JP2000231588A - Demand predictive system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To adjust a production plan for members constituting a device in a demand predictive stage. SOLUTION: A business information receiving means 21 fetches predictive information from business section through an input device 1 and stores it in a database 30. A predictive model producing means 22 produces a predictive model on the basis of the predictive information. A device developing means 23 decomposes a device, etc., to be used for the predictive model to the members of a minimum unit. A monthly production quantity accumulating means 24 performs monthly accumulation according to a delivery period to a customer of the predictive information for every decomposed member. A demand predictive situation adjusting means 25 compares the production quantity of monthly produceable members with predictive quantity of accumulated members, subsequently adjusts the predictive model, executes demand predictive simulation and shows results on an input-output device 4. An order acceptance determination receiving means 26 fetches firm order information and a predictive model difference adjusting means 27 analyzes the difference between the firm order information and demand prediction due to the predictive model and corrects the predictive model.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a demand forecasting system for predicting demand from a customer in build-to-order manufacturing.

[0002]

2. Description of the Related Art Conventionally, for example, telephone exchanges are made to order, but since there are many special parts that require a long manufacturing period of, for example, six months or more, production starts after receiving an order. Therefore, there is a problem that the delivery date is significantly delayed. Therefore, it is necessary to procure various components in advance by predicting demand before accepting an order and drafting a production plan of various components.

[0003]

The above-mentioned demand forecasting is performed based on empirical rules or intuition.
In addition to the fact that it is difficult to stably predict demand with high accuracy for actually determined orders, it is also difficult to formulate and adjust the production plan for each required component, for example. was there. The present invention has been made in view of the above circumstances, in the demand forecast in build-to-order manufacturing,
It is an object of the present invention to provide a demand forecasting system capable of adjusting a production plan of members constituting a target device in a demand forecasting stage.

[0004]

In order to solve the above-mentioned problems and to achieve the above object, a demand forecasting system according to the present invention is a demand forecasting system for simulating demand based on a forecasting model. And sales information receiving means for capturing forecast information including sales order accuracy, model name, number, amount, delivery date, etc. of a customer from a business, and a forecast model creating means for creating a forecast model based on the forecast information. A device deployment unit for disassembling a device targeted by the prediction model into members of a minimum unit constituting the device;
Reference is made to a monthly production volume accumulating means for summing up the predicted quantity of the component for each month by the delivery date to the customer in the prediction information, and for each component, and referring to the production quantity of the component that can be produced monthly. A simulation is performed on the predicted quantity of the member using the prediction information as a variable, and the predicted quantity and the production quantity of the member are compared and displayed based on the simulation result, thereby displaying the production plan of the member. And a demand forecast situation adjusting means for adjusting the demand forecasting situation.

[0005] In the demand forecasting system having the above configuration, a demand simulation is performed by a forecasting model based on forecasting information such as the order accuracy of a customer, the model name, the number of units, the amount of money, the delivery date and the like input from the business. At
For example, it is possible to adjust the monthly production plan of each component constituting the device to be delivered to the customer by, for example, changing the delivery date of delivery to the customer. Based on the above, for example, it is possible to adjust the delivery date and propose the model of the device to the customer.

The demand forecasting system according to claim 2, further comprising: an order confirmation accepting means for receiving order confirmation information; analyzing a difference between the order confirmation information and a demand forecast by the prediction model; And a prediction model difference adjusting means for correcting the model.

[0007] In the demand forecasting system having the above-described configuration, the demand forecast by the forecast model created on the basis of the forecast information is compared with the order confirmation information. Thus, the prediction accuracy of the prediction model can be improved, and the reliability of the simulation result of the demand prediction using the prediction model can be improved.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of a demand forecasting system according to the present invention will be described below with reference to the accompanying drawings.
FIG. 1 is a block diagram of a demand forecasting system according to an embodiment of the present invention. The demand forecasting system according to the present embodiment is used, for example, for demand forecasting of a telephone exchange, and as shown in FIG. 1, an input device 1, a data processing device 2, a storage device 3, and an input / output device. And an apparatus 4. The data processing device 2 includes a business information receiving unit 2
1, predicting means creating means 22, device developing means 23,
Monthly production volume accumulation means 24 and demand forecast situation adjustment means 25
, An order confirmation accepting unit 26 and a prediction model difference adjusting unit 27. The storage device 3 has a database 30.

