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

Abstract

This information processing device: acquires current and past demand results for a product; creates, on the basis of the demand results, a plurality of demand forecast patterns indicating daily demand in a future first period; calculates, for each of the plurality of demand forecast patterns, daily production plan amounts, sales plan amounts, and inventory plan amounts using a dynamic buffer management algorithm; creates, for each of the plurality of demand forecast patterns, a plurality of sets of PSI (production / sales / inventory) plan information including monthly production plan amounts, sales plan amounts and inventory plan amounts obtained by aggregating the daily production plan amounts, sales plan amounts and inventory plan amounts by month; and selects one set of the PSI plan information from among the plurality of sets of PSI plan information on the basis of at least one of cost, the number of times the inventory runs out, and the number of times the inventory becomes excessive.

Description

Information Processing Method, Information Processing Apparatus, and Information Processing Program

[0001] The present disclosure relates to a technique for creating product PSI (Production / Sales / Inventory) plan information.

[0002] For example, in Patent Document 1, the incoming quantity of items stored in a warehouse is predicted, the outgoing quantity of items shipped from the warehouse is predicted, a future inventory quantity prediction value is calculated from the prediction results of the incoming quantity and the outgoing quantity, and the upper and lower quantity reference values as preset appropriate inventory quantities are compared with the future inventory quantity prediction value to formulate an adjusted incoming / outgoing plan to avoid excess or shortage inventory associated with the predicted inventory quantity fluctuations. An adjusted incoming / outgoing quantity value is calculated from the adjusted incoming / outgoing plan, and a final predicted future inventory quantity value for each day is calculated based on the future inventory quantity prediction value and the adjusted incoming / outgoing quantity value. Also, an inventory management apparatus is disclosed that calculates an optimized adjusted incoming / outgoing quantity to maintain the standing inventory quantity within the range of the upper and lower quantity reference values based on the final predicted future inventory quantity value.

[0003] Also, for example, in Patent Document 2, a control is executed to create a PSI plan editing table that can switch between daily display, monthly display, and partial daily display by referring to PSI information indicating planned values of production, sales, and inventory and lead time indicating the period from order placement to delivery of a product, and display in a cell the lead time and the inventory quantity indicated by the PSI information associated with the date. An input field for inputting to the cell is provided in the PSI plan editing table, and a control is executed to input or calculate in a batch the numerical values or calculation formulas of the cells across a plurality of cells. A control is also executed to calculate the numerical values of the cells according to the calculated values in which the lead time is reflected in time series. An inventory management apparatus that executes these controls is disclosed.

[0004] However, with the above conventional techniques, it has been difficult to achieve overall supply chain efficiency improvement and cost reduction, and further improvement has been required.

[0005] Japanese Patent Application Laid-Open No. 2000-90171 Japanese Patent Application Laid-Open No. 2016-184378

[0006] This disclosure is made to solve the above-mentioned problems and aims to provide a technology that can reduce the risk of stockouts and excess inventory, thereby achieving overall supply chain efficiency and cost reduction.

[0007] An information processing method according to one aspect of the present disclosure is an information processing method performed by a computer, comprising: acquiring current and past demand performance for a product; creating a plurality of demand forecast patterns that show daily demand in a first future period based on the demand performance; calculating daily production plan quantities, sales plan quantities, and inventory plan quantities for each of the plurality of demand forecast patterns using a dynamic buffer management algorithm; creating a plurality of PSI (production / sales / inventory) plan information for each of the plurality of demand forecast patterns, including monthly production plan quantities, sales plan quantities, and inventory plan quantities obtained by aggregating the daily production plan quantities, sales plan quantities, and inventory plan quantities on a monthly basis; and selecting one PSI plan information from the plurality of PSI plan information based on at least one of cost, the number of times inventory was out of stock, and the number of times inventory was in excess.

[0008] According to this disclosure, it is possible to reduce the risk of stockouts and decrease excess inventory, thereby achieving overall supply chain efficiency and cost reduction.

[0009] This figure shows the configuration of the information processing device according to Embodiment 1. This figure shows an example of PSI plan information created by the PSI plan creation unit in Embodiment 1. This figure shows an example of demand performance acquired by the performance acquisition unit in Embodiment 1. This figure shows an example of a plurality of demand forecast patterns created by the demand forecast pattern creation unit in Embodiment 1. This figure shows an example of daily provisional production plan quantity, provisional sales plan quantity, and provisional inventory plan quantity for each of the plurality of demand forecast patterns calculated by the inventory management simulation unit in Embodiment 1. This figure shows an example of a plurality of provisional PSI plan information created by the provisional PSI plan creation unit in Embodiment 1. This figure shows an example of optimal provisional PSI plan information selected by the provisional PSI plan selection unit in Embodiment 1. This figure shows an example of difference information presented by the difference presentation unit in Embodiment 1. This is a first flowchart for explaining the information processing by the information processing device in Embodiment 1 of the present disclosure. This is a second flowchart for explaining the information processing by the information processing device in Embodiment 1 of the present disclosure. This is a flowchart for explaining the demand forecast pattern creation process in step S5 of Figure 9. This figure shows the configuration of the information processing device according to Embodiment 2. This is a first flowchart illustrating the information processing by the information processing apparatus in Embodiment 2 of the present disclosure. This is a second flowchart illustrating the information processing by the information processing apparatus in Embodiment 2 of the present disclosure.

[0010] (Knowledge forming the basis of this disclosure) Conventionally, highly accurate inventory management is required in order to respond quickly to customer demand while minimizing inventory costs. For example, in Patent Document 1, inventory quantities are managed by adjusting incoming and outgoing quantities using upper and lower limit quantity reference values ​​that are set in advance as appropriate inventory quantities. In Patent Document 2, a PSI planning editing table is created in which the date and the inventory quantity indicated by the lead time and PSI information corresponding to the date are displayed in cells, and the values ​​in the cells are calculated according to the calculated value in which the lead time is reflected in the time series.

[0011] In conventional inventory management systems, the PSI planning and inventory management functions operate independently, meaning that production plans and actual inventory levels are not linked, which can lead to problems such as stockouts or excess inventory.

[0012] To address the above challenges, the following technologies are disclosed.

