JP2001167081A - Ordered quantity predicting device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately predict a quantity to be ordered even concerning an item of articles having the considerable dispersion of daily shipping by predicting the quantity to be ordered within a predictive period after considering the dispersion of shipping. SOLUTION: This device is provided with a receiving part 11 for receiving the sales information of shipping from an agency to a retail shop and receiving stock information such as the stock quantity of the agency, the number of ordered articles and the number of articles to be supplied, a data base 12 for storing the received sales information and stock information as a data base and a predictive ordered quantity calculating part 13 for calculating the predictive quantity to be ordered by the item of articles from the sales information and stock information stored in this data base 12 and this predictive ordered quantity calculating part 13 calculates the predictive quantity to be ordered within the predictive period while considering the dispersion of shipping quantities, which are contained in the sales information, for a fixed period in the past concerning the preset prescribed predictive period and calculates the predictive quantity to be ordered within the predictive period on the basis of this predictive shipping quantity and the quantity of stock, the number of ordered articles and the number of articles to be supplied, which are contained in the stock information.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an order quantity predicting apparatus which is applied to a distribution system in which products are sequentially shipped from a manufacturer to a retail store via one or a plurality of agents, and which is based on a prospective production system.

[0002]

2. Description of the Related Art In a conventional production planning method based on expected production, a production quantity is determined for each product number by predicting an order quantity (order quantity) from an agent ordered by an order point management method. There is a need to. In other words, the distributor sets the ordering point and replenishment point in advance for each part number, and when the stock quantity of the target part number falls below the ordering point, determines the order quantity so that the stock amount reaches the replenishment point and To order.

For example, when the stock amount, order point and replenishment point of a certain agency are as shown in Table 1, the order amount to the manufacturer (manufacturer) is as shown in Table 2.

[0004]

[Table 1]

[0005]

[Table 2]

That is, the stock number (AAA) is ordered to the maker A because the stock quantity (50) is less than the order point (150), and the order quantity is as follows: replenishment point (200) -stock quantity (50) = 150. Become. The product number BBB is similarly calculated, and the order quantity is 100. In addition, since the stock quantity is larger than the order point, no order is placed for the part number CCC, and the order quantity becomes zero. FIG. 4 is a graph showing the relationship between the change of the stock amount and the order point.

[0007] In this way, it is necessary for the manufacturer to predict the order quantity (order quantity) for each product number ordered from the agent, and to formulate a production plan.

In this case, in the conventional order prediction method, when estimating the stock amount of the agency when a predetermined prediction period has elapsed, the shipping amount is predicted by multiplying the past average shipment amount by the prediction period, The predicted shipping amount is subtracted from the stock amount to calculate the predicted stock amount. Then, when the predicted inventory amount is lower than the order point, a method of subtracting the predicted inventory amount from the replenishment point and calculating the predicted order amount is adopted. In other words, it is assumed that shipments will be made at the same pace as the past average.

[0009]

However, in such a conventional order forecasting method, a product number whose daily shipment is relatively stable can be accurately predicted, but a variation in the daily shipment may occur. For product numbers that tend to be large,
If a large number of shipments occur suddenly, the quantity of orders will exceed the forecast, production will be out of stock in time, and there will be problems such as a drop in CS and loss of opportunity. In other words, in the conventional order prediction method, for a product number that tends to have a large variation in daily shipment, the prediction accuracy is reduced. Therefore, the manufacturer has to stock up on the product number to prevent shortage. It was the current situation that did not get.

On the other hand, there has been proposed a conventional technique for predicting the future shipment from the shipment results of the previous year and this year (see Japanese Patent Application Laid-Open No. 4-364574). In this conventional technique, a shipment forecast within a predetermined forecast period is not averaged in the past, but a past maximum shipment quantity is searched, thereby making it possible to cope with a product number having a large variation in shipment. . However, also in this conventional technique, since the shipment is predicted based on the past maximum shipment amount, it is inevitable that the stock amount at the manufacturer increases.

The present invention has been made in order to solve such a problem, and an object of the present invention is to reduce the amount of orders received during a forecast period, taking into account the dispersion of shipments from agents to retailers. An object of the present invention is to provide an order quantity predicting apparatus capable of accurately predicting an order quantity even for a product number that tends to have a large variation in daily shipment by predicting the product quantity.

