[go: up one dir, main page]

WO2018139029A1 - Dispositif de prévision de demande, système de prévision de demande, procédé de prévision de demande, et programme - Google Patents

Dispositif de prévision de demande, système de prévision de demande, procédé de prévision de demande, et programme Download PDF

Info

Publication number
WO2018139029A1
WO2018139029A1 PCT/JP2017/042375 JP2017042375W WO2018139029A1 WO 2018139029 A1 WO2018139029 A1 WO 2018139029A1 JP 2017042375 W JP2017042375 W JP 2017042375W WO 2018139029 A1 WO2018139029 A1 WO 2018139029A1
Authority
WO
WIPO (PCT)
Prior art keywords
shipments
order
demand
product
value
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/JP2017/042375
Other languages
English (en)
Japanese (ja)
Inventor
陽介 斉藤
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Mitsubishi Electric Corp
Original Assignee
Mitsubishi Electric Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Mitsubishi Electric Corp filed Critical Mitsubishi Electric Corp
Priority to JP2018564124A priority Critical patent/JPWO2018139029A1/ja
Publication of WO2018139029A1 publication Critical patent/WO2018139029A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising

Definitions

  • the present invention relates to a demand prediction device, a demand prediction system, a demand prediction method, and a program.
  • Demand forecasting device that forecasts demand for goods is used to manage product production. There has been proposed a demand forecasting device that manages past orders of products separately into normal orders and orders other than normal orders and reflects them in demand forecasting. Orders other than normal orders include sudden orders and orders with special demand.
  • Patent Document 1 describes a sales plan creation support system that classifies business negotiations into normal business negotiations and collective business negotiations, and manages the planning of sales plans by subtracting the delivery results of the collective business negotiations from the delivery results. Yes.
  • This invention is made in view of this subject, Comprising: It aims at providing the demand forecast apparatus, demand forecast system, demand forecast method, and program which can perform the demand forecast which followed the fluctuation
  • a demand prediction apparatus includes a threshold value calculation unit, a shipment result totaling unit, an expansion rate calculation unit, and a demand prediction value calculation unit.
  • the threshold calculation unit calculates a threshold for dividing the order into a normal order and a large order based on the number of past shipments of the product.
  • the shipment performance totaling unit divides orders for products before the date of demand prediction into normal orders with the shipment quantity smaller than the threshold value and large orders with the shipment quantity equal to or greater than the threshold value. Tally.
  • the expansion rate calculation unit calculates an expansion rate, which is an increase rate of the number of shipments from the date when the demand is predicted to the demand prediction target date, based on the past number of shipments of the product.
  • the demand forecast value calculation unit calculates the predicted value of the normal order from the number of shipments of the normal order products and the growth rate that are aggregated by the shipment performance aggregation unit, and from the number of product shipments of the large orders that are aggregated by the shipment performance aggregation unit Calculate the forecast value for large orders.
  • a demand prediction that can perform a demand prediction that follows a fluctuation in a demand trend by calculating a threshold for dividing an order into a normal order and a large order based on the number of past shipments of the product.
  • Block diagram which shows the structure of the demand prediction apparatus which concerns on embodiment of this invention.
  • Block diagram showing the hardware configuration of the demand forecasting device Diagram showing the configuration of the model master Flow chart of threshold calculation processing
  • Figure showing normal order shipment results and large order shipment results
  • Flow chart of expansion rate calculation processing Diagram showing growth rate by model group
  • Flow chart of demand forecast value calculation processing Diagram showing normal order forecast value and large order forecast value Diagram showing the structure of negotiation information Diagram showing the order accuracy coefficient master configuration
  • Block diagram showing the configuration of the demand forecasting system
  • Embodiment 1 A demand prediction apparatus 10 according to Embodiment 1 of the present invention will be described with reference to FIGS. In the present specification, description will be made on the assumption that the demand prediction for predicting the demand of the product at the end of Y is performed on the Y month Z year of the year X. The end of the month Y, which is the target of demand forecast, is also called the demand forecast target day. Moreover, although it demonstrates using a model as a reference
  • the demand prediction device 10 is a device that predicts the demand for a product based on past shipment results of the product, order for the product, and information on the negotiation related to the order for the product.
  • FIG. 1 is a block diagram showing the configuration of the demand prediction apparatus 10.
  • the demand prediction device 10 includes a processing unit 100 that performs processing for calculating demand prediction information, a storage unit 200 that stores data for calculating demand prediction information, and calculated demand prediction information. .
  • FIG. 2 is a block diagram illustrating a hardware configuration of the demand prediction apparatus 10.
  • the demand prediction device 10 includes a processing device 300 that functions as the processing unit 100, a storage device 400 that functions as the storage unit 200, a ROM 500 that stores a program executed by the processing device 300, and an external device.
  • a communication unit 600 that communicates data, an input unit 700 that receives input from the user, and an output unit 800 that presents information to the user.
  • Each of these parts is electrically connected to each other via a bus.
  • the processing apparatus 300 is composed of a CPU (Central Processing Unit).
  • the storage device 400 includes a memory and a hard disk.
  • the ROM 500 is composed of ROM (Read Only Memory).