The sales information receiving means 21 receives forecast information from the sales via the input device 1, for example, the order accuracy of the customer, country name, project name, predicted property name, model name, number of lines (number of telephone subscribers). The number, sales representative, estimated amount, delivery date and the like are fetched and stored in the database 30. The prediction model creation unit 22 searches the database 30 for a prediction model estimated in the past based on the prediction information. If the search fails and the prediction model estimated in the past cannot be found, a new prediction model is created and stored in the database 30. The device expanding means 23 expands the members of the device used in the prediction model, that is, decomposes the members into the minimum unit constituting the device. Monthly production volume accumulation means 24
Collects the developed members on a monthly basis according to the delivery date of the forecast information to the prospective customer, and stacks, that is, totals, the members. The demand forecasting situation adjusting means 25 compares the quantity of components that can be produced each month with the number of stacked components, and then simulates demand forecasting by, for example, changing the delivery date to the customer in the forecasting model. Is executed, and the result of the simulation is displayed on the input / output device 4 by, for example, graphing. Note that the simulation result can be modified as needed. The order confirmation accepting means 26 takes in the order confirmation information via the input device 1 and stores it in the database 30. The prediction model difference adjusting unit 27 corrects the prediction model by analyzing the difference between the order confirmation information and the demand prediction by the prediction model.

The demand forecasting system according to the present embodiment has the above-described configuration. Next, the operation of the demand forecasting system will be described with reference to FIGS. FIG. 2 is a flowchart showing the operation of the demand forecasting system,
FIG. 3 is a diagram showing the operation of the demand forecasting system. First,
The business information receiving means 21 is activated. The input device 1 has, for example, an interface function for data linking the sales information from the sales system. When the sales information is linked from the sales system, the sales information reception program starts, and the prediction information is selected from the sales information. For example, customer order accuracy, country name, project name, forecast property name, model name,
Information such as the number of lines, the number of units, the sales representative, the estimated amount, and the delivery date is stored in the sales information file 31 (step S1).
At the same time, as shown in FIG. 3, when the order confirmation accepting means 26 is activated and the order confirmation information exists in the sales information, the order confirmation information is selected and, for example, the country name of the customer, the project name , Forecasted property name, model name, number of lines,
The number of devices, device specifications, and the like are stored in the order confirmation information file 40 (step T1 in FIG. 3).

After the link of the sales information is completed, the predictive model creating means 22 is started. Here, first, a program for selecting the order accuracy is activated and the sales information file 31 is selected.
, Prediction information having a predetermined order receiving probability is selected (step S2). It should be noted that the selected prediction information is provided with a selected signature so that the prediction information is excluded from the target at the time of the next selection of the order accuracy. The selected prediction information is sorted based on the customer's delivery date, model, etc. (Step S
3). If the model name of the prediction information is determined (step S
(YES side of No. 4), a model model selection program is started, and a model model having the same model name as the model name in the prediction information is extracted from the model model file 32 (step S5). Here, the selected model model is a prediction model.
If the model name of the prediction information is not determined (step S4
NO side), whether or not the number of lines of the prediction information has been determined is determined by determining the number of lines (step S6). The system starts up, searches the price model file 33 for a price model of the same amount as or close to the expected price, and sets a model corresponding to the selected price model as a standard model (step S7). Here, the selected money amount model is a prediction model. On the other hand, when the number of lines is determined (YES in step S6), it is determined by the number determination program whether the number of prediction information is determined (step S8), and the number must be determined. If it is (NO in step S8), a program for calculating the number of units is started and, for example, how many selected models are required in a range of an appropriate number of lines is calculated (step S8).
9).

After the number is determined, an option for the country name of the forecast information, for example, an appropriate option corresponding to the special circumstances of the country of delivery is selected from the country option file 34 and added to the forecast model (step S10). Next, a program for selecting a prediction model is started, and it is checked whether the created prediction model already exists in the prediction model file 35. If the prediction model does not exist (step S1)
(NO side of No. 2), the prediction model is stored in the prediction model file 35 (step S13). Thereafter, a case creation program is activated, and the case model created by combining the prediction model and the prediction information is stored in the case model file 36 (step S14).