[0012]

In order to solve the above-mentioned problems, an order quantity predicting apparatus according to the present invention is an order quantity predicting apparatus for predicting an order received from an agent ordered by an order point management system, Receive sales information shipped from the agency to the retailer, as well as inventory, reorder points,
Receiving means for receiving stock information such as replenishment points; storage means for storing the received sales information and stock information as a database; and order receiving for calculating a forecasted order quantity for each product number based on the sales information and stock information stored in the database. A forecasted quantity calculating means, the order forecasted quantity calculating means for the predetermined forecasting period being set within the forecasting period in consideration of the variation in the shipping amount in the past certain period included in the sales information. Is calculated, and the estimated shipping amount and the inventory amount, ordering point,
It is characterized in that a predicted order quantity within the prediction period is calculated based on the replenishment point.

Further, in the order quantity predicting apparatus according to the present invention, in the above configuration, the order quantity predicting means calculates a shipment quantity for each of the categories while classifying a predetermined period in the past for each predetermined number of days. The maximum value among the shipment amounts for each of the divided sections is set as the estimated shipment amount in the estimation period.

Further, in the order quantity predicting apparatus of the present invention, in the above configuration, the order quantity predicting means may sequentially divide a predetermined period in the past by a predetermined number of days and one day at a time. Is determined, and the maximum value of the determined shipment amounts for each category is used as the estimated shipment amount in the estimation period.

[0015]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 3 is an explanatory view schematically showing a distribution system to which the order quantity forecasting apparatus of the present invention is applied.

A product distribution system to which the order quantity forecasting apparatus of the present invention is applied is a manufacturer (factory) A that is a manufacturer.
From one or more distributors B, B ...
One or a plurality of retail stores C, C.
The system is designed so that products are shipped to general consumers through retail stores C, C... The flow of such a product is shown by a solid line arrow in the figure. on the other hand,
The flow of the order of the product (order) is completely opposite to the flow, and is indicated by a broken arrow in the figure.

FIG. 2 shows an example of a communication system constructed to realize the distribution system shown in FIG.

That is, in order to receive sales information to the retail stores C, C... From the agents B, B. The host computer A1 of the maker A is connected to a communication network L such as a VAN or a public line. Then, the order quantity prediction device 1 of the present invention is connected to the host computer A1 of the manufacturer A.

FIG. 1 is a functional block diagram showing the system configuration of the order quantity predicting apparatus 1.

The order quantity predicting apparatus 1 receives sales information (in this case, information on the shipping amount) of the agents B, B... From the host computer A1, and inventory information (in this case, the inventory amount, order point, replenishment point). Receiving unit 11, a data storage unit (database) 12 for storing the received sales information and stock information, and an order for each product number based on the sales information and stock information stored in the database 12. A calculation unit 13 for calculating a predicted amount.

In the agents B, B,...
.. Periodically transmit sales information and stock information to maker A. The transmission cycle is a cycle in which the order quantity prediction device 1 of the manufacturer A performs order quantity prediction (for example, 4
Daily cycle, etc.), but it is desirable to transmit every day if possible.

The manufacturer A stores the sales information and the stock information received from the agents B, B,. At this time, the data items include, for example, [agency code] [part number] [sales (shipping) date] [sales (shipment) quantity] as sales information. The stock information includes, for example, [agency code] [part number] [stock date (transmission date)] [stock quantity (stock quantity)] [order point] [supplement point].

The calculating unit 13 calculates the predicted order quantity for each product number from the sales information and the stock information stored in the database 12 during the order quantity prediction processing. That is, for a predetermined prediction period (for example, one week or the like) set in advance, a certain past period (for example,
(For example, one month) to calculate the estimated shipping amount within the forecasting period, taking into account the estimated shipping amount and the inventory amount, ordering point, and replenishment point included in the inventory information. Calculate the forecasted order quantity within the forecast period. Here, it is desirable that the predetermined prediction period is determined as a parameter, taking into account the production capacity of the factory and the characteristics of the product. As a result, even for a product number having a large variation in shipment, the accuracy of the order forecast can be improved by predicting the order quantity in consideration of the sudden shipment.

Next, a specific example of the order quantity prediction processing in the calculation unit 13 will be described below.