[0013] (1) An information processing method relating to one aspect of the present disclosure is an information processing method performed by a computer, comprising: acquiring current and past demand performance for a product; creating a plurality of demand forecast patterns that show daily demand in a first future period based on the demand performance; calculating daily production plan quantities, sales plan quantities, and inventory plan quantities for each of the plurality of demand forecast patterns using a dynamic buffer management algorithm; creating a plurality of PSI (production / sales / inventory) plan information for each of the plurality of demand forecast patterns, including monthly production plan quantities, sales plan quantities, and inventory plan quantities that aggregate the daily production plan quantities, sales plan quantities, and inventory plan quantities on a monthly basis; and selecting one PSI plan information from the plurality of PSI plan information based on at least one of cost, the number of times inventory has run out, and the number of times inventory has been in excess.

[0014] With this configuration, the inventory level simulation results from the dynamic buffer management algorithm are reflected in the PSI plan, allowing the PSI plan to be dynamically updated and production plans to be synchronized and adjusted with actual inventory levels. This reduces the risk of stockouts and excess inventory, resulting in overall supply chain efficiency and cost reduction.

[0015] (2) The information processing method described in (1) above may further include obtaining PSI plan information created by the user, calculating the difference between the selected PSI plan information and the created PSI plan information, and presenting the user with difference information indicating the difference.

[0016] With this configuration, the difference between the PSI plan information, which reflects the inventory level simulation results from the dynamic buffer management algorithm, and the PSI plan information created by the user is calculated, and difference information showing the calculated difference is presented, so that the user can be given information to make decisions to correct the PSI plan information they have created.

[0017] (3) The information processing method described in (2) above may further include receiving an update instruction from the user to update the created PSI plan information to the selected PSI plan information, and updating the created PSI plan information to the selected PSI plan information when the user input of the update instruction is received.

[0018] In this configuration, when a user inputs an update instruction, the PSI plan information created by the user is updated to reflect the inventory level simulation results from the dynamic buffer management algorithm. Therefore, the user can decide whether or not to update their own PSI plan information, taking into account future business plans that are not reflected in the PSI plan information created by the computer.

[0019] (4) The information processing method described in (1) above may further include obtaining PSI plan information created by the user, calculating the difference between the selected PSI plan information and the created PSI plan information, and updating the created PSI plan information to the selected PSI plan information based on the difference.

[0020] This configuration allows the PSI planning information created by the user to be automatically updated to reflect the inventory level simulation results from the dynamic buffer management algorithm.

[0021] (5) In the information processing method described in (4) above, the updating of the created PSI plan information may include updating the created PSI plan information to the selected PSI plan information if the difference in production plan quantity, the difference in sales plan quantity, and the difference in inventory plan quantity are all within a predetermined range.

[0022] With this configuration, when the difference is small, the PSI plan information created by the user can be automatically updated to PSI plan information that reflects the inventory level simulation results from the dynamic buffer management algorithm.

[0023] (6) In the information processing method described in any one of (1) to (5) above, the creation of the plurality of demand forecast patterns may include: extracting a reference demand pattern from the actual demand data showing the daily demand for the period from the present to the second period prior; identifying a plurality of third periods from the actual demand data having a plurality of first demand patterns similar to the reference demand pattern; extracting a plurality of second demand patterns showing the daily demand for the period from immediately after each of the plurality of third periods to the second period ahead; creating a plurality of demand distribution curves that approximate the distribution of a plurality of demands by curve fitting based on the plurality of second demand patterns; creating a plurality of third demand patterns for the period from the present to the second period ahead by sampling each of the plurality of demand distribution curves on a daily basis; and creating a plurality of daily demand forecast patterns for the first period in the future by repeatedly creating and connecting the plurality of third demand patterns for each of the second periods from the present to the first period ahead.

[0024] According to this configuration, multiple third periods are identified that have a first demand pattern similar to the base demand pattern, which shows the daily demand from the present to two periods prior to the current period. From multiple second demand patterns, which show the daily demand from immediately after each of the multiple third periods to two periods ahead, the daily demand from the present to two periods ahead is predicted, thereby improving the accuracy of demand forecasting.

[0025] (7) In the information processing method described in (6) above, the creation of the plurality of demand forecast patterns may include creating a demand occurrence pattern that indicates whether or not the demand occurred on a daily basis based on the actual demand, and multiplying the demand occurrence pattern by each of the plurality of third demand patterns.

[0026] Demand does not necessarily occur every day; there are days when no demand occurs. Since the demand distribution curve approximates the distribution of multiple demands, it does not reflect days when no demand occurs. However, with this configuration, a demand occurrence pattern is created that shows whether or not demand occurred on a daily basis based on actual demand data. This demand occurrence pattern is then multiplied by each of the multiple third demand patterns, so that the demand on days when no demand occurred can be reflected in the multiple third demand patterns.

[0027] (8) In the information processing method described in (6) or (7) above, the second period may be a period of the same length as the production lead time for producing the product.

[0028] With this configuration, the second period is the same length as the production lead time for producing the product, so demand can be forecasted taking the production lead time into account.

[0029] (9) In the information processing method described in any one of (1) to (8) above, the selection of PSI plan information in 1 may include selecting from the plurality of PSI plan information the PSI plan information in which at least one of the cost, the number of times inventory was out of stock, and the number of times inventory was in excess is minimized.

[0030] This configuration allows for the selection of the most optimal PSI plan information from among multiple PSI plan information sources.

[0031] Furthermore, this disclosure can be implemented not only as an information processing method that performs the characteristic processing described above, but also as an information processing device having a characteristic configuration corresponding to the characteristic processing performed by the information processing method. It can also be implemented as a computer program that causes a computer to execute the characteristic processing included in such an information processing method. Therefore, the same effects as the above-described information processing method can be achieved in the following other embodiments.

[0032] (10) An information processing device according to another aspect of the present disclosure is an information processing device comprising a processor, the processor acquiring current and past demand performance for a product, creating a plurality of demand forecast patterns showing daily demand in a first future period based on the demand performance, calculating daily production plan quantities, sales plan quantities and inventory plan quantities for each of the plurality of demand forecast patterns using a dynamic buffer management algorithm, creating a plurality of PSI (production / sales / inventory) plan information for each of the plurality of demand forecast patterns, including monthly production plan quantities, sales plan quantities and inventory plan quantities obtained by aggregating the daily production plan quantities, sales plan quantities and inventory plan quantities on a monthly basis, and selecting one PSI plan information from the plurality of PSI plan information based on at least one of cost, the number of times inventory was out of stock, and the number of times inventory was in excess.

