Demand prediction device, demand prediction method, and demand prediction program

Abstract

A future inventory estimation unit (22) estimates the future inventory level of the supply destination by assuming that the inventory increases due to ordering in accordance with the ordering conditions and the ordering amount indicated by the ordering model information (61) and that the inventory decreases in accordance with the past inventory level indicated by the past inventory information (62) at the supply destination of the predicted object. The demand prediction unit (23) predicts that a demand for the order amount indicated by the order model information (61) is generated for the supply source at a point in time when the order condition indicated by the order model information (61) is satisfied in the future inventory amount estimated by the future inventory estimation unit (22).

Description

Technical Field

[0001] The present disclosure relates to techniques for performing demand forecasting. Background Art

[0002] In supply chains such as manufacturing and agriculture, a series of business transactions, from the procurement of raw materials to the sale of final products, are carried out by multiple operators. In these supply chains, methods for predicting demand for products are being studied to reduce inventory management costs.

[0003] Here, forecasting demand means predicting the future order quantity from the product supply destination. Below, the case of specifying forecast demand is described as demand forecast.

[0004] Conventional demand forecasting techniques use time-series data, such as past order performance, to learn from. Patent Document 1 describes a technique that improves demand forecast accuracy by learning past demand trends based on order performance and adjusting the demand forecasting method.

[0005] Prior art literature

[0006] Patent Literature

[0007] Patent Document 1: Japanese Patent Application Laid-Open No. 2022-101879 Summary of the Invention

[0008] Problems to be solved by the invention

[0009] The method described in Patent Document 1, which estimates and predicts the generation tendency of long-term demand such as seasonality or cyclicality based on order performance data, cannot perform demand forecasting that follows detailed daily demand fluctuations.

[0010] An object of the present disclosure is to enable demand forecasting that follows detailed daily demand fluctuations.

[0011] Means for solving problems

[0012] The demand forecasting device disclosed in the present invention comprises: a future inventory estimation unit, which assumes that at the supply destination of the forecast object, the inventory increases due to ordering according to the ordering conditions and order quantity indicated by the ordering model information, and the inventory decreases according to the progress of the past inventory quantity indicated by the past inventory information, thereby estimating the future inventory quantity of the supply destination; and a demand forecasting unit, which predicts that, under the future inventory quantity estimated by the future inventory estimation unit, at the point in time when the ordering conditions indicated by the ordering model information are met, a demand for the order quantity indicated by the ordering model information will be generated for the supply source.

[0013] Effects of the Invention

[0014] In the present disclosure, past inventory changes are used to estimate future inventory levels, and demand is predicted based on future inventory levels. This allows for demand prediction that tracks detailed daily demand fluctuations by taking recent inventory fluctuations into account. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is an explanatory diagram of the target of demand forecast in the first embodiment.

[0016] Figure 2 This is a configuration diagram of the demand forecasting system 100 according to the first embodiment.

[0017] Figure 3 This is a configuration diagram of the demand forecasting device 10 according to the first embodiment.

[0018] Figure 4 This is a flowchart of the input acceptance processing in the first embodiment.

[0019] Figure 5 This is an explanatory diagram of the order model information 61 according to the first embodiment.

[0020] Figure 6 This is an explanatory diagram of the past inventory information 62 according to the first embodiment.

[0021] Figure 7 This is a flowchart of the demand forecasting process in the first embodiment.

[0022] Figure 8 This is an explanatory diagram of the future inventory estimation process according to the first embodiment.

[0023] Figure 9 This is an explanatory diagram of the demand forecast result 64 according to the first embodiment.

[0024] Figure 10 This is a configuration diagram of the demand forecasting device 10 according to Modification 1.

[0025] Figure 11 This is a configuration diagram of the demand forecasting device 10 according to the second embodiment.

[0026] Figure 12 This is an explanatory diagram of the business flow structure information 65 according to the second embodiment.

[0027] Figure 13 This is a flowchart of the demand forecast processing in the second embodiment.

[0028] Figure 14 This is an explanatory diagram of the future inventory estimation process according to the second embodiment.

[0029] Figure 15 This is an explanatory diagram of the target of demand forecast in the second embodiment.

[0030] Figure 16 This is a configuration diagram of a demand forecasting device 10 according to a third embodiment.

[0031] Figure 17 This is an explanatory diagram of the logistics information 66 of the third embodiment.

[0032] Figure 18 This is a flowchart of the demand forecast processing in the third embodiment.

[0033] Figure 19 This is an explanatory diagram of the past inventory estimation process in the third embodiment.

[0034] Figure 20 This is a configuration diagram of a demand forecasting device 10 according to a fourth embodiment.