[Specific Example 1] Agents B, B ... ship (sell) products based on orders from retailers C, C .... The agents B, B,... Have product inventory and respond to orders. When the product inventory falls below a certain amount (order point), the order (order amount) is transmitted to the maker A so that the replenishment point is satisfied, and inventory replenishment is performed.

The agent B periodically (daily) totals sales information (shipment amount information) to the retail store and sends it to the maker A, and also sends inventory information including order reference information to the maker A. Send. As described above, the format example of the data to be transmitted is [agency code] [part number] [sales (shipment) date] [sales (shipment) quantity] for sales information, and [agency] for inventory information. Code] [article number] [stock date (send date)] [stock quantity (stock quantity)] [order point] [supplement point]. However, these are the minimum necessary information, and other additional information can be added.

Here, assuming that a predetermined prediction period is L, the prediction period L is a parameter to be determined in advance in consideration of the production capacity of the manufacturer A and the lead time (LT). Usually, the number of days after the order quantity prediction processing date is such that the product can be shipped as a finished product (for example, the procurement lead time of members + production lead time + α). In this specific example 1, the member adjustment lead time is set to 2
Assuming that the day and the production lead time are 1 day and α is 1 day, the prediction period L = 4 days.

As another parameter, a past shipping period (constant period) M to be predicted is determined.
This parameter is determined in consideration of the life cycle of the product, the degree of variation in shipment, and the above-described prediction period L. In the first specific example, the prediction period M is set to 20 days.

Based on the above conditions, the following product numbers AA
A specific calculation example of the order quantity prediction processing will be described using A and BBB as models.

First, it is assumed that the order-amount prediction processing date is September 1st. Table 3 is transmitted from the agency B and is stored in the data storage unit 12.
, The stock information (stock quantity, order point, replenishment point) of the product numbers AAA and BBB stored in Table 4 shows the shipping information of the product number AAA for the past 20 days (forecast period M), and Table 5 shows the product number BBB Of the past 20 days (forecast period M).

[0032]

[Table 3]

[0033]

[Table 4]

[0034]

[Table 5]

As can be seen from Tables 4 and 5, the product number A
AA is modeled on a product number whose shipping is relatively stable, and model number BBB is a model whose shipping variation is large. However, the stock amount and the average shipment amount in the past 20 days are the same.

Based on the shipping information shown in Tables 4 and 5, the calculation unit 13 calculates the order amount prediction processing date September 1
Back calculation from the date, each part number A for each prediction period L (4 days)
The total of the shipment amounts of AA and BBB is calculated. Tables 6 and 7 show the calculation results. However, if there is a remainder in the number of days (that is, the remainder of M / L), it is truncated, but in this example, there is no remainder.

[0037]

[Table 6]

[0038]

[Table 7]

That is, for the product number AAA, 8/2
The total shipment volume during the period from 8 to 8/31 is 160, from 8/24
The total shipment volume for the period of 8/27 is 154, 8 / 20-8 /
Total shipment volume in 23 periods is 165, 8/16 to 8/19
Is 163, and the total shipment volume in the period from 8/12 to 8/15 is 158. Therefore, the maximum value 165 is set as the predicted shipment amount in the prediction period L.

For the product number BBB,
The total shipment volume during the period of August 31 was 300, August 24 to August 8
The total shipment volume during the period 27 is 0, the total shipment volume during the period 8/20 to 8/23 is 500, and the total shipment volume during the period 8/16 to 8/19 is 0, 8/12 to 8/15. Since the total shipment volume in the period is 0, the maximum value of 500 is set as the predicted shipment volume in the prediction period L.

After the estimated shipping amount is calculated in this manner, the estimated stock amount of each of the product numbers AAA and BBB at the agency B after the estimation period L (four days later) is calculated.

That is, for the product number AAA, the estimated inventory amount = the current inventory amount (350) −the estimated shipment amount (165) = 1
It becomes 85.

For the product number BBB, the estimated inventory amount = current warehouse amount (350) −estimated shipment amount (500) = − 15
It becomes 0.

After the predicted inventory amount is calculated in this way, the calculated predicted inventory amount is compared with the order point.
If the predicted inventory amount> the order point, the predicted order amount is set to 0.
If the predicted inventory amount ≦ the ordering point, the replenishment point−the predicted inventory amount is calculated as the predicted order amount.