[0033] (11) Information processing programs according to other embodiments of the present disclosure acquire current and past demand performance for a product, create a plurality of demand forecast patterns showing daily demand in a first future period based on the demand performance, calculate daily production plan quantities, sales plan quantities and inventory plan quantities for each of the plurality of demand forecast patterns using a dynamic buffer management algorithm, create a plurality of PSI (production / sales / inventory) plan information for each of the plurality of demand forecast patterns, including monthly production plan quantities, sales plan quantities and inventory plan quantities obtained by aggregating the daily production plan quantities, sales plan quantities and inventory plan quantities on a monthly basis, and cause the computer to function to select one PSI plan information from the plurality of PSI plan information based on at least one of cost, the number of times inventory was out of stock, and the number of times inventory was in excess.

[0034] (12) Non-temporary computer-readable recording media relating to other aspects of the present disclosure record the information processing programs described in (11) above.

[0035] Embodiments of this disclosure will be described below with reference to the attached drawings. Note that the embodiments described below are all specific examples of this disclosure. The numerical values, shapes, components, steps, and order of steps shown in the following embodiments are examples only and are not intended to limit this disclosure. Furthermore, components in the following embodiments that are not described in the independent claim representing the highest-level concept will be described as optional components. Also, in all embodiments, the contents of each can be combined.

[0036] (Embodiment 1) Figure 1 is a diagram showing the configuration of the information processing device 1 according to this embodiment 1.

[0037] The information processing device 1 includes an input unit 101, a PSI plan creation unit 102, a PSI plan acquisition unit 103, a performance acquisition unit 104, a demand forecast pattern creation unit 105, an inventory management simulation unit 106, a provisional PSI plan creation unit 107, a provisional PSI plan selection unit 108, a difference calculation unit 109, a difference presentation unit 110, a display unit 111, a PSI plan update unit 112, a PSI plan storage unit 201, a product information storage unit 202, a demand performance storage unit 203, and a provisional PSI plan storage unit 204.

[0038] The information processing device 1 includes, for example, a computer system comprising a control program, a processing circuit such as a processor or logic circuit that executes the control program, and a recording device such as an internal memory or an accessible external memory that stores the control program. The information processing device 1 may be implemented, for example, by hardware implementation using a processing circuit, by execution of a software program held in memory by a processing circuit or distributed from an external server, or by a combination of these hardware and software implementations. The information processing device 1 may also be a server.

[0039] The PSI plan creation unit 102, PSI plan acquisition unit 103, actual data acquisition unit 104, demand forecast pattern creation unit 105, inventory management simulation unit 106, provisional PSI plan creation unit 107, provisional PSI plan selection unit 108, difference calculation unit 109, difference presentation unit 110, and PSI plan update unit 112 are implemented by a processor. In addition, the PSI plan storage unit 201, product information storage unit 202, demand actual data storage unit 203, and provisional PSI plan storage unit 204 are implemented by a recording device.

[0040] The input unit 101 is, for example, a keyboard, mouse, or touch panel, and accepts information input from the user. The input unit 101 may be connected to the information processing device 1 via a wired or wireless connection so that they can communicate with each other. Alternatively, the input unit 101 may 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.

[0041] The input unit 101 receives input by the user of PSI plan information including the production plan quantity, sales plan quantity, and inventory plan quantity of the product. The user inputs PSI plan information including the monthly production plan quantity, sales plan quantity, and inventory plan quantity of the product for the first period. The first period is, for example, one year or six months.

[0042] The PSI plan creation unit 102 creates PSI plan information including the monthly production plan quantity, sales plan quantity, and inventory plan quantity of the product for the future first period. The PSI plan creation unit 102 creates PSI plan information for each product. The PSI plan creation unit 102 stores the created PSI plan information in the PSI plan storage unit 201.

[0043] The PSI plan storage unit 201 stores the PSI plan information for each product created by the PSI plan creation unit 102.

[0044] FIG. 2 is a diagram showing an example of PSI plan information created by the PSI plan creation unit 102 in the first embodiment.

[0045] The PSI plan information shown in FIG. 2 includes a product ID for identifying the product and the monthly production plan quantity, sales plan quantity, and inventory plan quantity for the future first period.

[0046] The PSI plan acquisition unit 103 acquires the PSI plan information created by the user stored in the PSI plan storage unit 201. The PSI plan acquisition unit 103 outputs the acquired PSI plan information to the difference calculation unit 109.

[0047] The product information storage unit 202 stores product information related to the product in advance. The product information includes the production lead time of the product and the defective rate of the product. The production lead time represents the period from the start of manufacturing to the completion of manufacturing of the product. The defective rate represents the probability that a defective product occurs. Note that the product information may include the order lead time of the product. The order lead time represents the period from placing an order for the product to receiving the delivery.

[0048] The demand performance storage unit 203 stores the past demand performance of the product. The demand performance storage unit 203 stores the past daily demand performance.

[0049] The performance acquisition unit 104 acquires past demand data for the product. The performance acquisition unit 104 acquires daily demand data for the product during a past first period from the demand data storage unit 203. The first period is, for example, one year. The performance acquisition unit 104 acquires daily demand data from the previous day to one year ago.

[0050] Figure 3 shows an example of demand data acquired by the performance data acquisition unit 104 in this embodiment 1.

[0051] The demand data shown in Figure 3 includes a product ID to identify the product and daily demand for the past first period.

[0052] Furthermore, the performance acquisition unit 104 acquires current demand performance data. The input unit 101 accepts user input of current demand performance data. Current demand performance data refers to the demand for the current day. The performance acquisition unit 104 acquires the current demand performance data entered by the input unit 101.

[0053] Furthermore, the performance acquisition unit 104 acquires product information, including the product's production lead time and the product's defect rate, from the product information storage unit 202.

[0054] The demand forecast pattern creation unit 105 creates multiple demand forecast patterns that show the daily demand for the first period in the future, based on the actual demand acquired by the actual demand acquisition unit 104.

[0055] Specifically, the demand forecast pattern creation unit 105 extracts a base demand pattern from actual demand data that shows the daily demand for the period from the present to the second period prior. The second period is a period of the same length as the production lead time for producing the product. The second period may also be a period of the same length as the order lead time. The demand forecast pattern creation unit 105 identifies multiple third periods from actual demand data that have multiple first demand patterns similar to the base demand pattern. The demand forecast pattern creation unit 105 extracts multiple second demand patterns that show the daily demand for the period from immediately after each of the multiple third periods to the second period ahead. Based on the multiple second demand patterns, the demand forecast pattern creation unit 105 creates multiple demand distribution curves that approximate multiple demand distributions by curve fitting.