[0035] Figure 21 This is an explanatory diagram of the component configuration information 68 according to the fourth embodiment. DETAILED DESCRIPTION

[0036] Implementation Method 1

[0037] The demand forecast in the first embodiment is performed independently for each operator and each product. Figure 1 As shown, the item to be subject to demand forecasting, i.e., the forecast target item, is set as "Part A." Furthermore, the demand forecast for "Part A" generated from "Carrier A," the supply destination of "Part A," to "Carrier X," the supply source, is described as an example.

[0038] Here, when demand forecasting is performed for "Operator B" or "Component B", demand forecasting can also be performed by the same processing as when forecasting demand for "Component A" generated from "Operator A" for "Operator X".

[0039] The demand forecast in Embodiment 1 can be performed in unit periods, which are arbitrary time units such as days or weeks. Hereinafter, a case where the unit period is one day and the demand forecast is performed in daily units will be described as an example.

[0040] ***Description of the structure***

[0041] Reference Figure 2 The configuration of the demand forecast system 100 according to the first embodiment will be described.

[0042] The demand forecasting system 100 includes a demand forecasting device 10 , an input terminal 51 , and an output terminal 52 .

[0043] The demand forecasting device 10 is a computer such as a server, and receives order pattern information 61 and past inventory information 62 from an input terminal 51 , performs demand forecasting, and outputs a demand forecast result 64 to an output terminal 52 .

[0044] The input terminal 51 is a terminal for operating the demand forecasting device 10. Specifically, the input terminal 51 is a PC, tablet, or smartphone. PC stands for Personal Computer. The demand forecasting device 10 may also serve as the input terminal 51. Furthermore, there may be multiple input terminals 51.

[0045] The output terminal 52 is a terminal for outputting the processing details of the demand forecasting device 10 to the inventory planner. Specifically, the output terminal 52 is a PC, tablet computer, or smartphone. The input terminal 51 may also serve as the output terminal 52. Furthermore, there may be multiple output terminals 52.

[0046] Reference Figure 3 The configuration of the demand forecasting device 10 according to the first embodiment will be described.

[0047] The demand forecasting device 10 is a computer.

[0048] The demand forecasting device 10 includes hardware such as a processor 11, a memory 12, a storage 13, and a communication interface 14. The processor 11 is connected to the other hardware via a signal line and controls the other hardware.

[0049] Processor 11 is an integrated circuit (IC) that performs processing. IC stands for Integrated Circuit. Specifically, processor 11 is a CPU, DSP, or GPU. CPU stands for Central Processing Unit. DSP stands for Digital Signal Processor. GPU stands for Graphics Processing Unit.

[0050] The memory 12 is a storage device that temporarily stores data. Specifically, the memory 12 is an SRAM or a DRAM. SRAM is the abbreviation for Static Random Access Memory. DRAM is the abbreviation for Dynamic Random Access Memory.

[0051] Memory 13 is a storage device for storing data. Specifically, memory 13 is an HDD. HDD stands for Hard Disk Drive. Alternatively, memory 13 may be a removable recording medium such as an SD (registered trademark) memory card, CompactFlash (registered trademark), NAND flash memory, a floppy disk, an optical disk, a high-density disk, a Blu-ray (registered trademark) disk, or a DVD. SD stands for Secure Digital. DVD stands for Digital Versatile Disk.

[0052] The communication interface 14 is an interface for communicating with external devices. Specifically, the communication interface 14 is an Ethernet (registered trademark), USB, or HDMI (registered trademark) port. USB stands for Universal Serial Bus. HDMI stands for High-Definition Multimedia Interface.

[0053] The demand forecasting device 10 includes, as functional components, a reception unit 21, a future inventory estimation unit 22, a demand forecasting unit 23, and an output unit 24. The functions of the functional components of the demand forecasting device 10 are realized by software.

[0054] The memory 13 stores a program for realizing the functions of each functional component of the demand forecasting device 10. The program is read into the memory 12 by the processor 11 and executed by the processor 11. Thus, the functions of each functional component of the demand forecasting device 10 are realized.

[0055] In the memory 13 , during the operation of the demand forecasting device 10 , order pattern information 61 , past inventory information 62 , future inventory information 63 , and demand forecast results 64 are stored.

[0056] exist Figure 3 In FIG, only one processor 11 is shown. However, there may be multiple processors 11, and the multiple processors 11 may cooperate to execute programs that realize various functions.

[0057] ***Description of the action***

[0058] Reference Figures 4 to 9 The operation of the demand forecasting device 10 according to the first embodiment will be described.

[0059] The operation procedure of the demand forecasting apparatus 10 in the first embodiment corresponds to the demand forecasting method in the first embodiment. In addition, the program for realizing the operation of the demand forecasting apparatus 10 in the first embodiment corresponds to the demand forecasting program in the first embodiment.

[0060] The operation of the demand forecasting device 10 includes processing during input acceptance and processing during demand forecasting.

[0061] Reference Figure 4 The input acceptance processing of the first embodiment will be described.

[0062] (Step S101: Acceptance Processing)

[0063] Accepting unit 21 accepts input of ordering pattern information 61 and past inventory information 62 from input terminal 51. However, ordering pattern information 61 is required only for the first acceptance, and accepting unit 21 does not need to accept ordering pattern information 61 every time it executes input acceptance processing.

[0064] Reference Figure 5 The order model information 61 according to the first embodiment will be described.

[0065] Order model information 61 includes a set of one or more carrier names, order models, order conditions, order quantity, and procurement lead time. Order conditions are conditions for placing an order, and an order is placed if the conditions are met. The order quantity is the quantity ordered if the conditions are met. Procurement lead time is the period from when a product is ordered until it arrives and is managed as inventory.

[0066] For example, let's assume that operator A's ordering model uses the order point method. In this case, the information corresponding to the order point method includes the order point and order quantity, which represent order conditions. Therefore, order model information 61 contains a set of information including operator A, the order point method, the order point and order quantity, and the procurement lead time. The order point method refers to an ordering method that orders when inventory levels fall below a certain threshold, called the order point.

[0067] Reference Figure 6 The past inventory information 62 of the first embodiment will be described.

[0068] The past inventory information 62 includes one or more sets of a date, a company name, a product name, and an inventory amount.

[0069] (Step S102: Storage Process)

[0070] The receiving unit 21 writes the order pattern information 61 and the past inventory information 62 into the memory 13. However, the order pattern information 61 is recorded only when the order pattern information 61 is received in step S101.

[0071] Here, the order pattern information 61 and the past inventory information 62 are written into the memory 13. However, the order pattern information 61 and the past inventory information 62 may be written into a storage device provided outside the demand forecasting device 10 instead of the memory 13.

[0072] Reference Figure 7 The demand forecasting process according to the first embodiment will be described.

[0073] (Step S201: Reading Process)

[0074] The future inventory estimation unit 22 reads information about operator A, the supply destination of component A, a forecasted item, from the ordering pattern information 61 stored in the memory 13. Specifically, the future inventory estimation unit 22 reads operator A's ordering pattern, ordering point, order quantity, and procurement lead time.

[0075] Furthermore, the future inventory estimation unit 22 reads information on the inventory quantity of component A at the supplier A, from the past inventory information 62 stored in the memory 13. Specifically, the future inventory estimation unit 22 reads the date and inventory quantity of component A at the supplier A during the most recent base period.

[0076] (Step S202: Future Inventory Estimation Process)

[0077] The future inventory estimation unit 22 estimates the future inventory level of carrier A, the supply destination, based on the ordering pattern information 61 and past inventory information 62 read in step S201. At this point, the future inventory estimation unit 22 assumes that at carrier A, inventory will increase due to orders placed according to the ordering conditions and order quantities indicated by the ordering pattern information 61, and that inventory will decrease according to the past inventory levels indicated by the past inventory information 62. For carrier A, the ordering condition is the order point.

[0078] Reference Figure 8 Provide specific instructions.

[0079] The future inventory estimation unit 22 calculates the future inventory level change for each day during the forecast period starting from the day following the day on which the process is executed, based on the order pattern information 61 and past inventory information 62 read in step S201. The future inventory estimation unit 22 then writes the daily future inventory level change into the memory 13 as future inventory information 63.

[0080] The future inventory estimation unit 22 sets each day in the forecast period, starting from the day immediately following the day of the process execution time, as the target day, in descending order. The future inventory estimation unit 22 calculates the inventory level on the target day by adding the inventory level increased on the target day to the inventory level on the day before the target day and subtracting the inventory level decreased on the target day. In this way, the future inventory estimation unit 22 estimates the inventory level for each day in the forecast period.

[0081] If the ordering conditions are met on the day before the procurement lead time of the target day, the future inventory estimation unit 22 sets the order quantity as the inventory quantity to be increased on the target day. Here, the ordering model uses the order point method. Therefore, if the inventory quantity on the day before the procurement lead time is below the order point indicated by the order model information 61, the inventory quantity equal to the order quantity indicated by the order model information 61 is used as the inventory quantity to be increased. If the ordering conditions are not met on the day before the procurement lead time of the target day, the future inventory estimation unit 22 sets the inventory quantity to be increased on the target day to zero.

[0082] The future inventory estimation unit 22 calculates a statistical value of the daily inventory reduction based on the inventory level changes during the most recent reference period read in step S201. The future inventory estimation unit 22 uses the statistical value as the inventory level reduction on the target day. The statistical value may be an average value or a median value.

