ｔ is the maximum inventory amount z ０ Since the cost is larger than the cost when the forecast inventory amount is smaller than z ｔ is the maximum inventory amount z ０ States greater than 0 are not selected.

[0093] In addition, the smaller the remaining order quantity and inventory quantity, the better. ｔ and forecast inventory quantity z ｔ The forecast backlog order quantity s that minimizes the total value of ｔ and forecast inventory quantity z ｔ A combination of is selected.

[0094] Furthermore, the fluctuation of production volume (o ｔ -o ｔ－１) is taken into consideration, a state (combination of forecast backorder amount and forecast inventory amount) is selected such that the monthly production volume remains the same.

[0095] The forecasted inventory quantity calculation unit 114 calculates the cost of a combination of forecasted backorder quantities and forecasted inventory quantities for each of multiple production quantities in unit period t based on equation (2) and selects the combination of forecasted backorder quantities and forecasted inventory quantities that results in the lowest cost. The forecasted inventory quantity calculation unit 114 calculates the forecasted inventory quantity from the selected combination as the forecasted inventory quantity for unit period t. Then, the forecasted inventory quantity calculation unit 114 also calculates the forecasted inventory quantities for unit periods t+1 and beyond using the same method as above.

[0096] If the predicted sales volume in the first unit period exceeds the maximum production volume, the predicted inventory quantity calculation unit 114 may increase the predicted inventory quantity in a second unit period that precedes the first unit period. In this case, the predicted inventory quantity calculation unit 114 may add the amount obtained by subtracting the maximum production volume from the predicted sales volume in the first unit period to the predicted inventory quantity in the second unit period.

[0097] Furthermore, when a specific product passes through multiple warehouses from the time it is produced until it is sold and the specific product moves through multiple warehouses for each unit period to be predicted, the predicted inventory quantity calculation unit 114 may calculate the predicted inventory quantity for each of the multiple warehouses. In this way, even if there is a lead time between production and sale, the predicted inventory quantity and predicted production quantity can be calculated taking the lead time into consideration.

[0098] 4 , next, in step S6, the forecast production volume calculation unit 115 calculates a forecast production volume for each future unit period of a predetermined product based on the actual inventory volume, the forecast sales volume, and the forecast inventory volume. Here, the forecast production volume calculation unit 115 calculates the forecast production volume by adding the forecast inventory volume to the forecast sales volume and subtracting the actual inventory volume from the added amount.

[0099] Next, in step S7, the output unit 116 outputs the forecast sales volume, forecast upper sales volume, forecast lower sales volume, forecast inventory volume, and forecast production volume for each unit period to the display unit 3.

[0100] Next, in step S8, the display unit 3 displays the forecast sales volume, forecast upper sales volume, forecast lower sales volume, forecast inventory volume, and forecast production volume for each unit period.

[0101] FIG. 6 is a diagram showing an example of the forecast sales volume, forecast upper sales volume, forecast lower sales volume, forecast inventory volume, and forecast production volume for each unit period, which are displayed on the display unit 3 in this embodiment.

[0102] 6 , the display unit 3 displays a forecasted sales volume 31, a forecasted upper sales volume 32, a forecasted lower sales volume 33, a forecasted inventory volume 34, and a forecasted production volume 35 for each month for the six months from November to April. Furthermore, for each month, a straight line connects the forecasted sales volume 31 and the forecasted upper sales volume 32, and a straight line connects the forecasted sales volume 31 and the forecasted lower sales volume 33. This allows the user to recognize how far the forecasted sales volume 31 is from the forecasted sales volume 31, and how far the forecasted upper sales volume 32 and the forecasted lower sales volume 33 are from the forecasted sales volume 31.

[0103] In this embodiment, the forecast sales volume, forecast upper limit sales volume, forecast lower limit sales volume, forecast inventory volume, and forecast production volume for each unit period are displayed, but the present disclosure is not limited to this, and only the forecast production volume may be displayed.

[0104] Fig. 7 is a diagram showing an example of a user interface screen displayed on the display unit 3 in this embodiment. While the display screen shown in Fig. 6 displays only the prediction results, the user interface screen shown in Fig. 7 displays the prediction results and also accepts data input by the user.