  • the communication unit 600 includes a NIC (Network Interface Card) and an antenna.
  • the input unit 700 includes a mouse and a keyboard.
  • the output unit 800 includes a display and a speaker.
  • the demand prediction apparatus 10 is composed of a computer including a personal computer, a server, a mainframe, and a workstation.
  • the processing unit 100 is a processing device that calculates demand prediction information based on data stored in the storage unit 200.
  • the processing unit 100 includes a threshold value calculation unit 110 that calculates a threshold value for dividing an order into a normal order and a large order, and a shipment result totaling unit that totals the total value of the number of shipments according to the normal order and the total value of the number of shipments according to the large order. 120, an expansion rate calculation unit 130 that calculates an expansion rate from the number of shipments of past normal orders, a predicted value of demand for normal orders, a predicted value of demand for large orders, and an error rate of past demand prediction A demand forecast value calculation unit 140.
  • the processing unit 100 may include a CPU, but is not limited thereto.
  • the threshold calculation unit 110 calculates a threshold for dividing an order into a normal order and a large order based on the order-specific shipping information stored in the storage unit 200, and stores the threshold as a model group-specific threshold 212 in the storage unit 200.
  • the shipment information by order includes an order reception date, a shipment date, a model name, and the number of shipments.
  • the shipment information by order used by the threshold calculation unit 110 to calculate the threshold is typically shipment information by order for the past year, but is not limited thereto.
  • the shipment result totaling unit 120 acquires order-specific shipment information stored in the storage unit 200, and based on the model group-specific threshold 212 calculated by the threshold calculation unit 110, orders for the current month for each model group and a large order. Divide into orders. Then, the total value of the number of shipments according to the normal order and the total value of the number of shipments according to the large order are totaled for each model group, and are stored in the storage unit 200 as the normal order shipment result 221 and the large order order shipment result 222, respectively.
  • the expansion rate calculation unit 130 calculates the expansion rate from the number of shipments based on past normal orders stored in the storage unit 200 and stores it in the storage unit 200 as the model group expansion rate 213.
  • the growth rate is the growth rate of the average number of shipments per day at the end of the month with respect to the average number of shipments per day on the demand forecast date.
  • the number of shipments based on normal orders used by the expansion rate calculation unit 130 for calculating the expansion rate is typically the number of shipments for the past five years, but is not limited to this.
  • the growth rate calculation unit 130 calculates the average growth rate by taking the average of the calculated growth rates for the past five years, and stores it in the storage unit 200 as the model group growth rate 213.
  • the demand forecast value calculation unit 140 calculates the forecast value of demand up to the end of the current month for normal orders, the forecast value of demand up to the end of the month of large orders, and the error rate of the past demand forecast for each model group.
  • the demand predicted value calculating unit 140 stores the calculated predicted value of demand up to the end of the current month for normal orders, predicted value of demand up to the end of the current month of large orders, and past demand prediction error rates in the storage unit 200, respectively.
  • the demand predicted value calculation unit 140 calculates a predicted value of demand up to the end of the current month of the normal order. calculate.
  • the demand prediction value calculation unit 140 stores the calculated prediction value in the storage unit 200 as the normal order prediction value 223.
  • the demand predicted value calculation unit 140 acquires the order information, the negotiation information, and the order accuracy coefficient stored in the storage unit 200, and calculates the predicted value of demand up to the end of the current month for large orders for each model group.
  • the demand predicted value calculation unit 140 stores the calculated predicted value in the storage unit 200 as a large order predicted value 224.
  • the order information includes, but is not limited to, the order reception date, delivery date, model name, and planned shipment quantity.
  • the negotiation information includes, but is not limited to, the model name, planned delivery date, planned shipment quantity, and order receipt accuracy.
  • the demand prediction value calculation unit 140 is configured to calculate the demand forecast from the normal order shipment result 221, the large order shipment result 222 and the order information stored in the storage unit 200, and the past normal order prediction value 223 and the large order order prediction value 224. Calculate the error rate of the actual shipment quantity.
  • the demand predicted value calculation unit 140 stores the calculated error rate in the storage unit 200 as the predicted value error rate 226.
  • the past normal order forecast value 223 and the large order forecast value 224 used by the demand forecast value calculation unit 140 are typically forecast values for the past year, but are not limited thereto.
  • the storage unit 200 is a storage device that stores data for calculating demand prediction information and the demand prediction information calculated by the processing unit 100.
  • the storage unit 200 includes a model information database 210 that stores information on models for which demand prediction is performed, a shipping performance prediction management database 220 that stores data necessary for performing demand prediction and calculated demand prediction values, order information, A transaction database 230 for storing order-specific shipping information and negotiation information.
  • the model information database 210 is a database that stores information on models for which demand is to be predicted.