That is, in the case of the product number AAA, the predicted inventory quantity (185) ≦ the order point (300), the predicted order quantity = the replenishment point (500) −the predicted inventory quantity (185) = 310.
Becomes

In the case of the product number BBB, the predicted inventory amount (−150) ≦ the order point (300), the predicted order amount = the replenishment point (500) −the predicted inventory amount (−150) = 65.
It becomes 0.

Here, for comparison with the above results, a calculation example in the case where the order quantity is predicted using the conventional prediction method will be described below.

[Conventional example] In the method of the conventional example, first, for each of the product numbers AAA and BBB, the past 20 days (forecast period M)
Find the average shipment volume at From Tables 4 and 5, it can be seen that the average shipment quantity of the product number AAA is 40, and the average shipment quantity of the product number BBB is also 40. Next, a value obtained by multiplying the average shipment amount by the prediction period L (4 days) is set as a shipment prediction amount in the prediction period L. That is, each part number AAA, BBB
In both cases, the estimated shipment amount = the average shipment amount (40) × the estimation period L
(4th) = 160. Next, after the forecast period (after 4 days)
Calculates the estimated inventory amount of the agency B. That is, for each of the product numbers AAA and BBB, the predicted inventory amount = the current warehouse amount (35)
0) -estimated shipping quantity (160) = 190. Next, the predicted inventory amount is compared with the ordering point of each of the product numbers AAA and BBB.
If the predicted inventory amount ≦ the order point, the replenishment point−the predicted inventory amount is calculated as the predicted order amount. That is, each part number AAA,
For both BBB, predicted inventory quantity (190) ≦ order point (30
0), the predicted order quantity = the replenishment point (500) −the predicted inventory quantity (190) = 310.

Comparing this conventional example with the above specific example 1, there is no significant difference between the specific example 1 and the conventional example with respect to the product number AAA whose shipping volume is relatively stable (in this case, the same). However, for the product number BBB having a large variation in shipment, there is a clear difference in the predicted order amount. In the first specific example, since the forecasted order quantity is determined after the peak of the shipment is determined, it is possible to cope with a sudden change in the shipment. In addition, the order quantity prediction device 1 of the present invention has an advantage that it can be applied to a part number whose shipment is relatively stable, as is clear from the specific example 1, and that it has excellent versatility. ing.

In the above specific example 1, in the past prediction period M
(20 days) is simply divided for each prediction period L (4 days). On the other hand, in the specific example 2 described below, the way of division according to the prediction period L is different.

[Specific Example 2] In this specific example 2 as well, the stock information (stock quantity, order point, replenishment point) of the product numbers AAA and BBB stored in the data storage unit 12 and the past 20 days of the product number AAA (forecast) Shipping information for period M) and part number BB
The shipping information of B for the past 20 days (predicted period M) is the same as Table 3, Table 4, and Table 5 shown in the above specific example 1, so the information of Table 3, Table 4, and Table 5 is also used here. Will be explained.

Based on the shipping information shown in Tables 4 and 5, the calculation unit 13 calculates each part number AAA for each prediction period L (four days) shifted by one day from August 12, which is the data start date. , BBB is calculated. That is, in the specific example 2, each prediction period L (4
Day) is calculated. Tables 8 and 9 show the calculation results.

[0053]

[Table 8]

[0054]

[Table 9]

That is, for the product number AAA, 8/1
The total shipment volume for the period from 2 to 8/15 is 158, and the total shipment volume for the period from 8/13 to 8/16, which is shifted by one day, is 158, and this period is shifted by one day, 8/14 to 8 / The total shipment volume for the 17 period is 157, ..., last 8 / 28-
Since the total shipment volume in the period of 8/31 is 160, the maximum value 165 (period of 8/20 to 8/23) is set as the predicted shipment amount in the prediction period.

For the product number BBB,
The total shipment volume in the period of 8/15 is 0, the total shipment volume in the period of 8/13 to 8/16, which is one day shifted from this period, is 0, ...
・ The total shipment volume during the period from August 18 to August 21 was 200,
The total shipment volume during the period from August 19 to August 22 was 200, August
The total shipment volume for the period from 20 to 8/23 is 500, ...,
The total shipment volume during the last 8 / 28-8 / 31 period is 300
Therefore, the maximum value is 500 (8/20 to 8 /
The period of 23 or the period of 8/21 to 8/24) is the estimated shipment amount in the estimation period.