[0083] (Step S203: Demand Forecast Processing)

[0084] The demand forecasting unit 23 calculates the demand forecast value for each day in the forecast period. Then, the demand forecasting unit 23 writes the demand forecast value for each day into the memory 13 as the demand forecast result 64. Figure 9 As shown, the demand forecast result 64 includes a set of date, business name, product name, and demand quantity.

[0085] Specifically, the demand forecasting unit 23 forecasts that the order quantity indicated by the order pattern information 61 will be required of the carrier X as the supply source when the order conditions indicated by the order pattern information 61 are satisfied under the future inventory quantity estimated in step S202 .

[0086] (Step S204: Output Processing)

[0087] The output unit 24 outputs the demand forecast result 64 to the output terminal 52 .

[0088] ***Effects of Implementation Method 1***

[0089] As described above, the demand forecasting device 10 of Embodiment 1 estimates future inventory levels using past inventory level changes and forecasts demand based on future inventory levels. This allows for demand forecasting that follows detailed daily demand fluctuations by taking recent inventory level fluctuations into account.

[0090] Specifically, the demand forecasting device 10 performs demand forecasting based on the daily inventory information of carrier A, the supply destination, rather than the actual order results of carrier X. This allows for the detection of demand fluctuations before they are reflected in actual order results, tracking detailed daily demand fluctuations. In other words, by considering recent inventory fluctuations, even in the event of significant market fluctuations, highly accurate demand forecasting can be performed without waiting for the accumulation of actual order results.

[0091] ***Other structures***

[0092] <Variation 1>

[0093] In the first embodiment, each functional component is implemented by software. However, as a first modification, each functional component may be implemented by hardware. This first modification will be described with respect to the differences from the first embodiment.

[0094] Reference Figure 10 The configuration of the demand forecasting device 10 according to Modification 1 will be described.

[0095] When each functional component is implemented by hardware, the demand forecasting device 10 includes an electronic circuit 15 instead of the processor 11, memory 12, and storage 13. The electronic circuit 15 is a dedicated circuit that implements the functions of each functional component and the memory 12 and storage 13.

[0096] The electronic circuit 15 may be a single circuit, a complex circuit, a programmable processor, a parallel programmable processor, a logic IC, a GA, an ASIC, or an FPGA. GA stands for Gate Array. ASIC stands for Application Specific Integrated Circuit. FPGA stands for Field-Programmable Gate Array.

[0097] Each functional component may be realized by one electronic circuit 15 , or may be realized by distributing a plurality of electronic circuits 15 .

[0098] <Variation 2>

[0099] As a second modification, some of the functional components may be implemented by hardware, and the other functional components may be implemented by software.

[0100] The processor 11, the memory 12, the storage 13, and the electronic circuit 15 are referred to as a processing circuit. That is, the functions of each functional component are realized by the processing circuit.

[0101] Implementation Method 2

[0102] The difference between the second embodiment and the first embodiment is that the inventory reduction amount is determined based on the demand forecast results for downstream operators in the flow of forecast target products between operators. In the second embodiment, this difference is described, and the description of the similarities is omitted.

[0103] ***Description of the structure***

[0104] Reference Figure 11 The configuration of the demand forecasting device 10 according to the second embodiment will be described.

[0105] Demand forecasting device 10 and Figure 3 The difference is that the business flow structure information 65 is stored in the memory 13.

[0106] ***Description of the action***

[0107] Reference Figure 4 The input acceptance processing of the second embodiment will be described.

[0108] (Step S101: Acceptance Processing)

[0109] In addition to order pattern information 61 and past inventory information 62, the accepting unit 21 also accepts input of business flow structure information 65 from the input terminal 51. However, order pattern information 61 is required only for the initial acceptance, and the accepting unit 21 does not need to accept order pattern information 61 every time it executes the input acceptance process.

[0110] Reference Figure 12 The business flow structure information 65 according to the second embodiment will be described.

[0111] The business flow structure information 65 indicates the flow of the component A, which is a prediction target product, between operators. The business flow structure information 65 includes operator names, upstream operator names, and downstream operator names.

[0112] The upstream operator name indicates the operator upstream of the operator indicated by the operator name. The downstream operator name indicates the operator downstream of the operator indicated by the operator name. The upstream operator is the source of component A for the operator indicated by the operator name. The downstream operator is the destination of component A for the operator indicated by the operator name.

[0113] (Step S102: Storage Process)

[0114] In addition to the order pattern information 61 and the past inventory information 62, the receiving unit 21 also writes the business flow structure information 65 into the memory 13. However, the order pattern information 61 is recorded only when the order pattern information 61 is received in step S101.