[0056] The demand forecast pattern creation unit 105 creates multiple third demand patterns for the period from the present to the second period ahead by sampling each of the multiple demand distribution curves on a daily basis. Based on actual demand, the demand forecast pattern creation unit 105 creates a demand occurrence pattern that indicates whether or not demand occurred on a daily basis. The demand forecast pattern creation unit 105 multiplies each of the multiple third demand patterns by the demand occurrence pattern. The demand forecast pattern creation unit 105 creates multiple daily demand forecast patterns for the future first period by repeatedly creating and connecting the multiple third demand patterns for each second period from the present to the first period ahead.

[0057] Figure 4 shows an example of a plurality of demand forecast patterns created by the demand forecast pattern creation unit 105 in this embodiment 1.

[0058] The demand forecast pattern shown in Figure 4 includes a product ID for identifying the product and the daily demand for the first future period. The demand forecast pattern creation unit 105 creates the first to Nth demand forecast patterns. The number of multiple demand forecast patterns is predetermined. The number of demand distribution curves created by curve fitting is equal to the number of demand forecast patterns created.

[0059] The inventory management simulation unit 106 calculates the daily provisional production plan quantity, provisional sales plan quantity, and provisional inventory plan quantity for each of the multiple demand forecast patterns created by the demand forecast pattern creation unit 105 using a dynamic buffer management algorithm.

[0060] The dynamic buffer management algorithm is a known technique. Here's a brief explanation of the dynamic buffer management algorithm. In the dynamic buffer management algorithm, the buffer is divided into three equal parts, from top to bottom: the green zone, the yellow zone, and the red zone. Then, it is determined which of the three zones the current inventory level is in. If the inventory level is in the green or red zone, the number of consecutive occurrences in each zone is counted. If the number of consecutive occurrences in the green zone reaches a threshold, the target inventory level, which is the buffer's upper limit, is decreased. If the number of consecutive occurrences in the red zone reaches a threshold, the target inventory level is increased. If the current inventory level is less than the target inventory level, the difference between the current inventory level and the target inventory level becomes the production quantity.

[0061] The initial value of the target inventory quantity, the threshold number of consecutive times to increase or decrease the target inventory quantity, and the amount by which the target inventory quantity is increased or decreased are all pre-set.

[0062] Figure 5 shows an example of the daily provisional production plan quantity, provisional sales plan quantity, and provisional inventory plan quantity for each of the multiple demand forecast patterns calculated by the inventory management simulation unit 106 in this embodiment 1.

[0063] As shown in Figure 5, the inventory management simulation unit 106 calculates the daily provisional production plan, provisional sales plan, and provisional inventory plan for the first period in the future for each of the multiple demand forecast patterns. The inventory amount is expressed as the previous day's inventory amount + incoming quantity - outgoing quantity. Outgoing quantity is demand. Incoming quantity is the shortfall to reach the target inventory amount. The target inventory amount is determined by the dynamic buffer management algorithm. The daily provisional production plan is the daily incoming quantity, the daily provisional sales plan is the daily demand, and the daily provisional inventory plan is the daily inventory amount.

[0064] The inventory management simulation unit 106 may also reflect the defect rate of the products obtained by the performance acquisition unit 104 in the provisional production plan quantity. For example, if the defect rate is 1 / 100 and 1000 units of the product are to be produced, the inventory management simulation unit 106 may set the provisional production plan quantity to 1010 units.

[0065] Furthermore, the inventory management simulation unit 106 calculates costs, the number of times inventory was out of stock (in days), and the number of times inventory was in excess (in days) for each of several demand forecast patterns. The inventory management simulation unit 106 calculates costs as the sum of the production costs required to produce the products, the storage costs required to store the products, the losses incurred due to product shortages, and the overtime pay incurred due to exceeding the daily production capacity. If the provisional sales plan quantity exceeds the provisional inventory plan quantity, the inventory management simulation unit 106 increments the number of times inventory was out of stock. Also, if the provisional inventory plan quantity exceeds a threshold, the inventory management simulation unit 106 increments the number of times inventory was in excess.

[0066] The provisional PSI planning unit 107 creates multiple provisional PSI (production / sales / inventory) plan information for each of the multiple demand forecast patterns, including monthly provisional production plans, provisional sales plans, and provisional inventory plans, which are aggregated monthly from the daily provisional production plans, provisional sales plans, and provisional inventory plans created by the inventory management simulation unit 106. The provisional PSI planning unit 107 stores the created provisional PSI plan information in the provisional PSI plan storage unit 204. The inventory management simulation unit 106 also stores the calculated costs, the number of times inventory ran out, and the number of times inventory was in excess, associating them with the multiple provisional PSI plan information corresponding to the multiple demand forecast patterns, and stores them in the provisional PSI plan storage unit 204.

[0067] The provisional PSI plan storage unit 204 stores multiple provisional PSI plan information created by the provisional PSI plan creation unit 107. The provisional PSI plan storage unit 204 also stores the cost calculated by the inventory management simulation unit 106, the number of times inventory was out of stock, and the number of times inventory was in excess, in association with the multiple provisional PSI plan information.

[0068] Figure 6 shows an example of multiple provisional PSI plan information created by the provisional PSI plan creation unit 107 in this embodiment 1.

[0069] The provisional PSI plan information shown in Figure 6 includes a product ID for identifying the product, and provisional production plan quantities, provisional sales plan quantities, and provisional inventory plan quantities for the first period in the future on a monthly basis. The provisional PSI plan creation unit 107 creates the first to Nth provisional PSI plan information. The provisional PSI plan creation unit 107 calculates the monthly provisional production plan quantities, provisional sales plan quantities, and provisional inventory plan quantities by aggregating the daily provisional production plan quantities, provisional sales plan quantities, and provisional inventory plan quantities calculated by the inventory management simulation unit 106 on a monthly basis. The provisional PSI plan creation unit 107 creates multiple provisional PSI plan information sets including the calculated monthly provisional production plan quantities, provisional sales plan quantities, and provisional inventory plan quantities.

[0070] The provisional PSI plan selection unit 108 selects one provisional PSI plan from among multiple provisional PSI plan information based on at least one of the following: cost, number of times inventory was out of stock, and number of times inventory was in excess. From among the multiple provisional PSI plan information, the provisional PSI plan selection unit 108 selects one provisional PSI plan information that minimizes at least one of the following: cost, number of times inventory was out of stock, and number of times inventory was in excess.