  • the model information database 210 includes a model master 211 including a correspondence relationship between models and model groups, a model group threshold 212 including a threshold for dividing an order into a normal order and a large order, and a model including an expansion rate for each model group. And an expansion rate 213 for each group.
  • FIG. 3 is a diagram showing the configuration of the model master 211.
  • the model master 211 is data representing a correspondence relationship between an actual model and a model group which is a unit for performing demand prediction.
  • the model master 211 is stored in the storage unit 200 when the demand predictor inputs the models by grouping them into units for performing demand prediction in advance.
  • model A, model B, and model C are model group A
  • model D and model E are model group D.
  • the model F is grouped into a model group F.
  • the shipment record prediction management database 220 stores shipment record data necessary for forecasting demand, the calculated predicted value of demand, and the order accuracy coefficient and the error rate of the predicted value, which are related information. Database.
  • the shipment record prediction management database 220 includes a normal order shipment record 221 including the total value of shipments by normal orders, a large order shipment record 222 including the total value of shipments by large orders, and the demand up to the end of the current month for normal orders.
  • a normal order forecast value 223 including a forecast value, a large order forecast value 224 including a demand forecast value until the end of the current month for a large order, an order accuracy coefficient master 225 including a correspondence relationship between the order accuracy and the order accuracy coefficient receipt, A predicted value error rate 226 including an error rate of the actual number of shipments with respect to the demand forecast.
  • the transaction database 230 is necessary to perform demand prediction, and includes order information used by the threshold value calculation unit 110, the shipment result totaling unit 120, the expansion rate calculation unit 130, and the demand prediction value calculation unit 140, and shipment information by order. This is a database that stores business negotiation information.
  • the processing unit 100 of the demand prediction apparatus performs a threshold value calculation process, a shipment result totaling process, and an expansion rate calculation process in order to calculate data necessary for calculating the demand prediction value. Below, the process which the process part 100 performs is demonstrated in order.
  • the threshold value calculation process executed by the threshold value calculation unit 110 of the processing unit 100 will be described with reference to the flowchart of FIG. Since the threshold value calculation process is performed for each model group, the threshold value calculation unit 110 repeatedly performs the threshold value calculation process for the number of all model groups for which the demand predictor is to calculate the threshold value.
  • the threshold calculation unit 110 Based on the order-specific shipment information for the past year stored in the transaction database 230, the threshold calculation unit 110 refers to the model master 211 in the model information database 210 to determine the number of shipments per order for each model group. Aggregate (step S101).
  • the threshold value calculation unit 110 calculates the corrected average value ⁇ ′ of the number of shipments that has been corrected with emphasis on the latest trend, using equation (1) (step S102).
  • the latest is typically three months, but is not limited thereto.
  • ⁇ ′ ⁇ (Number of shipments per order in the past) + ⁇ (Number of shipments in the last order) ⁇ / ⁇ (Number of orders in the past) + (Number of orders in the last order) ⁇ (1)
  • the seasonal index Sn indicates the ratio of the number of shipments in n months to the number of shipments per year, and is obtained by the following formula (8).
  • the corrected average value ⁇ ′ may be calculated by the following formula (9) in consideration of the seasonality of each month.
  • ⁇ ′ ⁇ ⁇ Sn ⁇ (number of shipments per order in n month) / (number of orders in n month) ⁇ (9)
  • the threshold value calculation unit 110 calculates the square of the standard deviation ⁇ of the number of shipments using the formula (2) (step S103).
  • ⁇ 2 ⁇ ⁇ (Past number of shipments per order) ⁇ (Average value of past shipments) ⁇ 2 ⁇ / (Past number of orders) (2)
  • the threshold calculation unit 110 uses the sum ⁇ ′ + ⁇ of the corrected average value ⁇ ′ and the standard deviation ⁇ as a threshold, and the threshold for each model group in the storage unit 200 212 (step S104).
  • the threshold value calculation unit 110 ends the threshold value calculation process.
  • FIG. 5 is a diagram illustrating an example of the model group-specific threshold 212 calculated and stored by the threshold calculation process.
  • the model group threshold 212 stores the corrected average value ⁇ ′, standard deviation ⁇ , and threshold ⁇ ′ + ⁇ calculated in steps S102 to S104 for each model group.
  • the corrected average value ⁇ ′ 1.151
  • the standard deviation ⁇ 1.352
  • the threshold value ⁇ ′ + ⁇ 2.503.
  • the shipping result totaling process executed by the shipping result totaling unit 120 of the processing unit 100 will be described. Since the shipment record totaling process is performed for each model group, the shipment record totaling unit 120 repeats the threshold value calculation process by the number of all model groups for which the demand predictor intends to calculate the threshold value.
  • the shipment result totaling unit 120 extracts the number of shipments due to orders in the current month, that is, the month for which the demand is predicted, from the order-specific shipment information stored in the transaction database 230 (step S201).
  • step S202 If it is determined that the number of shipments by order is smaller than ⁇ ′ + ⁇ (step S202: YES), the shipment record totaling unit 120 totals the number of shipments by shipment date and by model group, and the shipment record prediction management as the shipment record of normal orders.
  • the normal order shipment result 221 of the database 220 is stored (step S203), and the process proceeds to step S205.
  • step S202 If it is determined that the number of shipments by order is greater than or equal to ⁇ ′ + ⁇ (step S202: NO), the shipment record totaling unit 120 totals the number of shipments by shipment date and by model group, and the large order in the shipment record prediction management database 220 is obtained.