[0105] The user interface screen shown in Figure 7 includes a PSI data input area 301, a product selection area 302, a forecast period selection area 303, a maximum production volume input area 304, a number of production volume options input area 305, a maximum inventory quantity input area 306, a maximum backlog of orders input area 307, a coefficient input area 308, a sales volume display area 309, an inventory quantity display area 310, and a production volume display area 311.

[0106] The PSI data input area 301 accepts user input of past PSI data. The user drags and drops a file storing past PSI data into the PSI data input area 301. This inputs the past PSI data to be used for prediction. The user may also select a file storing past PSI data from multiple files.

[0107] The product selection area 302 displays a drop-down list showing multiple selectable product names, and accepts the user's selection of the product name to be predicted. The user selects the desired product name from the drop-down list displayed in the product selection area 302.

[0108] The prediction period selection area 303 displays a drop-down list from which the year and month from which the prediction is to start can be selected, and accepts the user's selection of the year and month from which the prediction is to start. The user selects the year and month from which the prediction is to start from the drop-down list displayed in the prediction period selection area 303. When the year and month from which the prediction is to start are selected, the prediction results for up to six months from the selected year and month are displayed.

[0109] The maximum production volume input area 304 accepts input of the maximum monthly production volume by the user. The minimum unit of production volume is predetermined for each product. Therefore, products are produced in units of, for example, 100 units, 1,000 units, or 10,000 units. The user inputs the maximum monthly production volume by pressing the plus button or minus button displayed in the maximum production volume input area 304.

[0110] The production volume option number input area 305 accepts user input of the number of production volume options to be listed when calculating the forecasted inventory amount. The production volume options are used when calculating multiple combinations of forecasted backorder amounts and forecasted inventory amounts for each unit period. A large number of options improves the forecast accuracy of the forecasted backorder amounts and forecasted inventory amounts, but increases the time required for the calculation process. On the other hand, a small number of options reduces the forecast accuracy of the forecasted backorder amounts and forecasted inventory amounts, but decreases the time required for the calculation process. For example, if the number of production volume options input by the user is 10 and the maximum production amount is 10,000, then 0, 1,000, 2,000, 3,000, ..., 10,000 are used as production volume options (11 options including 0).

[0111] The maximum inventory amount input area 306 accepts input by the user of the maximum inventory amount for one month.

[0112] The maximum remaining order quantity input area 307 accepts input by the user of the maximum remaining order quantity for one month.

[0113] The coefficient input area 308 accepts user input of a coefficient for making the monthly production volume the same. The input coefficient is the coefficient η in the above formula (2), and is multiplied by the difference between the current month's production volume and the previous month's production volume. As the coefficient η increases, the fluctuation in monthly production volume decreases, and as the coefficient η decreases, the fluctuation in monthly production volume increases. Note that the coefficient η may be 0. When the coefficient η becomes 0, fluctuations in monthly production volume are not taken into account when calculating the predicted inventory amount.

[0114] The sales volume display area 309 displays the forecast sales volume, forecast upper sales volume, and forecast lower sales volume for each unit period. In FIG. 7 , the forecast sales volume, forecast upper sales volume, and forecast lower sales volume for six months from January 2024 are displayed for each month. Note that the sales volume display area 309 displays not only the forecast sales volume, forecast upper sales volume, and forecast lower sales volume for each unit period, but also the actual sales volume for each unit period included in the PSI data. In FIG. 7 , the actual sales volume for 11 months from April 2023 is displayed for each month. Note that the forecast sales volume and actual sales volume may be displayed overlapping each other.

[0115] The inventory quantity display area 310 displays the predicted inventory quantity for each unit period. In FIG. 7 , the predicted inventory quantity for six months from January 2024 is displayed on a monthly basis. Note that the inventory quantity display area 310 displays not only the predicted inventory quantity for each unit period, but also the actual inventory quantity for each unit period included in the PSI data. In FIG. 7 , the actual inventory quantity for 11 months from April 2023 is displayed on a monthly basis. Note that the predicted inventory quantity and actual inventory quantity may be displayed overlapping each other.