[0071] Figure 7 shows an example of the optimal provisional PSI plan information selected by the provisional PSI plan selection unit 108 in this embodiment 1.

[0072] The provisional PSI planning information shown in Figure 7 includes a product ID to identify the product, and provisional production plans, sales plans, and inventory plans for the first period in the future on a monthly basis.

[0073] The provisional PSI plan selection unit 108 may select one provisional PSI plan from among multiple provisional PSI plan information that minimizes all of the following: cost, number of times inventory was out of stock, and number of times inventory was in excess. If there is no provisional PSI plan information that minimizes all of the following: cost, number of times inventory was out of stock, and number of times inventory was in excess, the provisional PSI plan selection unit 108 may select one provisional PSI plan information according to the priority of cost, number of times inventory was out of stock, and number of times inventory was in excess.

[0074] The priorities for cost, the number of times inventory ran out, and the number of times inventory became excessive may be set in advance. If cost has the highest priority, the provisional PSI plan selection unit 108 may select one provisional PSI plan from among multiple provisional PSI plan information that has the minimum cost. If the number of times inventory ran out has the highest priority, the provisional PSI plan selection unit 108 may select one provisional PSI plan from among multiple provisional PSI plan information that has the minimum number of times inventory ran out. If the number of times inventory became excessive has the highest priority, the provisional PSI plan selection unit 108 may select one provisional PSI plan from among multiple provisional PSI plan information that has the minimum number of times inventory became excessive.

[0075] The difference calculation unit 109 calculates the difference between the provisional PSI plan information selected by the provisional PSI plan selection unit 108 and the user-created PSI plan information acquired by the PSI plan acquisition unit 103.

[0076] The difference presentation unit 110 presents the user with difference information showing the difference calculated by the difference calculation unit 109. The difference presentation unit 110 outputs the difference information showing the difference calculated by the difference calculation unit 109 to the display unit 111.

[0077] The display unit 111 is, for example, a liquid crystal display device that displays information to be presented to the user. The display unit 111 may be connected to the information processing device 1 via wired or wireless means so as to be able to communicate with each other. Alternatively, the display unit 111 may be connected to the information processing device 1 via a network so as to be able to communicate with each other. The network may be a local area network or a wide area network.

[0078] The display unit 111 displays the difference information output by the difference presentation unit 110. The user can check the difference between the provisional PSI plan information created by the information processing device 1 and the PSI plan information created by the user. The display unit 111 may also display the provisional PSI plan information created by the information processing device 1 and the PSI plan information created by the user, along with the difference information.

[0079] Figure 8 shows an example of difference information presented by the difference presentation unit 110 in this embodiment 1.

[0080] The difference calculation unit 109 calculates, on a monthly basis, the difference in production plan quantity obtained by subtracting the production plan quantity from the provisional production plan quantity in the provisional PSI plan information, the difference in sales plan quantity obtained by subtracting the sales plan quantity from the provisional sales plan quantity in the provisional PSI plan information, and the difference in inventory plan quantity obtained by subtracting the inventory plan quantity from the provisional inventory plan quantity in the provisional PSI plan information.

[0081] The difference information shown in Figure 8 includes a product ID for identifying the product, and the difference values ​​for the monthly production plan quantity, sales plan quantity, and inventory plan quantity for the first future period.

[0082] For example, the provisional inventory plan quantity for June 2024 in the provisional PSI planning information shown in Figure 7 is 400 units, while the inventory plan quantity for June 2024 in the PSI planning information shown in Figure 2 is 600 units. In this case, the difference in the inventory plan quantity for June 2024 shown in Figure 8 is -200 units.

[0083] The input unit 101 receives an update instruction from the user to update the PSI plan information created by the user with the provisional PSI plan information selected by the provisional PSI plan selection unit 108. The user checks the presented difference and decides whether or not to update the PSI plan information created by the user with the provisional PSI plan information selected by the provisional PSI plan selection unit 108. The display unit 111 may display an update button to update the PSI plan information created by the user with the provisional PSI plan information selected by the provisional PSI plan selection unit 108. The input unit 101 may accept the user pressing the update button.

[0084] When a user input for an update instruction is received, the PSI plan update unit 112 updates the PSI plan information created by the user stored in the PSI plan storage unit 201 with the temporary PSI plan information selected by the temporary PSI plan selection unit 108. When the update button is pressed, the PSI plan update unit 112 updates the PSI plan information stored in the PSI plan storage unit 201 with the temporary PSI plan information selected by the temporary PSI plan selection unit 108.

[0085] Next, the information processing by the information processing device 1 in Embodiment 1 of this disclosure will be described.

[0086] Figure 9 is a first flowchart illustrating the information processing by the information processing device 1 in Embodiment 1 of the present disclosure, and Figure 10 is a second flowchart illustrating the information processing by the information processing device 1 in Embodiment 1 of the present disclosure.

[0087] The information processing shown in Figures 9 and 10 is performed, for example, once a day.

[0088] First, in step S1, the PSI planning unit 102 creates PSI planning information including the monthly production plan, sales plan, and inventory plan for the first future period of the product. The input unit 101 receives user input of the monthly production plan, sales plan, and inventory plan for the first future period of the product.

[0089] Next, in step S2, the PSI plan creation unit 102 stores the created PSI plan information in the PSI plan storage unit 201.

[0090] Next, in step S3, the PSI plan acquisition unit 103 acquires the PSI plan information stored in the PSI plan storage unit 201.

[0091] Next, in step S4, the performance acquisition unit 104 acquires the product's past demand performance, current demand performance, and product information. The performance acquisition unit 104 acquires the product's daily demand performance for the past first period from the demand performance storage unit 203. The performance acquisition unit 104 also acquires the product's current demand performance entered by the input unit 101. The performance acquisition unit 104 also acquires the product's product information from the product information storage unit 202.

[0092] Next, in step S5, the demand forecast pattern creation unit 105 creates multiple demand forecast patterns that show the daily demand for the first period in the future, based on past demand data, current demand data, and product information acquired by the data acquisition unit 104.

[0093] Here, we will explain the demand forecast pattern creation process performed by the demand forecast pattern creation unit 105.

[0094] Figure 11 is a flowchart illustrating the demand forecast pattern creation process in step S5 of Figure 9.

[0095] First, in step S21, the demand forecast pattern creation unit 105 extracts a standard demand pattern from current and past demand data, showing the daily demand for the period from the present to the period before the second period. The second period is a period of the same length as the production lead time included in the product information.

[0096] Next, in step S22, the demand forecast pattern creation unit 105 identifies a plurality of third periods from the actual demand data that have a plurality of first demand patterns similar to the standard demand pattern. The number of each demand included in the first demand pattern may be the same as the number of each demand included in the standard demand pattern. Also, the time series trajectory of each demand included in the first demand pattern may be the same as the time series trajectory of each demand included in the standard demand pattern.

[0097] Next, in step S23, the demand forecast pattern creation unit 105 extracts multiple second demand patterns that show the daily demand for the period from immediately after each of the multiple third periods to the second period ahead.

[0098] Next, in step S24, the demand forecast pattern creation unit 105 creates multiple demand distribution curves that approximate the distribution of multiple demands by curve fitting based on the multiple second demand patterns. The demand forecast pattern creation unit 105 plots the multiple demand values ​​included in each of the multiple second demand patterns on a coordinate system where the horizontal axis represents time and the vertical axis represents quantity, and creates multiple demand distribution curves by curve fitting. The number of multiple demand distribution curves to be created is predetermined.

[0099] Next, in step S25, the demand forecast pattern creation unit 105 creates multiple third demand patterns for the period from the present to the second period ahead by sampling each of the multiple demand distribution curves on a daily basis. The demand forecast pattern creation unit 105 samples multiple daily demands from each of the multiple demand distribution curves and creates multiple third demand patterns for the period from the present to the second period ahead using the sampled multiple daily demands.

[0100] Next, in step S26, the demand forecast pattern creation unit 105 creates a demand occurrence pattern that indicates whether or not demand occurred on a daily basis, based on actual demand. That is, demand does not necessarily occur every day, and there are days when it does not occur. The above demand distribution curve is a curve that approximates the distribution of multiple demands, so it does not reflect days when no demand occurs. Therefore, the demand forecast pattern creation unit 105 calculates the probability of demand occurring in the period from immediately after each of the multiple third periods to the second period after, using a Bernoulli distribution. The demand forecast pattern creation unit 105 creates a demand occurrence pattern by designating days when the probability of demand occurring is higher than a threshold as days when demand occurs, and days when the probability of demand occurring is below the threshold as days when no demand occurs. The demand forecast pattern creation unit 105 creates a demand occurrence pattern in which the demand on days when demand occurs is set to 1 and the demand on days when no demand occurs is set to 0.

[0101] Next, in step S27, the demand forecast pattern creation unit 105 multiplies the demand generation pattern by each of the multiple third demand patterns. As a result, among the multiple demands included in the third demand pattern, the demand for days on which no demand occurs becomes 0.

[0102] Next, in step S28, the demand forecast pattern creation unit 105 connects each of the multiple third demand patterns created this time to each of the multiple third demand patterns created in previous steps. That is, multiple third demand patterns are created for each second period, and by sequentially connecting each of these multiple third demand patterns, multiple daily demand forecast patterns for future first periods are created.

[0103] Next, in step S29, the demand forecast pattern creation unit 105 determines whether multiple demand forecast patterns for the future first period have been created. If it is determined that multiple demand forecast patterns for the future first period have been created (YES in step S29), the demand forecast pattern creation process ends.

[0104] On the other hand, if it is determined that no multiple demand forecast patterns have been created for the future first period (NO in step S29), in step S30, the demand forecast pattern creation unit 105 creates an average demand pattern by averaging the daily demand for the period from immediately following each of the multiple third periods to the second period ahead. The process then moves to step S22, where the demand forecast pattern creation unit 105 identifies multiple third periods that have multiple first demand patterns similar to the average demand pattern among the actual demand.

[0105] Returning to Figure 9, in step S6, the inventory management simulation unit 106 selects one of the multiple demand forecast patterns created by the demand forecast pattern creation unit 105.

[0106] Next, in step S7, the inventory management simulation unit 106 calculates the daily provisional production plan quantity, provisional sales plan quantity, and provisional inventory plan quantity using a dynamic buffer management algorithm based on the selected demand forecast pattern.

[0107] Next, in step S8, the provisional PSI plan creation unit 107 creates provisional PSI plan information including monthly provisional production plan quantities, provisional sales plan quantities, and provisional inventory plan quantities by aggregating the daily provisional production plan quantities, provisional sales plan quantities, and provisional inventory plan quantities created by the inventory management simulation unit 106 on a monthly basis.

[0108] Next, in step S9, the provisional PSI plan creation unit 107 stores the created provisional PSI plan information in the provisional PSI plan storage unit 204.

[0109] Next, in step S10, the inventory management simulation unit 106 calculates the cost, the number of times inventory ran out, and the number of times inventory was in excess, based on the selected demand forecast pattern.

[0110] Next, in step S11, the inventory management simulation unit 106 stores the calculated costs, the number of times inventory was out of stock, and the number of times inventory was in excess in the provisional PSI plan storage unit 204, associating them with provisional PSI plan information corresponding to one demand forecast pattern.

[0111] Next, in step S12, the inventory management simulation unit 106 determines whether or not provisional PSI planning information has been created for all demand forecast patterns. If it is determined that provisional PSI planning information has not been created for all demand forecast patterns (NO in step S12), the process returns to step S6. Then, in step S6, the inventory management simulation unit 106 selects one unselected demand forecast pattern from among the multiple demand forecast patterns created by the demand forecast pattern creation unit 105.

[0112] On the other hand, if it is determined that provisional PSI plan information has been created for all demand forecast patterns (YES in step S12), in step S13, the provisional PSI plan selection unit 108 selects one provisional PSI plan from among the multiple provisional PSI plan information stored in the provisional PSI plan storage unit 204 based on at least one of the following: cost, the number of times inventory was out of stock, and the number of times inventory was in excess.

[0113] Specifically, the provisional PSI plan selection unit 108 selects one provisional PSI plan from among multiple provisional PSI plan information that has the minimum cost, the number of times inventory was out of stock, and the number of times inventory was in excess. If there is no provisional PSI plan information that has the minimum cost, the number of times inventory was out of stock, and the number of times inventory was in excess, the provisional PSI plan selection unit 108 may select one provisional PSI plan information according to the priority of cost, the number of times inventory was out of stock, and the number of times inventory was in excess. The priority of cost, the number of times inventory was out of stock, and the number of times inventory was in excess is set in advance.

[0114] Next, in step S14, the difference calculation unit 109 calculates the difference between the provisional PSI plan information selected by the provisional PSI plan selection unit 108 and the PSI plan information created by the user and acquired by the PSI plan acquisition unit 103.

[0115] Next, in step S15, the difference presentation unit 110 displays the difference calculated by the difference calculation unit 109 on the display unit 111. At this time, the difference presentation unit 110 outputs the difference calculated by the difference calculation unit 109 to the display unit 111. The display unit 111 displays the difference output by the difference presentation unit 110.

[0116] Next, in step S16, the input unit 101 receives an update instruction from the user to update the PSI plan information created by the user with the provisional PSI plan information selected by the provisional PSI plan selection unit 108.

[0117] Next, in step S17, the PSI plan update unit 112 determines whether or not the user has entered an update instruction. If it is determined that the user has not entered an update instruction (NO in step S17), the process ends.

[0118] On the other hand, if it is determined that the user has entered an update instruction (YES in step S17), in step S18, the PSI plan update unit 112 updates the PSI plan information created by the user stored in the PSI plan storage unit 201 with the temporary PSI plan information selected by the temporary PSI plan selection unit 108.

[0119] As described above, according to this embodiment 1, the simulation results of inventory levels by the dynamic buffer management algorithm are reflected in the PSI plan, allowing the PSI plan to be dynamically updated and production plans and actual inventory levels to be synchronized and adjusted. This reduces the risk of stockouts and excess inventory, thereby achieving overall supply chain efficiency and cost reduction.

[0120] For example, if sales volume is expected to increase in the future due to a sales promotion campaign, production volume needs to be increased in advance. Since the dynamic buffer management algorithm does not take into account plans to increase production volume, the user who created the PSI plan information must decide whether or not to update the current PSI plan information. By presenting the difference to the user, the system can provide the user with the information needed to decide whether or not to update the current PSI plan information.

[0121] Note that the processes in steps S1 and S2 and the processes in steps S3 to S18 do not need to be performed consecutively in terms of time. Also, after the user has created the PSI plan information and the created PSI plan information has been stored in the PSI plan storage unit 201, the processes in steps S1 and S2 may be skipped and the process in step S3 may begin. Furthermore, if the PSI plan information stored in the PSI plan storage unit 201 is updated, the PSI plan acquisition unit 103 may acquire the updated PSI plan information from the PSI plan update unit 112.

[0122] Furthermore, a first information processing device managed by a business unit that sells products may include an input unit 101, a PSI plan creation unit 102, a difference presentation unit 110, a display unit 111, a PSI plan update unit 112, and a PSI plan storage unit 201, while a second information processing device managed by a factory that produces products may include a PSI plan acquisition unit 103, a performance acquisition unit 104, a demand forecast pattern creation unit 105, an inventory management simulation unit 106, a provisional PSI plan creation unit 107, a provisional PSI plan selection unit 108, a difference calculation unit 109, a product information storage unit 202, a demand performance storage unit 203, and a provisional PSI plan storage unit 204.

[0123] Furthermore, a first information processing device managed by a business unit that sells products may include an input unit 101, a PSI plan creation unit 102, a PSI plan acquisition unit 103, a performance acquisition unit 104, a demand forecast pattern creation unit 105, an inventory management simulation unit 106, a provisional PSI plan creation unit 107, a provisional PSI plan selection unit 108, a difference calculation unit 109, a difference presentation unit 110, a display unit 111, a PSI plan update unit 112, a PSI plan storage unit 201, and a provisional PSI plan storage unit 204, while a second information processing device managed by a factory that produces products may include a product information storage unit 202 and a demand performance storage unit 203.

[0124] (Embodiment 2) In Embodiment 1, the difference between the selected provisional PSI plan information and the created PSI plan information is presented to the user, and when the user inputs an update instruction to update the created PSI plan information to the selected provisional PSI plan information, the created PSI plan information is updated to the selected provisional PSI plan information. In contrast, in Embodiment 2, the created PSI plan information is updated to the selected provisional PSI plan information based on the difference between the selected provisional PSI plan information and the created PSI plan information.

[0125] Figure 12 shows the configuration of the information processing device 1A according to this second embodiment.

[0126] The information processing device 1A includes an input unit 101, a PSI plan creation unit 102, a PSI plan acquisition unit 103, a performance acquisition unit 104, a demand forecast pattern creation unit 105, an inventory management simulation unit 106, a provisional PSI plan creation unit 107, a provisional PSI plan selection unit 108, a difference calculation unit 109, a PSI plan update unit 112A, a difference determination unit 113, a PSI plan storage unit 201, a product information storage unit 202, a demand performance storage unit 203, and a provisional PSI plan storage unit 204.

[0127] The information processing device 1A comprises, for example, a computer system including a control program, a processing circuit such as a processor or logic circuit that executes the control program, and a recording device such as an internal memory or an accessible external memory that stores the control program. The information processing device 1A may be implemented, for example, by hardware implementation using a processing circuit, by execution of a software program held in memory by a processing circuit or distributed from an external server, or by a combination of these hardware and software implementations. The information processing device 1A may also be a server.

[0128] The PSI plan creation unit 102, PSI plan acquisition unit 103, actual data acquisition unit 104, demand forecast pattern creation unit 105, inventory management simulation unit 106, provisional PSI plan creation unit 107, provisional PSI plan selection unit 108, difference calculation unit 109, PSI plan update unit 112A, and difference determination unit 113 are implemented by a processor. In addition, the PSI plan storage unit 201, product information storage unit 202, demand actual data storage unit 203, and provisional PSI plan storage unit 204 are implemented by a recording device.

[0129] In this second embodiment, the same reference numerals are used for components that are the same as those in the first embodiment, and their descriptions are omitted.

[0130] The difference determination unit 113 determines whether the difference values ​​of the monthly production plan quantity, the monthly sales plan quantity, and the monthly inventory plan quantity are within a predetermined range. The predetermined range is, for example, between -200 units and +200 units.

[0131] The PSI plan update unit 112A updates the created PSI plan information with the selected provisional PSI plan information based on the difference calculated by the difference calculation unit 109. The PSI plan update unit 112A updates the created PSI plan information with the selected provisional PSI plan information if the difference values ​​for monthly production plan quantities, monthly sales plan quantities, and monthly inventory plan quantities are all within a predetermined range.

[0132] Next, the information processing by the information processing device 1A in Embodiment 2 of this disclosure will be described.

[0133] Figure 13 is a first flowchart illustrating the information processing by the information processing device 1A in Embodiment 2 of the present disclosure, and Figure 14 is a second flowchart illustrating the information processing by the information processing device 1A in Embodiment 2 of the present disclosure.

[0134] The processes in steps S41 to S54 are the same as the processes in steps S1 to S14 shown in Figures 9 and 10, so their explanation will be omitted.

[0135] Next, in step S55, the difference determination unit 113 determines whether all monthly difference values ​​are within a predetermined range. That is, the difference determination unit 113 determines whether each of the monthly difference values ​​for the planned production quantity, the monthly difference value for the planned sales quantity, and the monthly difference value for the planned inventory quantity are within a predetermined range.

[0136] If it is determined that all monthly difference values ​​are not within the predetermined range (NO in step S55), the process ends.

[0137] On the other hand, if it is determined that all monthly difference values ​​are within a predetermined range (YES in step S55), in step S56, the PSI plan update unit 112A updates the PSI plan information created by the user stored in the PSI plan storage unit 201 with the provisional PSI plan information selected by the provisional PSI plan selection unit 108.

[0138] Thus, according to this embodiment 2, the PSI plan information created by the user can be automatically updated to PSI plan information that reflects the inventory quantity simulation results by the dynamic buffer management algorithm.

[0139] Furthermore, some or all of the functions of the apparatus according to the embodiments of this disclosure may be realized by a processor such as a CPU executing a program.

[0140] Furthermore, all figures used above are illustrative examples provided to illustrate this disclosure, and this disclosure is not limited to these illustrative figures.

[0141] Furthermore, the order in which the steps shown in the flowchart above are performed is illustrative for the purpose of specifically illustrating this disclosure, and other orders are acceptable as long as similar effects are achieved. Also, some of the steps above may be performed simultaneously (in parallel) with other steps.

[0142] The technology disclosed herein can reduce the risk of stockouts and decrease excess inventory, thereby achieving overall supply chain efficiency and cost reduction. Therefore, it is useful as a technology for creating product PSI (Production / Sales / Inventory) planning information.

Claims

1. An information processing method performed by a computer, comprising: acquiring current and past demand performance for a product; creating a plurality of demand forecast patterns that show daily demand in a first future period based on the demand performance; calculating daily production plan quantities, sales plan quantities, and inventory plan quantities for each of the plurality of demand forecast patterns using a dynamic buffer management algorithm; creating a plurality of PSI (production / sales / inventory) plan information for each of the plurality of demand forecast patterns, including monthly production plan quantities, sales plan quantities, and inventory plan quantities obtained by aggregating the daily production plan quantities, sales plan quantities, and inventory plan quantities on a monthly basis; and selecting one PSI plan information from the plurality of PSI plan information based on at least one of cost, the number of times inventory was out of stock, and the number of times inventory was in excess.

2. The information processing method according to claim 1, further comprising: obtaining PSI plan information created by a user; calculating the difference between the selected PSI plan information and the created PSI plan information; and presenting the user with difference information indicating the difference.

3. The information processing method according to claim 2, further comprising: receiving input from the user of an update instruction to update the created PSI plan information with the selected PSI plan information; and, when the user input of the update instruction is received, updating the created PSI plan information with the selected PSI plan information.

4. The information processing method according to claim 1, further comprising: obtaining PSI plan information created by a user; calculating the difference between the selected PSI plan information and the created PSI plan information; and updating the created PSI plan information to the selected PSI plan information based on the difference.

5. The information processing method according to claim 4, wherein the update of the created PSI plan information includes updating the created PSI plan information to the selected PSI plan information if the difference in production plan quantity, the difference in sales plan quantity, and the difference in inventory plan quantity are all within a predetermined range.

6. The information processing method according to claim 1, wherein the creation of the plurality of demand forecast patterns includes: extracting a reference demand pattern from the actual demand data showing the daily demand for the period from the present to the period before the second period; identifying a plurality of third periods from the actual demand data having a plurality of first demand patterns similar to the reference demand pattern; extracting a plurality of second demand patterns showing the daily demand for the period from immediately after each of the plurality of third periods to the second period ahead; creating a plurality of demand distribution curves that approximate the distribution of a plurality of demands by curve fitting based on the plurality of second demand patterns; creating a plurality of third demand patterns for the period from the present to the second period ahead by sampling each of the plurality of demand distribution curves on a daily basis; and creating a plurality of daily demand forecast patterns for the future first period by repeatedly creating and connecting the plurality of third demand patterns for each of the second periods from the present to the first period ahead.

7. The information processing method according to claim 6, wherein the creation of the plurality of demand forecast patterns includes: creating a demand occurrence pattern that indicates whether or not the demand occurred on a daily basis based on the actual demand; and multiplying the demand occurrence pattern by each of the plurality of third demand patterns.

8. The information processing method according to claim 6, wherein the second period is a period of the same length as the production lead time for producing the product.

9. The information processing method according to claim 1, wherein the selection of the PSI plan information in 1 includes selecting the PSI plan information in 1 from among the plurality of PSI plan information such that at least one of the cost, the number of times the inventory ran out, and the number of times the inventory was in excess is minimized.

10. An information processing device comprising a processor, wherein the processor acquires current and past demand performance for a product, creates a plurality of demand forecast patterns showing daily demand in a first future period based on the demand performance, calculates daily production plan quantities, sales plan quantities, and inventory plan quantities for each of the plurality of demand forecast patterns using a dynamic buffer management algorithm, creates a plurality of PSI (production / sales / inventory) plan information for each of the plurality of demand forecast patterns, including monthly production plan quantities, sales plan quantities, and inventory plan quantities obtained by aggregating the daily production plan quantities, sales plan quantities, and inventory plan quantities on a monthly basis, and selects one PSI plan information from the plurality of PSI plan information based on at least one of cost, the number of times inventory was out of stock, and the number of times inventory was in excess.

11. An information processing program that obtains current and past demand data for a product, creates multiple demand forecast patterns showing daily demand for a first future period based on the demand data, calculates daily production plans, sales plans, and inventory plans for each of the multiple demand forecast patterns using a dynamic buffer management algorithm, creates multiple PSI (Production / Sales / Inventory) plan information for each of the multiple demand forecast patterns, including monthly production plans, sales plans, and inventory plans obtained by aggregating the daily production plans, sales plans, and inventory plans on a monthly basis, and causes a computer to function to select one PSI plan information from the multiple PSI plan information based on at least one of cost, the number of times inventory was out of stock, and the number of times inventory was in excess.

Images