  • the shipment result 222 is stored (step S204), and the process proceeds to step S205.
  • the shipment record totaling unit 120 determines whether all orders of the current month have been totaled as the shipment record (step S205). When it is determined that there is an unaggregated order (step S205: NO), the process returns to step S202.
  • step S205 When it is determined that all the order results for the current month have been totaled as the shipping results (step S205: YES), the shipping result totaling unit 120 ends the shipping result totaling process.
  • FIG. 7 is a diagram showing the normal order shipment record 221 and the large order shipment record 222 calculated and stored by the shipment record totaling process.
  • the shipping result totaling unit 120 classifies orders of the current month into normal orders and large orders, and totals the number of shipments by shipping date and by model group. For example, when forecasting demand on June 11, 2014, orders for the current month up to June 10, 2014 are classified into normal orders and large orders, and the number of shipments is tabulated by shipment date and model group. .
  • the expansion rate calculation process executed by the expansion rate calculation unit 130 of the processing unit 100 will be described with reference to the flowchart of FIG. Since the expansion rate calculation process is performed for each model group, the expansion rate calculation unit 130 repeats the threshold value calculation process by the number of all the model groups for which the demand predictor intends to calculate the threshold value.
  • the growth rate calculation unit 130 extracts the number of shipments made in the past order from the shipment information by order for the past five years stored in the transaction database 230 (step S301).
  • the expansion rate calculation unit 130 extracts the number of orders shipped and the threshold ⁇ ′ + ⁇ registered in the model group threshold 212 of the model group to which the ordered model belongs for each order. Compared to the above, normal orders whose number of orders shipped is smaller than ⁇ ′ + ⁇ are extracted and tabulated every day (step S302).
  • the expansion rate calculation unit 130 calculates the expansion rate according to equation (3) for each model group and year based on the number of shipments based on normal orders (step S303).
  • M represents the year before the demand forecast date
  • M X-5, X-4,..., X-1.
  • (M / Y / Y growth rate) ⁇ (Number of shipments due to normal order in year Y / M) / (Number of working days in month / month M / Y) ⁇ / ⁇ (Number of shipments due to order normal to month / month Y / Z-1) / (M year) Number of working days until month Y (Z-1)) ⁇ (3)
  • the growth rate calculation unit 130 calculates the average growth rate by calculating the average value of the calculated growth rates for five years (step S304).
  • the expansion rate calculation unit 130 stores the calculated expansion rate and the average expansion rate in the model group-specific expansion rate 213 of the model information database 210 (step S305).
  • the expansion rate calculation unit 130 ends the expansion rate calculation process.
  • the expansion rate calculation unit 130 executes an expansion rate calculation process for all model groups for which the demand predictor intends to calculate the expansion rate.
  • FIG. 9 is a diagram showing the model group expansion rate 213 calculated and stored by the expansion rate calculation process.
  • the model group growth rate 213 stores the yearly growth rate and average growth rate calculated in steps S303 and S304 for each model group. Note that FIG. 9 also shows the number of shipments and the number of working days for normal orders as reference values.
  • the growth rate of the model group A in June 2013 is calculated by the following formula.
  • the demand forecast value calculation process executed by the demand forecast value calculation unit 140 will be described with reference to the flowchart of FIG.
  • the demand predicted value calculation unit 140 is based on the normal order shipment result 221 stored in the storage unit 200 by the shipment result totaling unit 120 and the calculated model group expansion rate 213 stored in the storage unit 200 by the expansion rate calculation unit 130. Then, the predicted value of the normal order is calculated by the equation (4) (step S401).
  • (Predicted value for regular orders in Y month) ⁇ (Number of shipments of normal orders up to Y month Z day of X) / Number of working days up to Y month Z day of X ⁇ ⁇ (Average growth rate) ⁇ (Number of working days of Y month) (4)
  • the demand predicted value calculation unit 140 stores the calculated predicted value of the normal order in the normal order predicted value 223 of the shipment performance prediction management database 220 (step S402).
  • the demand predicted value calculation unit 140 extracts orders from the order information stored in the transaction database 230 that have not been shipped and the delivery date is the end of Y (step S403).
  • the demand forecast value calculation unit 140 counts the number of orders shipped and the threshold ⁇ ′ + ⁇ registered in the model group threshold 212 of the model group to which the ordered model belongs. Are extracted for each of the extracted orders, and large orders whose planned shipment quantity is greater than or equal to ⁇ ′ + ⁇ are extracted and tabulated daily (step S404).
  • the demand forecast value calculation unit 140 extracts the planned shipment number of large negotiations whose number is greater than ⁇ ′ + ⁇ from the negotiation information stored in the transaction database 230 (step S405). ).
  • the demand forecast value calculation unit 140 extracts the large-order order shipment result 222, the extracted number of large-order orders that are not yet shipped and the delivery date is the end of Y, and the extracted large-scale negotiation shipments. From the number and the order accuracy coefficient stored in the order accuracy coefficient master 225, the predicted value of the large order is calculated by the equation (5) (step S406).
  • the demand predicted value calculation unit 140 stores the predicted value of the large order calculated for each model group in the large order predicted value 224 of the shipment performance prediction management database 220 (step S407).
  • the demand forecast value calculation unit 140 calculates the demand forecast value by the formula (6) from the calculated forecast value of the normal order and the calculated forecast value of the large order (step S408).