After the predicted shipping amount is calculated in this way, the predicted stock amount of each of the product numbers AAA and BBB at the agency B after the prediction period (four days) is calculated.

That is, for the item number AAA, the predicted inventory amount = the current storage amount (350) −the predicted shipment amount (165) = 1
It becomes 85.

For the product number BBB, the estimated inventory amount = the current inventory amount (350) −the estimated shipment amount (500) = − 15.
It becomes 0.

When the estimated inventory amount is calculated in this way, the calculated estimated inventory amount is compared with the order point.
If the predicted inventory amount> the order point, the predicted order amount is set to 0.
If the predicted inventory amount ≦ the ordering point, the replenishment point−the predicted inventory amount is calculated as the predicted order amount.

That is, in the case of the product number AAA, the predicted inventory quantity (185) ≦ the order point (300), the predicted order quantity = the replenishment point (500) −the predicted inventory quantity (185) = 310.
Becomes

In the case of the product number BBB, the predicted inventory amount (−150) ≦ the order point (300), the predicted order amount = the replenishment point (500) −the predicted inventory amount (−150) = 65.
It becomes 0.

Comparing the specific example 2 with the above conventional example, there is no significant difference between the specific example 2 and the conventional example for the product number AAA whose shipping amount is relatively stable (in this case, the same is true). However, for the product number BBB having a large variation in shipment, there is a clear difference in the predicted order amount. In the specific example 2, since the forecasted order quantity is determined after the peak of the shipment is obtained, it is possible to cope with a sudden change in the shipment. In addition, the order quantity predicting apparatus 1 of the present invention has an advantage that it can be applied to a part number whose shipment is relatively stable, as is clear from the specific example 2, and is excellent in versatility. ing.

[0064]

As described above, according to the order quantity predicting apparatus of the present invention, the order quantity within the forecast period is estimated while taking into account the dispersion of the shipment from the agency to the retail store. Even for product numbers that tend to have large variations in daily shipments, it is possible to accurately predict the order quantity. In other words, it is possible to respond to sudden orders, so that it is possible to reduce the occurrence of shortage. In addition, since the same logic can be used to predict a part number whose shipment is relatively stable, it is possible to provide an order quantity prediction device having excellent versatility.

[Brief description of the drawings]

FIG. 1 is a functional block diagram showing a system configuration of an order quantity prediction device.

FIG. 2 is an explanatory diagram showing an example of a communication system constructed to realize the distribution system shown in FIG.

FIG. 3 is an explanatory diagram schematically showing a distribution system to which the order quantity prediction device of the present invention is applied.

FIG. 4 is a graph showing a relationship between a change in stock quantity and an order point.

[Explanation of symbols]

 DESCRIPTION OF REFERENCE NUMERALS 1 order receiving device 11 receiving unit 12 database (data storage unit) 13 calculating unit (order receiving predicting amount calculating means)

Claims (3)

[Claims] An order quantity forecasting apparatus for predicting an order received from an agent ordered by an order point management system, comprising: receiving sales information shipped from an agent to a retailer; Receiving means for receiving inventory information such as stock quantity, order points, replenishment points, storage means for storing the received sales information and stock information as a database, and, based on the sales information and stock information stored in this database, by product number. Means for calculating a forecasted order quantity, wherein the forecasted order quantity calculation means is configured to calculate, based on a predetermined forecast period set in advance, a variation in a shipment amount in a past fixed period included in the sales information. Is calculated in consideration of the estimated shipment amount within the forecast period, and the estimated shipment amount within the forecast period is calculated based on the estimated shipment amount and the inventory amount, order point, and replenishment point included in the stock information. Order quantity prediction apparatus and calculates the predicted amount. 2. The predicted order quantity calculating means divides a predetermined period in the past into predetermined days, obtains a shipment amount for each of the divisions, and calculates a maximum value among the obtained shipment amounts for each of the divisions. 2. The order amount prediction device according to claim 1, wherein is a predicted shipment amount in the prediction period. 3. The predicted order receiving quantity calculating means sequentially divides a predetermined period in the past by a predetermined number of days and one day at a time, obtains a shipping quantity for each of the categories, and obtains a shipping quantity for each of the obtained categories. The order quantity estimating apparatus according to claim 1, wherein a maximum value of the shipment quantity is set as a shipment estimation quantity in the estimation period.