[0115] Reference Figure 13 The demand forecast processing according to the second embodiment will be described.

[0116] (Step S211: Downstream Search Processing)

[0117] The future inventory estimation unit 22 refers to the business flow structure information 65 stored in the memory 13 and searches for a carrier on the downstream side of the carrier A, which is the supply destination of the component A that is the forecast target product.

[0118] Specifically, the future inventory estimation unit 22 identifies the operator indicated by the downstream operator name of operator A in the business flow structure information 65. The future inventory estimation unit 22 then identifies the operator indicated by the downstream operator name of each identified operator. This step is repeated until no operator indicated by the downstream operator name is found. In this way, all operators downstream of operator A are retrieved.

[0119] exist Figure 12 In the case of the business flow structure information 65 shown, operators C and D are identified as downstream operators of operator A. Operator F is identified as the downstream operator of operator C. Since operator D has no downstream operators, the search is aborted. Since operator F has no downstream operators, the search is aborted.

[0120] (Step S212: Reading Process)

[0121] The future inventory estimation unit 22 reads out the order pattern information 61 and the past inventory information 62 related to each operator (here, operators C, D, and F) searched in step S211 in addition to operator A.

[0122] Specifically, the future inventory estimation unit 22 sets operator A and each operator retrieved in step S211 as the target operators. The future inventory estimation unit 22 reads the target operator information from the order model information 61. Furthermore, the future inventory estimation unit 22 reads the inventory quantity information related to component A at the target operator from the past inventory information 62.

[0123] The target operators are selected in order from the downstream operator, and the processes of step S213 and step S214 are executed.

[0124] (Step S213: Future Inventory Estimation Process)

[0125] The future inventory estimation unit 22 estimates the future inventory amount of the target operator based on the order pattern information 61 and the past inventory information 62 read out in step S212 .

[0126] Specifically, similar to the first embodiment, the future inventory estimation unit 22 sets each day in the forecast period, starting from the day following the day of the process execution time point, as the target day, in descending order of time. The future inventory estimation unit 22 calculates the inventory level on the target day by adding the inventory level increased on the target day to the inventory level on the day before the target day and subtracting the inventory level decreased on the target day.

[0127] However, in the second embodiment, when the target operator does not have a downstream operator, the future inventory estimation unit 22 uses the statistical value as the inventory reduction on the target day, similar to the first embodiment. On the other hand, when the target operator has a downstream operator, as shown in FIG. Figure 14 As shown in FIG, the future inventory estimation unit 22 uses the demand of the immediately downstream operator as the inventory amount to be reduced on the target day. Figure 14 In the example, the ordering model is the order point method. In addition, when there are multiple operators immediately downstream, the future inventory estimation unit 22 uses the total demand of the multiple operators as the inventory reduction amount on the target day.

[0128] Reference Figure 15 A specific example will be described.

[0129] Here, the target operators are selected starting from the downstream operator. Therefore, first, operator F or operator D is selected as the target operator. After the processing for operator F is completed, operator C is selected as the target operator. After the processing for operators C and D is completed, operator A is selected as the target operator.

[0130] Operators F and D do not have downstream operators. Therefore, daily inventory levels are calculated in the same manner as in Embodiment 1. Operators C and A do have downstream operators. Therefore, daily inventory levels are calculated using the demand of the immediately downstream operator as the inventory reduction on the target day.

[0131] When calculating Carrier C's future inventory, the daily inventory reduction for Carrier C is determined based on the demand forecast results related to orders from Carrier F to Carrier C. When calculating Carrier A's future inventory, the daily inventory reduction for Carrier A is determined based on the demand forecast results related to orders from Carrier C and Carrier D to Carrier A.

[0132] (Step S214: Demand Forecast Processing)

[0133] The demand forecasting unit 23 forecasts that, under the future inventory estimated in step S213 , when the order conditions indicated by the order pattern information 61 are satisfied, a demand for the order quantity indicated by the order pattern information 61 will be generated for the immediately upstream carrier.

[0134] The processing of step S215 is the same as Figure 7 The processing of step S204 is the same.

[0135] ***Effects of Implementation 2***

[0136] As described above, the demand forecasting device 10 in Embodiment 2 determines the amount of inventory reduction based on the demand forecast results for downstream carriers in the flow of forecasted goods between carriers. This allows for highly accurate estimation of the future inventory levels of the carriers at the supply destination. Consequently, highly accurate demand forecasting is possible.

[0137] ***Other structures***

[0138] <Variation 3>

[0139] Demand forecasting for downstream operators of operator A may be performed using methods other than those described in Embodiments 1 and 2, such as known methods. Furthermore, when forecasting demand from operator A for operator X, the demand from downstream operators for operator A, as predicted by any method, may be used as the inventory reduction for operator A.

[0140] Implementation 3

[0141] The difference between the third embodiment and the second embodiment is that the inventory level for the day on which the process is executed is estimated by referring to the logistics information 66 indicating the migration of the predicted target product during the logistics process. In the third embodiment, this difference is described, and the description of the similarities is omitted.

[0142] Here, a case where functions are added to Embodiment 1 will be described. However, functions may also be added to Embodiment 2.

[0143] ***Description of the structure***

[0144] Reference Figure 16 The configuration of the demand forecasting device 10 according to the third embodiment will be described.

[0145] Demand forecasting device 10 and Figure 3 The difference between the present invention and the present invention is that the present invention has a past inventory estimation unit 25 as a functional component. Like other functional components, the function of the past inventory estimation unit 25 is realized by software or hardware.

[0146] In addition, the demand forecasting device 10 and Figure 3 The difference is that logistics information 66 and past inventory information 67 are stored in the memory 13.

[0147] ***Description of the action***

[0148] Reference Figure 4 The input acceptance processing of the third embodiment will be described.

[0149] (Step S101: Acceptance Processing)

[0150] Accepting unit 21 accepts input of ordering model information 61 and distribution information 66 from input terminal 51. However, ordering model information 61 is required only for the first acceptance, and accepting unit 21 does not need to accept ordering model information 61 every time it executes input acceptance processing.

[0151] Reference Figure 17 The logistics information 66 of the third embodiment will be described.

[0152] Logistics information 66 indicates the movement of the predicted product during the logistics process. Logistics information 66 includes the date, operator name, product name, event name, and quantity. Here, "event name" refers to the name of the event. An event refers to the movement of a product during the logistics process, such as production, shipment, and arrival. Quantity refers to the number of products that moved during the event.

[0153] (Step S102: Storage Process)

[0154] The receiving unit 21 writes the order model information 61 and the distribution information 66 into the memory 13. However, the order model information 61 is recorded only when the order model information 61 is received in step S101.

[0155] Reference Figure 18 The demand forecast processing according to the third embodiment will be described.

[0156] (Step S221: Reading Process)

[0157] The future inventory estimation unit 22 reads out the information of the carrier A from the order pattern information 61 stored in the memory 13 . Furthermore, the future inventory estimation unit 22 reads out the physical distribution information 66 stored in the memory 13 .

[0158] (Step S222: Past Inventory Estimation Process)

[0159] The past inventory estimation unit 25 estimates the inventory quantity of component A, a forecasted item, at the supplier A, a supply destination, for each day of the most recent base period preceding the day to which the process is executed, by referring to the physical distribution information 66. The past inventory estimation unit 25 writes the estimated daily inventory quantity into the memory 13 as past inventory information 67.

[0160] Reference Figure 19 To be explained in detail, the past inventory estimation unit 25 sets the target day in descending order, in the most recent base period preceding the day on which the processing execution time occurs. The past inventory estimation unit 25 sets an initial value for the inventory level on the day before the earliest day in the most recent base period. The past inventory estimation unit 25 calculates the inventory level on the target day by adding the inventory level increased on the target day as indicated by the logistics information 66 to the inventory level on the day before the target day and subtracting the inventory level decreased on the target day as indicated by the logistics information 66. This estimates the inventory level on the day on which the processing execution time occurs.

[0161] The inventory quantity increased on the target day is the quantity of components A arriving at the operator A on the target day in the physical distribution information 66. The inventory quantity decreased on the target day is the quantity of components A shipped from the operator A on the target day in the physical distribution information 66.

[0162] (Step S223: Future Inventory Estimation Process)

[0163] The future inventory estimation unit 22 estimates the future inventory of the operator A in the same manner as in the first embodiment, using the inventory indicated by the past inventory information 67 estimated in step S222 as the past inventory indicated by the past inventory information 62 .

[0164] The processing of step S224 and step S225 is the same as Figure 7 The processing of step S203 and step S204 is the same.

[0165] ***Effects of Implementation 3***

[0166] As described above, the demand forecasting device 10 of Embodiment 3 estimates the inventory level for the day on which the processing is executed by referring to the logistics information 66, thereby estimating the future inventory level. This makes it possible to estimate the future inventory level of operator A even if information on the inventory level for the day on which the processing is executed is unavailable. Consequently, demand forecasting is possible even if information on the inventory level for the day on which the processing is executed by operator A is unavailable.

[0167] Implementation 4

[0168] The fourth embodiment differs from the third embodiment in that the inventory level for the day on which the process is executed is estimated by utilizing the movement of used products, which are manufactured using the forecast target items as components, during the logistics process. This difference will be described in the fourth embodiment, and the description of the similarities will be omitted.

[0169] ***Description of the structure***

[0170] Reference Figure 20 The configuration of the demand forecasting device 10 according to the fourth embodiment will be described.

[0171] Demand forecasting device 10 and Figure 16 The difference is that component structure information 68 is stored in the memory 13 .

[0172] ***Description of the action***

[0173] Reference Figure 4 The input acceptance processing of the fourth embodiment will be described.

[0174] (Step S101: Acceptance Processing)

[0175] Accepting unit 21 accepts input of order model information 61, distribution information 66, and component structure information 68 from input terminal 51. However, order model information 61 is required only for the initial acceptance and does not need to be accepted every time input acceptance processing is executed.

[0176] Reference Figure 21 The component configuration information 68 according to the fourth embodiment will be described.

[0177] Component structure information 68 indicates the quantity of component A, a prediction target item, used in the product. Component structure information 68 includes a product name, a component name, and a quantity. The product name is the name of the product. The component name is the name of the component used in the product, such as component A. The quantity is the number of components used in the product.

[0178] (Step S102: Storage Process)

[0179] The receiving unit 21 writes the order model information 61, the physical distribution information 66, and the component structure information 68 into the memory 13. However, the order model information 61 is recorded only when the order model information 61 is received in step S101.

[0180] Reference Figure 18 The demand forecast processing according to the fourth embodiment will be described.

[0181] (Step S221: Reading Process)

[0182] The future inventory estimation unit 22 reads out the information of the carrier A from the order pattern information 61 stored in the memory 13 . The future inventory estimation unit 22 also reads out the physical distribution information 66 and the component structure information 68 stored in the memory 13 .

[0183] (Step S222: Past Inventory Estimation Process)

[0184] As in the third embodiment, the past inventory estimation unit 25 sets the days of the most recent base period preceding the day at the processing execution time point in descending order as target days, and estimates the inventory amount on the target days.

[0185] At this time, the past inventory estimation unit 25 refers to the logistics information 66 to determine whether operator A has produced a used product using component A as a component. If operator A has not produced a used product using component A as a component, the past inventory estimation unit 25 estimates the inventory level for the target day in the same manner as in Embodiment 3. If operator A has produced a used product using component A as a component, the past inventory estimation unit 25 determines the inventory level that decreased on the target day as follows to estimate the inventory level for the target day. This is the same as in Embodiment 3, except for the method for determining the inventory level that decreased on the target day.

[0186] Instead of calculating the number of component A shipped based on the physical distribution information 66, the past inventory estimation unit 25 calculates the production and shipment quantities of the used products produced based on component A. Based on the component structure information 68, the past inventory estimation unit 25 determines the quantity of component A used in the production of the used products or the quantity of component A shipped as incorporated into the used products, and sets this quantity as the inventory reduction.

[0187] Reference Figure 21 A specific example will be described.

[0188] To determine the shipment quantity of component A, the past inventory estimation unit 25 first refers to the physical distribution information 66 to determine whether operator A produces a product such as product A or product B that includes component A as a component. For example, if operator A produces product B, the past inventory estimation unit 25 determines the production or shipment quantity of product B on the target day based on the physical distribution information 66. The past inventory estimation unit 25 then determines the number of component A used in product B based on the component structure information 68 and calculates the total amount of component A used in the production or shipment of product B. The past inventory estimation unit 25 uses this calculated total amount as the inventory reduction on the target day.

[0189] ***Effects of Implementation 4***

[0190] As described above, the demand forecasting device 10 of Embodiment 4 utilizes the migration of products produced using the forecast item as a component, i.e., used products, during the logistics process, to estimate the inventory level on the day the process is executed. This allows for appropriate estimation of the inventory level of forecast items whose tracking is interrupted in the logistics information 66. This means that even if tracking of the forecast item in the logistics information 66 is interrupted, such as due to changes in its shape during distribution, the inventory level can be appropriately estimated.

[0191] In addition, the word "unit" in the above description may be rewritten as "circuit," "process," "step," "processing," or "processing circuit."

[0192] The above describes the embodiments and variations of the present disclosure. Several of these embodiments and variations may be implemented in combination. Furthermore, any one or several of these embodiments and variations may be implemented partially. Furthermore, the present disclosure is not limited to the above embodiments and variations, and various modifications may be made as needed.

[0193] Label Description

[0194] 10: Demand forecasting device; 11: Processor; 12: Memory; 13: Storage; 14: Communication interface; 15: Electronic circuit; 21: Receiving unit; 22: Future inventory estimation unit; 23: Demand forecasting unit; 24: Output unit; 25: Past inventory estimation unit; 51: Input terminal; 52: Output terminal; 61: Ordering model information; 62: Past inventory information; 63: Future inventory information; 64: Demand forecasting result; 65: Business flow structure information; 66: Logistics information; 67: Past inventory information; 68: Component structure information.

Claims

1. A demand forecasting device, comprising: a future inventory estimation unit that estimates a future inventory level at a supply destination of the forecasted product, assuming that inventory at the supply destination increases due to orders placed in accordance with the order conditions and order quantities indicated by the order model information, and that inventory decreases in accordance with a past inventory level indicated by the past inventory information; and The demand forecasting unit forecasts that, under the future inventory amount estimated by the future inventory estimating unit, a demand for the order quantity indicated by the order pattern information will be generated for a supply source at a point in time when an order condition indicated by the order pattern information is satisfied.

2. The demand forecasting device according to claim 1, wherein: The demand forecasting device forecasts the demand generated in each unit period. The future inventory estimation unit sets each unit period in the forecast period starting from the next unit period of the unit period to which the processing execution time point belongs in order from earliest to latest as an object, adds the inventory amount in the unit period before the object unit period to the inventory amount increased in the object unit period, and subtracts the inventory amount decreased in the object unit period to calculate the inventory amount in the object unit period, thereby estimating the inventory amount in each unit period in the forecast period.

3. The demand forecasting device according to claim 2, wherein: When the order condition is satisfied in a unit period before the procurement lead time of the target unit period, the future inventory estimation unit sets the order quantity as the inventory quantity increased in the target unit period.

4. The demand forecasting device according to claim 2 or 3, wherein: The future inventory estimation unit calculates a statistical value of an inventory reduction amount per the unit period based on a change in the inventory amount indicated by the past inventory information, and uses the statistical value as the inventory amount reduced in the target unit period.

5. The demand forecasting device according to claim 4, wherein: The demand forecasting device sets target operators in order from the downstream operator in the flow of the prediction target product between operators, and forecasts the demand generated for the supply source of the operator that is the upstream operator of the target operator. The future inventory estimation unit uses the statistical value as the inventory amount reduced during the target unit period when the target operator does not have a downstream operator, and uses the demand of the downstream operator as the inventory amount reduced during the target unit period when the target operator has a downstream operator.

6. The demand forecasting device according to any one of claims 2 to 5, wherein: The demand forecasting device further includes a past inventory estimation unit, which estimates the inventory quantity of the supply destination per unit period of a past base period up to the unit period to which the processing execution point belongs, with reference to logistics information indicating the migration of the forecast object product at the supply destination during the logistics process. The future inventory estimation unit estimates the future inventory quantity using the inventory quantity estimated by the past inventory estimation unit as the past inventory quantity indicated by the past inventory information.

7. The demand forecasting device according to claim 6, wherein: The past inventory estimation unit sets an initial value for the inventory amount in the unit period before the base period of the unit period to which the processing execution point belongs, and sets each unit period from the next unit period before the base period to the unit period to which the processing execution point belongs as an object in sequence, adds the inventory amount increased in the object unit period shown in the logistics information to the inventory amount in the unit period before the object unit period, and subtracts the inventory amount decreased in the object unit period shown in the logistics information to calculate the inventory amount in the object unit period, thereby estimating the inventory amount in the unit period to which the processing execution point belongs.

8. The demand forecasting device according to claim 7, wherein: The logistics information indicates the migration of products produced using the predicted product as a component during the logistics process. The past inventory estimation unit refers to component configuration information indicating the quantity of the prediction target product used in the used product, considers the usage quantity of the prediction target product calculated based on the production number of the used product, and determines the inventory quantity reduced during the target unit period.

9. A demand forecasting method, wherein: The computer estimates the future inventory level of the supply destination, assuming that inventory at the supply destination of the forecasted product increases due to orders placed according to the order conditions and order quantities indicated by the order model information, and that inventory decreases according to the past inventory level indicated by the past inventory information. The computer predicts that, under the estimated future inventory level, at a point in time when the ordering conditions indicated by the ordering pattern information are satisfied, a demand for the order quantity indicated by the ordering pattern information will be generated from the supply source.

10. A demand forecasting program, the program causing a computer to function as a demand forecasting device, the demand forecasting device performing the following processing: a future inventory estimation process for estimating the future inventory level of a supply destination of the forecasted product, assuming that inventory at the supply destination increases due to orders placed in accordance with the order conditions and order quantities indicated by the order model information, and that inventory decreases in accordance with the past inventory level indicated by the past inventory information; and The demand forecasting process forecasts that, under the future inventory amount estimated by the future inventory estimation process, a demand for the order quantity indicated by the order pattern information will be generated for a supply source at a time when the order conditions indicated by the order pattern information are satisfied.

Images