  • (Demand forecast value for year X, month Y) (Predicted value of normal order in year Y in X) + (Predicted value in large order in year Y in X) (6)
  • the demand prediction value calculation unit 140 stores the past normal order prediction value 223 stored in the storage unit 200, the past large order prediction value 224, the past normal order shipment result 221, and the past An error rate of the predicted value is calculated from the large-order order shipment result 222 using the equation (7) (step S409).
  • the demand predicted value calculation unit 140 stores the calculated error rate of the predicted value in the predicted value error rate 226 of the shipping performance prediction management database 220 (step S410).
  • the demand predicted value calculation unit 140 ends the demand predicted value calculation process.
  • FIG. 11 is a diagram showing the normal order prediction value 223 and the large order prediction value 224 calculated and stored by the demand prediction value calculation process.
  • FIG. 11 also shows numerical values necessary for calculating the predicted value.
  • the predicted value of the normal order of the model group A is calculated as follows using Equation (4).
  • the day before the demand forecast indicates the day of month Y (Z-1).
  • FIG. 12 is a diagram showing the structure of the negotiation information. As shown in FIG. 12, the negotiation information includes the model name, the scheduled delivery date, the number of shipments, and the order accuracy.
  • FIG. 13 is a diagram illustrating a configuration of the order accuracy coefficient master 225 stored in the storage unit 200.
  • the order accuracy coefficient master 225 stores an order accuracy coefficient for each order accuracy.
  • the order accuracy coefficient defines a weight for reflecting the negotiation in the predicted value according to the order accuracy included in the negotiation information.
  • the order accuracy coefficient in the case of the order accuracy a is 0.9.
  • the predicted value of the large order of model group A is calculated as follows using equation (5).
  • the demand prediction device 10 calculates a threshold value for dividing an order into a normal order and a large order based on the number of past shipments of the product, and changes in demand trends Demand forecasting can be performed.
  • the demand predictor can perform production management that follows fluctuations in the demand trend.
  • the demand prediction apparatus 10 Since the demand prediction apparatus 10 according to the present embodiment calculates the threshold value and the demand prediction value for each model group, even when the number of models is large, the demand prediction apparatus 10 can perform the demand prediction following the fluctuation of the demand trend for each model. .
  • the demand forecasting apparatus 10 uses the extracted large shipments and shipments of large negotiations during the current month that are not yet shipped and the delivery date for this month to calculate the demand forecast value of the large orders, thereby obtaining a special order belonging to the large orders. Demand can be predicted more accurately.
  • Embodiment 2 A demand prediction apparatus according to Embodiment 2 of the present invention will be described with reference to FIG.
  • Steps S501 to S510 in the flowchart in FIG. 14 are the same as steps S401 to S410 in the flowchart in FIG.
  • step S511 determines whether the predicted value of the normal order is larger than the calculated predicted value of the large order. If it is determined that the predicted value of the normal order is larger than the predicted value of the large order (step S511: YES), the process proceeds to step S508.
  • step S512 When it is determined that the predicted value of the normal order is less than or equal to the predicted value of the large order (step S511: NO), the demand predicted value calculation unit 140 outputs an alarm (step S512).
  • the demand forecast value calculation unit 140 accepts correction of the forecast value of the large order (step S513).
  • the demand prediction apparatus 10 has a large order predicted value in addition to the same effect as the demand prediction apparatus 10 according to the first embodiment. If the forecast value is larger than the predicted value of the order, an alarm is output to the demand forecaster and the forecast value of the large order is urged to be corrected, thereby preventing the large order from being reflected in the demand forecast. There is an effect.
  • Embodiment 3 A demand prediction system according to Embodiment 3 of the present invention will be described with reference to FIG.
  • FIG. 15 is a block diagram showing a configuration of the demand prediction system 1.
  • the demand prediction system 1 includes a demand prediction device 10, an ordering / ordering device 20, a shipping performance management device 30, a negotiation management device 40, and a production management device 50.
  • the demand prediction apparatus 10 is the same as the demand prediction apparatus 10 according to the first embodiment.
  • the ordering / ordering device 20 is a device that manages order information.
  • the order receiving / order receiving device 20 is connected to the demand prediction device 10 and can transmit / receive data to / from the demand prediction device 10.
  • the order receiving / order receiving device 20 transmits the order information to the demand prediction device 10.
  • the shipment record management device 30 is a device that manages shipment information by order.
  • the shipping record management device 30 is connected to the demand prediction device 10 and can transmit and receive data to and from the demand prediction device 10.
  • the shipment record management device 30 transmits order-specific shipment information to the demand prediction device 10.
  • the business negotiation management device 40 is a device that manages business negotiation information.
  • the business negotiation management device 40 is connected to the demand prediction device 10 and can transmit and receive data to and from the demand prediction device 10.
  • the negotiation management device 40 transmits the negotiation information to the demand prediction device 10.
  • the production management device 50 is a device that formulates a production plan by inputting demand prediction information and performs production management.
  • the production management device 50 is connected to the demand prediction device 10 and can transmit and receive data to and from the demand prediction device 10.
  • the production management device 50 receives demand prediction information from the demand prediction device 10.
  • the demand prediction information is data including a predicted value of an order calculated by the demand prediction device 10 and an error rate of the predicted value.
  • the ordering / order receiving device 20, the shipping result management device 30, the negotiation management device 40, and the production management device 50 are computers including a personal computer, a server, a mainframe, and a workstation, respectively.
  • the threshold value calculation unit 110 of the demand prediction device 10 calculates a threshold value for dividing an order into a normal order and a large order based on the order-specific shipping information transmitted from the shipping result management device 30, and stores it as a model group-specific threshold value 212. 200.
  • the shipment result totaling unit 120 of the demand prediction device 10 receives the order-specific shipment information from the shipment result management device 30, and based on the model group threshold value 212 calculated by the threshold value calculation unit 110, the order for the current month is classified as a normal order. Divide into orders.
  • the shipping result totaling unit 120 totals the total number of shipments according to the classified normal orders and the total number of shipments according to the large orders for each model, and stores them as normal order shipment results 221 and large order shipment results 222, respectively. To remember.
  • the expansion rate calculation unit 130 of the demand prediction device 10 receives the shipment information by order from the shipment result management device 30, calculates the expansion rate from the number of shipments in the past normal orders, and stores it in the storage unit 200 as the expansion rate 213 by model group.
  • the demand prediction value calculation unit 140 of the demand prediction device 10 includes large order information transmitted from the shipping result management device 30, business negotiation information transmitted from the business negotiation management device 40, and an order accuracy coefficient stored in the storage unit 200. Based on the above, the demand forecast value of the large order up to the end of the current month is calculated and stored in the storage unit 200 as the large order forecast value 224.
  • the demand prediction system 1 has the same effects as the effects of the demand prediction device 10 according to the first embodiment, in addition to orders, shipment results, negotiations, and By providing each device for managing production, there is an effect that production management can be realized by a single system.
  • the processing unit 100 includes the threshold value calculation unit 110, the shipping result totaling unit 120, the expansion rate calculation unit 130, and the demand prediction value calculation unit 140, but is not limited thereto.
  • the functions of the threshold value calculation unit 110 and the shipping result totaling unit 120 may be realized by one functional block.
  • the functions of the threshold value calculation unit 110, the shipping result totaling unit 120, the expansion rate calculation unit 130, and the demand prediction value calculation unit 140 may be realized by one functional block.
  • the demand for goods up to the end of Y on the Y, Z, X year is explained, it is not limited to this.
  • the demand up to the day before the end of Y for example, up to the 15th of Y may be predicted.
  • the demand after the end of Y for example, the end of (Y + 2) end may be predicted.
  • the threshold value calculation unit 110 calculates the correction average value ⁇ ′ of the number of shipments corrected with emphasis on the trend in the last three months, the present invention is not limited to this.
  • the period of emphasis may be longer than 3 months or shorter than 3 months.
  • the priority period may be 0, that is, the average value of the number of shipments that are not corrected by the priority period may be calculated.
  • the threshold value calculation unit 110 simply adds the most recent shipment number in Equation (1), that is, weights the most recent shipment number by a factor of 2, it is not limited to this.
  • the threshold calculation unit 110 may perform weighting with a coefficient smaller than 2 or larger than 2.
  • the demand forecast value calculation part 140 calculated the error rate of the past demand forecast, it is not restricted to this.
  • the demand prediction value calculation unit 140 may further calculate an average value, a minimum value, or a maximum value of the error rate of the prediction value.
  • the growth rate calculation unit 130 calculates the average growth rate from the calculated growth rates for the past five years, it is not limited to this. If there is a year in which the demand trend has fluctuated greatly, the growth rate for that year may be excluded from the growth rate used when calculating the average growth rate. Factors that cause large fluctuations in demand trends include large economic fluctuations.
  • the demand prediction apparatus and the demand prediction system according to the embodiment of the present invention may be realized by a dedicated system or may be realized by a normal computer system.
  • a demand prediction apparatus and a demand prediction system can be obtained by storing and distributing a program for executing the above-described operation in a computer-readable recording medium, installing the program in a computer, and executing the above-described processing. It may be configured. Alternatively, it may be stored in a disk device provided in a server device on the Internet network and downloaded to a computer. Further, the above-described functions may be realized by joint operation of the OS and application software. In this case, only the part other than the OS may be stored and distributed in a medium, or may be downloaded to a computer.
  • Recording media for recording the above programs include USB memory, flexible disk, CD, DVD, Blu-ray (registered trademark), MO, SD card, Memory Stick (registered trademark), magnetic disk, optical disk, magneto-optical disk, Computer-readable recording media including semiconductor memory and magnetic tape can be used.
  • the present invention can be used for a demand prediction device, a demand prediction system, a demand prediction method, and a program.

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Strategic Management (AREA)
  • Finance (AREA)
  • Game Theory and Decision Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

L'invention concerne un dispositif (10) de prévision de demande, comportant une unité (110) de calcul de valeur seuil, une unité (120) d'agrégation d'enregistrements d'expédition, une unité (130) de calcul de taux d'accroissement, et une unité (140) de calcul de valeur prévisionnelle de demande. L'unité (110) de calcul de valeur seuil calcule une valeur seuil d'après des quantités d'expédition passées d'un produit. D'après la valeur seuil, l'unité (120) d'agrégation d'enregistrements d'expédition trie des commandes concernant le produit, antérieures à une date à laquelle une prévision de demande est réalisée, en commandes ordinaires dans lesquelles les quantités d'expédition sont inférieures à la valeur seuil et en grosses commandes dans lesquelles les quantités d'expédition sont supérieures ou égales à la valeur seuil, et agrège celles-ci. D'après les quantités d'expédition passées du produit, l'unité (130) de calcul de taux d'accroissement calcule un taux d'accroissement qui est un taux d'augmentation des quantités d'expédition de la date à laquelle la prévision de demande est réalisée jusqu'à une date pour laquelle la prévision de demande est réalisée. L'unité (140) de calcul de valeur prévisionnelle de demande calcule une valeur prévisionnelle de commandes ordinaires à partir du taux d'accroissement et des quantités d'expédition du produit basées sur les commandes ordinaires, et calcule une valeur prévisionnelle de grosses commandes à partir des quantités d'expédition du produit basées sur les grosses commandes.
PCT/JP2017/042375 2017-01-27 2017-11-27 Dispositif de prévision de demande, système de prévision de demande, procédé de prévision de demande, et programme Ceased WO2018139029A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2018564124A JPWO2018139029A1 (ja) 2017-01-27 2017-11-27 需要予測装置、需要予測システム、需要予測方法及びプログラム

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2017-012946 2017-01-27
JP2017012946 2017-01-27

Publications (1)

Publication Number Publication Date
WO2018139029A1 true WO2018139029A1 (fr) 2018-08-02

Family

ID=62978278

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2017/042375 Ceased WO2018139029A1 (fr) 2017-01-27 2017-11-27 Dispositif de prévision de demande, système de prévision de demande, procédé de prévision de demande, et programme

Country Status (2)

Country Link
JP (1) JPWO2018139029A1 (fr)
WO (1) WO2018139029A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2021026402A (ja) * 2019-08-01 2021-02-22 富士通株式会社 情報処理装置、及び、情報処理プログラム

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004070633A (ja) * 2002-08-06 2004-03-04 Seiko Epson Corp 計画立案支援システム及び計画立案支援方法
JP2004161479A (ja) * 2002-11-15 2004-06-10 Shufu No Tomo Tosho Kk ピッキングリスト作成システム、ピッキングリスト作成方法及びピッキングリスト作成プログラム
JP2012108741A (ja) * 2010-11-18 2012-06-07 Mitsubishi Heavy Ind Ltd 需要予測システム
JP2015139283A (ja) * 2014-01-22 2015-07-30 国立大学法人名古屋大学 電力需要ピーク予測装置、電力需要ピーク予測方法、及び電力需要ピーク予測プログラム

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003242432A (ja) * 2002-02-14 2003-08-29 Ns Solutions Corp 需要予測装置、方法、コンピュータプログラム、及びコンピュータ読み取り可能な記録媒体

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004070633A (ja) * 2002-08-06 2004-03-04 Seiko Epson Corp 計画立案支援システム及び計画立案支援方法
JP2004161479A (ja) * 2002-11-15 2004-06-10 Shufu No Tomo Tosho Kk ピッキングリスト作成システム、ピッキングリスト作成方法及びピッキングリスト作成プログラム
JP2012108741A (ja) * 2010-11-18 2012-06-07 Mitsubishi Heavy Ind Ltd 需要予測システム
JP2015139283A (ja) * 2014-01-22 2015-07-30 国立大学法人名古屋大学 電力需要ピーク予測装置、電力需要ピーク予測方法、及び電力需要ピーク予測プログラム

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2021026402A (ja) * 2019-08-01 2021-02-22 富士通株式会社 情報処理装置、及び、情報処理プログラム

Also Published As

Publication number Publication date
JPWO2018139029A1 (ja) 2019-06-27

Similar Documents

Publication Publication Date Title
US7921061B2 (en) System and method for simultaneous price optimization and asset allocation to maximize manufacturing profits
US11301794B2 (en) Machine for labor optimization for efficient shipping
CN114926305B (zh) 电力物资管控方法、装置、计算机设备、存储介质
JP2018063598A (ja) 業務支援システム、および、業務支援方法
US11238482B1 (en) Method and system for managing clearance items
CN112215546B (zh) 一种对象页面生成方法、装置、设备及存储介质
CN109447355A (zh) 仓库货物的配送优化方法、装置、介质和计算设备
WO2025195484A1 (fr) Procédé et système de génération de commande de réapprovisionnement et de planification de plan de réapprovisionnement axés sur la demande
US20130103607A1 (en) Determination of Projected Carrier Assignment
CN114529086A (zh) 药品数据处理方法、装置及设备
CN116128408A (zh) 智能补货方法、系统以及计算机可读介质
JP2005015140A (ja) 発注量算出システム
JP2017072870A (ja) 損益予測装置及び損益予測プログラム
JP2011145960A (ja) 商品按分管理装置，商品按分管理プログラム
JP6193817B2 (ja) 需要予測装置およびプログラム
US12079828B2 (en) Multi-channel demand planning for inventory planning and control
CN114287018A (zh) 采购业务辅助系统、采购业务辅助装置以及采购业务辅助方法
CN114626660A (zh) 用于激增调整预报的方法和装置
JP2013131259A (ja) 統合需要予測装置、統合需要予測方法、及び統合需要予測プログラム
JP5847137B2 (ja) 需要予測装置及びプログラム
US11934365B1 (en) Autonomous supply chain data hub and platform
WO2018139029A1 (fr) Dispositif de prévision de demande, système de prévision de demande, procédé de prévision de demande, et programme
JP7579121B2 (ja) 在庫計画システム及び在庫計画方法
JP2020119029A (ja) 発注情報計算プログラム、装置、及び方法
JP6276655B2 (ja) 需要予測装置およびプログラム

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17894186

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2018564124

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17894186

Country of ref document: EP

Kind code of ref document: A1