[0116] The production volume display area 311 displays the predicted production volume for each unit period. In FIG. 7 , the predicted production volume for six months from January 2024 is displayed for each month. The production volume display area 311 displays not only the predicted production volume for each unit period, but also the actual production volume for each unit period included in the PSI data. In FIG. 7 , the actual production volume for 11 months from April 2023 is displayed for each month. The predicted production volume and the actual production volume may be displayed overlapping each other.

[0117] The minimum inventory quantity may be set based on an estimation error, which is the difference between the predicted sales quantity and the predicted sales upper limit quantity.

[0118] FIG. 8 is a diagram showing an example of a user interface screen displayed on the display unit 3 in this embodiment, which screen includes forecast sales volume, forecast inventory volume, and forecast production volume for each unit period.

[0119] The user interface screen shown in Figure 8 displays the PSI plan for product number x01. The PSI plan includes a sales plan (forecasted sales volume), an inventory plan (forecasted inventory volume), and a production plan (forecasted production volume). In Figure 8, the display unit 3 displays a forecasted sales volume 41, a forecasted upper sales volume 42, a forecasted lower sales volume 43, a forecasted inventory volume 44, and a forecasted production volume 45 for each of the three months of the current month, the next month, and the month after that. In addition, for each month, a line connects the forecasted sales volume 41 and the forecasted upper sales volume 42, and a line connects the forecasted sales volume 41 and the forecasted lower sales volume 43.

[0120] When the forecast sales volume calculation unit 113 obtains the actual sales volume of a predetermined product, it automatically calculates a future forecast sales volume 41, a forecast upper sales volume 42, and a forecast lower sales volume 43. The forecast sales volume 41 is represented by a dot, and the forecast upper sales volume 42 and the forecast lower sales volume 43 are displayed above and below the forecast sales volume 41. The difference between the forecast sales volume 41 and the forecast upper sales volume 42 is a positive estimation error 421, and the difference between the forecast sales volume 41 and the forecast lower sales volume 43 is a negative estimation error 422. For example, the positive estimation error 421 is +10 units, and the negative estimation error 422 is -10 units. The positive estimation error 421 and the negative estimation error 422 may be different values.

[0121] The positive estimation error 421 is the minimum inventory amount (minimum inventory amount 441) that should be maintained when creating an inventory plan. The production plan is created so that the predicted inventory amount 44 exceeds the minimum inventory amount 441. The production plan is created taking into account the maximum monthly production amount 451 and the adjustable production unit amount. The production unit amount indicates the smallest unit to be produced in one month, such as 100 units. The inventory plan is created so that the predicted inventory amount 44 is as close to the minimum inventory amount 441 as possible, and so that the predicted inventory amount 44 does not fall below the minimum inventory amount 441.

[0122] The input unit 2 may accept a user adjustment of the positive estimated error 421. The user may adjust the positive estimated error 421 by operating a mouse on the user interface screen. This allows the user's experience with discrepancies in the forecast sales volume 41 to be reflected. As the range of the positive estimated error 421 increases, the minimum inventory quantity 441 also increases accordingly.

[0123] Conversely, the positive estimated error 421 changes when the minimum inventory quantity 441 is changed. By changing the minimum inventory quantity 441, the user can intuitively understand the estimated error of the forecast sales volume 41.

[0124] Furthermore, when the predicted sales volume 41 exceeds the maximum production volume 451, the predicted production volume calculation unit 115 increases the predicted production volume 45 for the previous month. As a result, the predicted inventory volume 44 for the previous month increases. This makes it possible to avoid missing sales opportunities. For example, in the example of FIG. 8, the predicted sales volume 41 for the month after next exceeds the maximum production volume 451. Therefore, the predicted production volume calculation unit 115 increases the predicted production volume 45 for the next month in advance.

[0125] As the forecast production volume 45 increases, the forecast inventory volume 44 increases significantly above the minimum inventory volume 441. As a result, there will be too much inventory as of the following month. However, the month after that, the forecast sales volume 41 increases, so the forecast inventory volume 44 decreases, and the forecast inventory volume 44 returns to approximately the same amount as the minimum inventory volume 441.

[0126] The maximum production volume 451 may be adjusted between multiple product part numbers. For example, when a total of 100 units of two different product part numbers can be produced, the user may determine the maximum production volume 451 for each product. For example, the user may produce 50 units of one product and 50 units of the other product, or 70 units of one product and 30 units of the other product. The user may use the mouse to drag the horizontal line of the maximum production volume 451 up or down on the user interface screen of the PSI plan for each product part number of interest. When the maximum production volume 451 is changed, the inventory plan and production plan are recalculated accordingly. When the maximum production volume 451 for one product part number changes, the maximum production volumes 451 for all other product parts related to that product part number change accordingly, allowing the user to search for the optimal balance while viewing each PSI plan.

[0127] In each of the above embodiments, each component may be configured with dedicated hardware, or may be realized by executing a software program suitable for that component. Each component may be realized by a program execution unit such as a CPU or processor reading and executing a software program recorded on a recording medium such as a hard disk or semiconductor memory. Furthermore, the program may be executed by another independent computer system by recording the program on a recording medium and transferring it, or by transferring the program via a network.

[0128] Some or all of the functions of the device according to the embodiment of the present disclosure are typically realized as an LSI (Large Scale Integration), which is an integrated circuit. These may be individually integrated into a single chip, or some or all of them may be integrated into a single chip. Furthermore, the integrated circuit is not limited to an LSI, and may be realized using a dedicated circuit or a general-purpose processor. It is also possible to use an FPGA (Field Programmable Gate Array), which can be programmed after LSI manufacturing, or a reconfigurable processor, which can reconfigure the connections and settings of circuit cells within an LSI.

[0129] Furthermore, some or all of the functions of the device according to the embodiment of the present disclosure may be realized by a processor such as a CPU executing a program.

[0130] Furthermore, all the numbers used above are merely examples to specifically explain the present disclosure, and the present disclosure is not limited to the numbers used as examples.

[0131] The order in which the steps are executed in the above flowchart is merely an example for specifically explaining the present disclosure, and other orders may be used as long as similar effects are obtained. Also, some of the steps may be executed simultaneously (in parallel) with other steps.

[0132] The technology disclosed herein can determine the optimal forecast inventory quantity using the forecast sales quantity while taking into account constraints at the production site, and can automatically determine the forecast production quantity using the forecast inventory quantity, and is therefore useful as a technology for calculating the forecast production quantity for a specified product per unit period and outputting the forecast production quantity per unit period.

Claims

1. An information processing method executed by a computer, comprising: acquiring actual production volume, actual sales volume, actual inventory volume, and actual backorder volume for a specified product for each past unit period; acquiring a maximum inventory volume and a maximum backorder volume for the specified product for the unit period; calculating a forecast sales volume for the specified product for each future unit period based on the actual sales volumes, a forecast upper sales volume that is acceptable when the sales volume exceeds the forecast sales volume, and a forecast lower sales volume that is acceptable when the sales volume falls below the forecast sales volume; calculating a forecast inventory volume for each future unit period for the specified product based on the actual inventory volume, the actual backorder volume, the maximum inventory volume, the maximum backorder volume, the forecast sales volume, the forecast upper sales volume, and the forecast lower sales volume, satisfying a constraint that the forecast backorder volume does not exceed the maximum backorder volume when the forecast sales volume becomes the forecast upper sales volume, and that the forecast inventory volume does not exceed the maximum inventory volume when the forecast sales volume becomes the forecast lower sales volume; Calculating a forecast production volume for each future unit period of the specified product based on the actual inventory volume, the forecast sales volume, and the forecast inventory volume; and outputting the forecast production volume for each unit period.

2. The information processing method according to claim 1, wherein the output includes outputting the forecast sales volume, the forecast upper sales volume, the forecast lower sales volume, the forecast inventory volume, and the forecast production volume for each unit period.

3. An information processing method as claimed in claim 1 or 2, further comprising: obtaining a maximum production volume for the specified product in a unit period; and, if the forecasted sales volume in a first unit period exceeds the maximum production volume, increasing the forecasted inventory amount in a second unit period preceding the first unit period.

4. An information processing method as described in claim 1 or 2, wherein calculation of the forecast sales volume, the forecast upper sales volume, and the forecast lower sales volume includes calculating the forecast sales volume, the forecast upper sales volume, and the forecast lower sales volume by inputting the actual sales volume into a prediction model created by pre-learning, and obtaining the forecast sales volume, the forecast upper sales volume, and the forecast lower sales volume output from the prediction model.

5. An information processing method as claimed in claim 1 or 2, further comprising obtaining calendar information relating to past and future days of the week and holidays, weather information relating to past and future weather, and past actual sales volumes of products similar to the specified product, and calculating the forecast sales volume, the forecast upper sales volume, and the forecast lower sales volume includes calculating the forecast sales volume, the forecast upper sales volume, and the forecast lower sales volume based on the actual sales volume of the specified product, the calendar information, the weather information, and the actual sales volumes of the similar products.

6. An information processing method as claimed in claim 1 or 2, wherein the calculation of the forecasted inventory amount includes: calculating the forecasted inventory amount and the forecasted backorder amount for unit period t using the actual inventory amount and the actual backorder amount for unit period t-1, the forecasted sales amount for the unit period t, and a variable production amount for the unit period t; calculating a plurality of combinations of the forecasted inventory amount and the forecasted backorder amount for the unit period t by changing the production amount; selecting a combination from the plurality of combinations that satisfies the constraint condition and has the smallest sum of the forecasted inventory amount and the forecasted backorder amount; and calculating the forecasted inventory amount from the selected combination as the forecasted inventory amount for the unit period t.

7. An information processing method according to claim 1 or 2, wherein the calculation of the forecast production volume includes calculating the forecast production volume by adding the forecast inventory volume to the forecast sales volume and subtracting the actual inventory volume from the added volume.

8. An information processing method as claimed in claim 1 or 2, wherein the calculation of the predicted inventory amount includes, when the specified product passes through multiple warehouses from the time the specified product is produced to the time it is sold, and the specified product moves through the multiple warehouses for each predicted unit period, calculating the predicted inventory amount for each of the multiple warehouses.

9. A first acquisition unit that acquires actual production volume, actual sales volume, actual inventory volume, and actual backorder volume for a specified product for each past unit period; a second acquisition unit that acquires a maximum inventory volume and a maximum backorder volume for the specified product in a unit period; a first calculation unit that calculates a forecast sales volume for the specified product for each future unit period based on the actual sales volumes, a forecast upper sales volume that is acceptable when the sales volume exceeds the forecast sales volume, and a forecast lower sales volume that is acceptable when the sales volume falls below the forecast sales volume; and a second calculation unit that calculates a forecast inventory volume for the specified product for each future unit period based on the actual inventory volume, the actual backorder volume, the maximum inventory volume, the maximum backorder volume, the forecast sales volume, the forecast upper sales volume, and the forecast lower sales volume, satisfying a constraint that the forecast backorder volume does not exceed the maximum backorder volume when the forecast sales volume becomes the forecast upper sales volume, and that the forecast inventory volume does not exceed the maximum inventory volume when the forecast sales volume becomes the forecast lower sales volume. an information processing device comprising: a third calculation unit that calculates a predicted production volume for each future unit period of the specified product based on the actual inventory volume, the predicted sales volume, and the predicted inventory volume; and an output unit that outputs the predicted production volume for each unit period.

10. acquiring actual production volume, actual sales volume, actual inventory volume, and actual backorder volume for each past unit period of a specified product; acquiring a maximum inventory volume and a maximum backorder volume for the specified product in each unit period; calculating a forecast sales volume for each future unit period of the specified product based on the actual sales volume, a forecast upper sales volume that is acceptable when the sales volume exceeds the forecast sales volume, and a forecast lower sales volume that is acceptable when the sales volume falls below the forecast sales volume; calculating a forecast inventory volume for each future unit period of the specified product based on the actual inventory volume, the actual backorder volume, the maximum inventory volume, the maximum backorder volume, the forecast sales volume, the forecast upper sales volume, and the forecast lower sales volume, which satisfies the constraint that the forecast backorder volume does not exceed the maximum backorder volume when the forecast sales volume becomes the forecast upper sales volume, and the forecast inventory volume does not exceed the maximum inventory volume when the forecast sales volume becomes the forecast lower sales volume; calculating a forecast production volume for each future unit period of the specified product based on the actual inventory volume, the forecast sales volume, and the forecast inventory volume; an information processing program that causes a computer to function so as to output the predicted production amount for each unit period.

Images