Next, the device developing means 23 is activated. Here, first, as shown in FIG. 3, a program for device development is started, and based on, for example, the device number described in the prediction model of the case model file 36, the device configuration file 37 All the members constituting the corresponding device are extracted (step S15). After that, the monthly production volume stacking means 24 is activated, and the customer's delivery date is added to each of the developed members, and the monthly total is counted. This process is performed for all the predicted property names stored in the case model file 36, and the totaling result is stored in the stacking status file 38 (step S16). Next, the demand forecasting state adjusting means 25 is activated, and the number of stacked members for each month is compared with the number of members that can be produced for each month (production quantity) acquired from the production capacity file 39. If the number of components that can be produced exceeds the number of components that can be produced, for example, by adjusting the delivery date before or after the customer, the number of components accumulated in each month is equal to each other, that is, the load applied to monthly production is averaged. To be equal to the forecast quantity. Then, for example, a graph or the like in which the predicted quantity and the production quantity are compared by country, month, customer, model, etc., is displayed on the input / output device 4 (step S).
17).

The forecast quantity created as described above can be adjusted for the delivery date in accordance with, for example, another customer. A recalculation is performed. Further, if the order is confirmed after the above-described forecast quantity is created, the input device 1
The order confirmation information is stored in the order confirmation information file 40 via the server (step T1). When the input / output device 4 requests the adjustment of the difference between the prediction models, the prediction model difference adjustment unit 27 is activated, and the order determination information in the order determination information file 40 and the information in the stacking status file 41 and the forecast A comparison with the prediction model in the model file 35 is performed,
For example, for each predicted property name in the prediction information, the difference between the two is displayed on a display or the like of the input / output device 4. Based on the difference between the two, correction data of the prediction model is created and the prediction model is adjusted.

Demand forecasting system 10 according to the present embodiment
According to, for example, the order accuracy of the customer input from the sales,
At the stage where demand forecast is simulated by the forecast model based on forecast information such as model name, number of units, amount of money, delivery date, etc. It is possible to adjust the monthly production plan of each member to be performed, and it is possible to adjust the delivery date and propose the type of the device to the customer based on the demand prediction based on the simulation result. Furthermore, since the simulation result of the prediction model created based on the prediction information is compared with the order confirmation information, the difference between the two can be extracted and the prediction model can be adjusted, and the prediction accuracy of the prediction model can be adjusted. Can be improved, and the reliability of the simulation result of the demand forecast using the prediction model can be improved.

[0016]

As described above, according to the demand forecasting system of the first aspect of the present invention, at the stage where the demand forecast is simulated by the forecast model,
It is possible to adjust the monthly production plan of each component constituting the equipment to be delivered to the customer, and based on the simulation results of the demand forecast, for example, adjustment of the delivery date and proposal of the equipment model etc. can be made to the customer. It becomes possible. Furthermore, according to the demand forecasting system according to claim 2, a simulation result by a forecasting model created based on forecasting information is:
Since the comparison with order confirmation information is performed, the difference between the two can be extracted and the forecast model can be adjusted, the forecast accuracy of the forecast model can be improved, and the simulation results of the demand forecast using the forecast model can be obtained. Reliability can be improved.

[Brief description of the drawings]

FIG. 1 is a block configuration diagram of a demand forecasting system according to an embodiment of the present invention.

FIG. 2 is a flowchart showing the operation of the demand prediction system.

FIG. 3 is a diagram showing the operation of the demand forecasting system.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Input device 2 Data processing device 3 Storage device 4 Input / output device 21 Sales information receiving means 22 Prediction model creation means 23 Device development means 24 Monthly production volume accumulation means 25 Demand forecast situation adjustment means 26 Order confirmation acceptance means 27 Prediction model difference Adjustment means 30 Database

Claims (2)

[Claims] 1. A demand forecasting system for simulating demand based on a forecasting model, comprising: sales information receiving means for taking in forecast information including sales order accuracy, model name, number of units, amount of money, delivery date of delivery, etc. A prediction model creation unit that creates a prediction model based on the prediction information, an apparatus deployment unit that disassembles an apparatus targeted by the prediction model into members of a minimum unit that constitutes the apparatus, Forecasted quantity, monthly by the delivery date to the customer in the forecast information, and monthly production volume accumulating means to aggregate for each member, and referring to the production quantity of the member that can be produced every month, A simulation is performed on the predicted quantity of the member with the prediction information as a variable, and the predicted quantity and the production quantity of the member are calculated based on the simulation result. Demand forecasting system, characterized in that it comprises a forecast status adjusting means for adjusting the production plan of the member by compare to display. 2. An order confirmation accepting unit for receiving order confirmation information, a prediction model difference adjusting unit for analyzing a difference between the order confirmation information and the simulation result based on the prediction model, and correcting the prediction model. The demand forecasting system according to claim 1, further comprising: