JP2005035723A - Order recommendation number counting system, order recommendation number counting method, and its program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve a problem for frequently causing lack of commodities and excess stocks because order and purchase of the commodities depend upon experiences and feelings while an operation using a computer spreads in a field of selling commodities so as to easily manage sales and stocks. <P>SOLUTION: An order reference data creation means applies a least squares method to sales transition data of past eight weeks by an entering/exiting management means for managing entry and exist of each commodity, an entering/exiting management table, and a table for managing sales of last eight weeks of each commodity, so as to judge whether sales goes upward or downward. A multiplier decided on the basis of the sales trend is multiplied by the previous order number, so as to decide the order recommendation number and output instruction of proper replenishment of commodities. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention compares the sales trend obtained by the method of least squares from the long-term past number of sales of the product with the short-term past sales trend when determining the number of orders for the product at the store. The present invention relates to an order recommendation number calculation system that predicts an order quantity, an order recommendation number calculation method, and a program thereof.

  The use of computers has become popular in the field of product sales, and sales and inventory can be managed easily. However, because of the ordering and purchase of merchandise, it relies on the experience and intuition of the person in charge of ordering, resulting in frequent shortages and excessive inventory.

  In order to deal with such a problem, according to Patent Document 1, a continuous time-series change in the sales performance of a product is grasped as a function based on the relative relationship with the delivery performance of the product, Reflecting weather information such as conditions, the sales trend of the product under a predetermined condition is obtained, and the remaining number of products ordered on the same day of the previous week is obtained from the sales performance of the product and the ordering and delivery results. Thus, it is described that the order quantity suitable for the current situation can be determined in consideration of the day-of-week characteristic in the sales trend of the product, and the appropriateness of the order quantity can be improved.

Japanese Patent Laid-Open No. 9-274633 (page 2-3, FIG. 1)

  In the above-mentioned Patent Document 1, it is described that the order quantity is analyzed using a trend curve representing a sales trend. How is the trend curve and what is the curve? There is no description of what to ask for and there is a lack of concreteness. In the present invention, a sales trend graph based on the least square method is specifically obtained from past sales, and the ordering for the purpose of instructing the number of orders for proper product replenishment in order to eliminate the sales loss and the excess inventory according to the graph. A recommendation number calculation system, an order recommendation number calculation method, and a program thereof are proposed.

The first order recommendation number calculation system according to the present invention is a first means for registering the status of goods in / out as data in / out of the data storage unit, and the number of goods ordered by inputting the in / out information. An order recommendation number calculation system comprising: a second means for calculating,
The second means inputs the entry / exit information, takes out the number of sales of the product in a long-term period going back from the present time, calculates the first tendency value of sales based on the number, and calculates the current tendency from the entry / exit information. Calculating a second trend value based on the number of sales in the first short-term period going back from the third trend value based on the number of sales in the second short-term period going back from the current time, Comparing the first trend value with the second trend value, comparing the first trend value with the third trend value, and calculating the order quantity of the product according to the comparison result .

A second order recommendation number calculation system according to the present invention includes a first means for registering a state of goods in / out as goods / incoming information in a data storage unit, and the order information for the goods by inputting the goods / incoming information. An order recommendation number calculation system comprising: a second means for calculating,
The second means inputs the entry / exit information, extracts the number of sales in each period obtained by dividing the long-term period of the product from the current time into equal intervals, and based on this, minimizes the first tendency value of sales. A second trend value based on the number of sales in each period obtained by dividing the first short-term period retroactive from the current entry / exit information into the same interval and the second short-term retroactive from the present time A third tendency value based on the number of sales in each period divided into equal intervals of the target period, respectively, is calculated by the least square method, and the first tendency value and the second tendency value are compared, Comparing the tendency value of 1 with the third tendency value, and calculating the number of orders of the product according to the result of the comparison.

  According to a third aspect of the present invention, in the first or second invention, the second means sets the second short-term period as a half period of the first short-term period. Calculating an order quantity of the product.

  In a fourth order recommendation number calculation system according to the present invention, in the second invention, the second means sets a period obtained by dividing the long-term period and the short-term period into one week, and the product for each week. Applying the least squares method to the number of sales.

The fifth order recommendation number calculation system according to the present invention is a first means for registering the state of goods entering / exiting in the data storage unit as entry / exit information, and inputting the entry / exit information to determine the number of orders for the goods. An order recommendation number calculation system comprising: a second means for calculating,
The second means inputs the entry / exit information, takes out the number of sales of each week in the past year that goes back from the current time about the product, calculates a first tendency value of sales based on the number of sales by the least square method, A second trend value based on the number of sales in each week for the past eight weeks from the current entry / exit information and a third trend value based on the number of sales for each week for the past four weeks from the current time, respectively. Calculate by the least square method, compare the first trend value and the second trend value, compare the first trend value and the third trend value, and calculate the multiplication factor according to the comparison result And calculating the order quantity of the product by applying the multiplication factor.

According to a sixth order recommendation number calculation system of the present invention, in the fifth invention, when the second means compares the second tendency value ≧ | first tendency value |, the first sales tendency. Is upwards,
When the second trend value ≦ − | first trend value |, the first sales trend is downward,
In other cases, the first sales trend is flat, and when the third trend value ≧ | first trend value |, the second sales trend is upward.
When the third trend value ≦ − | first trend value |, the second sales trend is downward,
Otherwise, suppose the second sales trend is flat,
Determining the multiplication factor from a combination of upward, downward, and flattening of the first sales trend and the second sales trend.

A first order recommendation number calculation method according to the present invention includes: first means for registering a state of goods entering / exiting in a data storage unit as entry / exit information; and inputting the entry / exit information to determine an order number of the goods. An order recommendation number calculation method in an order recommendation number calculation system comprising: a second means for calculating,
A first step of inputting the entry / exit information and calculating a first tendency value of sales by the method of least squares based on the number of sales in each week in a long-term period retroactive from the present time about the product;
A third tendency based on the second trend value based on the number of sales in each week in the first short-term period retroactive from the current entry / exit information and the number of sales in each week in the second short-term period retroactive from the present time A second step of calculating values by the least square method,
A comparison between the first trend value and the second trend value and a comparison between the first trend value and the third trend value are performed, and the number of orders of the product is calculated according to the comparison result. A third step,
Is provided.

The second order recommendation number calculation method according to the present invention includes a first step of registering the status of goods in / out of goods as in / out information in the data storage unit;
A second step of inputting the entry / exit information and calculating a first tendency value of sales by the method of least squares based on the number of sales in each week in a long-term period retroactive from the present time for the product;
A second trend value based on the number of sales in each week in the first short-term period retroactive from the current entry / exit information and a third value based on the number of sales in each week in the second short-term period retroactive from the present time A third step of calculating a trend value of each by a least square method;
A comparison between the first trend value and the second trend value, and a comparison between the first trend value and the third trend value are performed, respectively, and the order number of the product is calculated according to the comparison result. 4 steps,
Is provided.

The first program of the present invention is:
On the computer,
A first step of registering the status of goods in / out of goods as in / out information in the data storage unit;
A second step of calculating the first tendency value of sales by the method of least squares based on the number of sales in each week of a long-term period that goes back from the present time for the product, by inputting the entry / exit information;
A second trend value based on the number of sales in each week in the first short-term period retroactive from the current entry / exit information and a third value based on the number of sales in each week in the second short-term period retroactive from the present time And a third step of calculating the trend value of each by the least square method,
A comparison between the first trend value and the second trend value, and a comparison between the first trend value and the third trend value are performed, respectively, and the order number of the product is calculated according to the comparison result. 4 steps,
Is executed.

  The present invention relies on the experience and intuition of the person in charge of ordering, and calculates the recommended order quantity from past sales volume for replenishment orders where shortage of goods and excess inventory frequently occur. By instructing, there is an effect that any salesperson can easily replenish proper products.

  Next, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram showing a configuration of an embodiment of the present invention.

  Referring to FIG. 1, for example, an ordering system 1 that operates the present invention in a computer installed in a supermarket headquarters or the like, a store controller 2 installed in a supermarket branch, and POS installed in a plurality of sales offices in the branch. Terminal 3.

  It is assumed that the POS terminal 3 and the store controller 2 are connected by, for example, a LAN, and the store controller 2 and the ordering system 1 are connected by, for example, the Internet.

  The ordering system 1 is connected to an input device 15 such as a keyboard, a display device 16 such as a display, an output device 17 such as a printer, and a data storage unit 20 such as a magnetic disk device.

  When the sales process of the product is performed at the POS terminal 3, information such as the product type, the unit price, and the number of products sold is sent to the store controller 2 and aggregated. The store controller 2 periodically transmits this total value to the ordering system 1. In the ordering system 1, record management is performed on merchandise information that has been sold and transmitted from a plurality of store controllers 2.

  The present invention is not particularly limited to the supermarket, and the ordering system 1 may be constructed on the store controller 2 or the POS terminal 3 in a retail store or the like having only one store.

  FIG. 2 is a block diagram showing a detailed configuration of the ordering system 1.

  The ordering system 1 includes a calculation unit 10, a data storage unit 20, and the like.

  The calculation unit 10 is a computer of the main body of the ordering system 1 including a CPU, a main storage device, and the like, and includes an entry / exit management unit 11 that operates as a software program, and an order reference data creation unit 12. ing.

  The data storage unit 20 stores and stores an entry / exit management table 21, a previous 8-week sales management table 22, and order reference data 23.

  The entry / exit management table 21 records information such as the date of ordering, the number of orders, the date of receipt, the number of receipts, the sales date, the number of sales, and the like for each product for a long term of the past one year or more.

  The previous eight-week sales management table 22 records the number of sales of each product in each week from the current week to a relatively short period of time, for example, eight weeks ago. In addition, 8 weeks corresponds to about 2 months, and in the embodiment of the present invention, for example, it is an optimal period length when assuming home appliances etc., but the product type (for example, lunch box) handled by the store or the store This period varies depending on the environment, etc., and is not particularly limited to 8 weeks. In addition, 4 weeks, which will be described later, means a half period of 8 weeks.

  Further, in this embodiment, explanation is made for obtaining the number of sales predicted in the next week based on the number of sales in each past week. For example, if the number of sales in the past two weeks is used as the basis, The number of sales predicted in a week can be obtained, and the number of sales in a future period that is the same as the past equally divided period length can be obtained.

  In the order reference data 23, the number of orders in the replenishment order for each product created by the order reference data creation means 12 is calculated, output and recorded as a recommended number every day.

  The entry / exit management means 11 receives an order from the manufacturer for each product of the entire system, or, when the ordering system 1 is in a branch, receipt by transfer from another store, sale information of the product input from the POS terminal 3, and the keyboard. The delivery management is performed based on the input information such as transfer information to another store and information such as returns, and the registration is sequentially registered in the entry / exit management table 21 and the previous 8 week sales management table 22.

  The order reference data creation means 12 refers to the previous eight-week sales management table 22 when activated by the input device 15 connected to the present system when it is necessary to calculate the number of orders for a product or at a predetermined date and time. Then, the order number is calculated by applying the method of least squares to the order quantity of each week of the previous eight weeks, and the result is output to the display device 16 and the output device 17.

  Next, the operation of this embodiment will be described based on the flowchart of FIG.

  The entry / exit management means 11 causes such an event that an order is received from a manufacturer or transfer from another store for each product of the entire system, and a product is sold, transferred to another store, or returned by return, etc. Each time it is started, the contents are registered in the storage / retrieval management table 21 and the previous 8 week sales management table 22 (step S1).

  For example, the order reference data creating means 12 is automatically activated at a fixed time every day, and calculates the order recommendation number for necessary products. Calculation of the long-term sales trend value based on the information in the storage / retrieval management table 21 and the trend value (trend value) of each week's sales in the previous 8 weeks from the previous 8 week sales management table 22 1) and the calculation process of the trend value (trend value 2) of the sales for the previous 4 weeks, which is half the period of the previous 8 weeks, is started (step S2).

The order reference data creation means 12 takes out the sales figures for the first week to the Nth week from the previous eight-week sales management table 22 for the corresponding product retroactively, with the current week as the Nth week (Step). S3). Here, N means 4 or 8.
(1) The sum of the number of sales of products for each week is (A). A = (number of sales in the first week) + (number of sales in the second week) + (number of sales in the third week) to + (N weeks) Number of sales)
(2) Number of sales × sum of each week of week (B) where B = (number of sales in the first week × 1) + (number of sales in the second week × 2) + (number of sales in the third week × 3) ~ + (Number of sales in week N × N)
(3) C = (1 × 1) + (2 × 2) + (3 × 3) to + (N × N) where the sum of the squares of each week is (C)
(4) D = 1 + 2 + 3 to + N, where D is the sum of each week
Using the above A, B, C, and D, the tendency value 1 and the tendency value 2 are obtained by the following formula of the least square method (step S4).
Trend value 1 = (8B-DA) / (8C-DD)
Trend value 2 = (4B-DA) / (4C-DD)
(5) The number of sales in each week (52 weeks or 53 weeks) of the past year, which is a relatively long period, is taken out of the storage management table 21, the week is plotted on the horizontal axis (x axis), and the number of sales is plotted vertically The least square method is applied to about 52 points as the axis (y-axis), and the slope a in the calculated straight line graph y = ax + b is examined and expressed as the trend (a) (step S5). This calculation may be calculated and stored in advance for each product as the sales trend of the previous year.
(6) When the trend value 1 ≧ | trend (a) |, the sales trend 1 is upward “↑”.

When the trend value 1 ≦ − | trend (a) |, the sales trend 1 is downward “↓”.
At other times, the sales trend 1 is flat “→”.
For sales trend 2 for trend value 2, upward “↑”, downward “↓”, and flat “→” are determined in the same manner as trend value 1 (step S6).

  Next, the content described above based on FIGS. 2 and 3 will be described with a simple specific example.

  The number of sales Y in FIG. 2 (d) describes the number of sales of a certain product in the past 8 weeks. The number of sales in the first week is 10, the second week is 11,. This shows how the number of sales increases by one each time. A graph of this state is shown in FIG. 2A, and this graph is y = x + 9. If the result of applying the method of least squares to the sales transition in the past year is y = 0.8x + b, the tendency (a) of the slope a is 0.8.

  The result applied to the mathematical formula based on the least square method for obtaining the tendency value 1 based on the numerical value of FIG. 2D is described in FIG.

Further, FIG. 2E shows the contents extracted from FIG. 2D for the sales for the past four weeks. Based on this, the calculation content of the trend value 2 is shown in FIG. 2C, but this value is also 1. From the above, when applying to the trend value 1 ≧ | trend (a) |, 1 ≧ 0.8, and the sales trend 1 is upward “↑”. Similarly, sales trend 2 is “↑”.

  Next, FIG. 3 will be described.

  The number of sales Y in FIG. 3D describes the number of sales of a certain product in the past 8 weeks, the number of sales in the first week is 10, the number in the second week is 9,. The number of sales is decreasing one by one. FIG. 3A is a graph of this state. This graph is y = −x + 11, and the result of applying the least square method to the sales transition of the past one year is y = −0.8x + b. Then, the tendency (a) of the inclination a is −0.8. FIG. 3B shows a result obtained by applying the mathematical formula for obtaining the trend value 1 based on FIG. 3D to −1.

Further, FIG. 3E shows the contents extracted from FIG. 3D for the sales for the past four weeks. Based on this, the formula for calculating the trend value 2 is shown in FIG. 3C, but this value is also -1. From the above, if the trend value 1 ≦ − | trend (a) | is applied, −1 ≦ −0.8 is satisfied and the sales trend 1 is downward “↓”.
Similarly, for the trend value 2, the sales trend 2 is downward “↓”.

  Returning to the above, when the upward “↑”, downward “↓”, and flat “→” trends, which are indicators of trend values, are obtained as described above, this result is obtained from sales trend 1, sales trend 2, and multiplication factor. The corresponding multiplication factor is determined from the sales trend 1 and the sales trend 2 by applying to the table as illustrated in FIG. 4 (step S7). For example, since the first row in FIG. 4 is “↑” for both trend 1 and trend 2, this indicates that 1.3 is applied as the multiplication factor.

This multiplier is used to determine the effective order recommendation number to eliminate sales losses and excess inventory.
Number of recommended orders = {(Number of sales since the previous order)-(Number of goods received from other stores since the previous order) + (Number of goods issued to other stores since the last order)} x Multiplier This recommended number of orders and trend 1 , 2 and the like, necessary information for proper product replenishment instruction is output to the order reference data 23 and also displayed on the display device 16, and the contents are confirmed by the operator (step S8). Further, it is also output to the output device 17 according to an instruction from the operator as necessary.

  The present invention can be used for calculating the number of orders when placing an order for products to be sold in a store.

It is a block diagram which shows the structure of the Example of this invention. It is a block diagram which shows the structure of the ordering system of this invention. It is a flowchart explaining operation | movement of the Example of this invention. It is explanatory drawing which calculates the tendency (upward) used as the parameter | index of a tendency value. It is explanatory drawing which calculates the tendency (downward) used as the parameter | index of a tendency value. It is a figure which sets the multiplier required for recommendation number calculation by the tendency 1,2.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Ordering system 2 Store controller 3 POS terminal 10 Calculation part 11 Entry / exit management means 12 Order reference data creation means 15 Input device 16 Display device 17 Output device 20 Data storage part 21 Entry / exit management table 22 Previous 8 weeks sales management table 23 Ordering Reference data

Claims (9)

Order recommendation comprising: first means for registering the status of goods in and out as data in and out of the data storage section; and second means for inputting the goods in / out information and calculating the number of orders for the goods. A number calculation system,
The second means inputs the entry / exit information, takes out the number of sales of the product in a long-term period going back from the present time, calculates the first tendency value of sales based on the number, and calculates the current tendency from the entry / exit information. Calculating a second trend value based on the number of sales in the first short-term period going back from the third trend value based on the number of sales in the second short-term period going back from the current time, A comparison between the first trend value and the second trend value, a comparison between the first trend value and the third trend value, and calculation of the number of orders of the product according to the comparison result An order recommendation number calculation system. Order recommendation comprising: first means for registering the status of goods in and out as data in and out of the data storage section; and second means for inputting the goods in / out information and calculating the number of orders for the goods. A number calculation system,
The second means inputs the entry / exit information, extracts the number of sales in each period obtained by dividing the long-term period of the product from the current time into equal intervals, and based on this, minimizes the first tendency value of sales. A second trend value based on the number of sales in each period obtained by dividing the first short-term period retroactive from the current entry / exit information into the same interval and the second short-term retroactive from the present time A third tendency value based on the number of sales in each period divided into equal intervals of the target period, respectively, is calculated by the least square method, and the first tendency value and the second tendency value are compared, An order recommendation number calculation system that compares a tendency value of 1 and a third tendency value, and calculates the order number of the product according to a result of the comparison. 3. The method according to claim 1, wherein the second means calculates the number of orders for the product with the second short-term period being half the first short-term period. An order recommendation number calculation system. 3. The method of claim 2, wherein the second means applies a least square method to the number of sales of the product in each week, with a period obtained by dividing the long-term period and the short-term period at equal intervals as one week. A system for calculating the number of recommended orders. Order recommendation comprising: first means for registering the status of goods in and out as data in and out of the data storage section; and second means for inputting the goods in / out information and calculating the number of orders for the goods. A number calculation system,
The second means inputs the entry / exit information, takes out the number of sales of each week in the past year that goes back from the current time about the product, calculates a first tendency value of sales based on the number of sales by the least square method, A second trend value based on the number of sales in each week for the past eight weeks from the current entry / exit information and a third trend value based on the number of sales for each week for the past four weeks from the current time, respectively. Calculate by the least square method, compare the first trend value and the second trend value, compare the first trend value and the third trend value, and calculate the multiplication factor according to the comparison result Then, an order recommendation number calculation system that calculates the order number of the product by applying the multiplication factor. In claim 5, the second means sets the first sales tendency upward when the comparison is second tendency value ≧ | first tendency value |
When the second trend value ≦ − | first trend value |, the first sales trend is downward,
In other cases, the first sales trend is flat, and when the third trend value ≧ | first trend value |, the second sales trend is upward.
When the third trend value ≦ − | first trend value |, the second sales trend is downward,
Otherwise, suppose the second sales trend is flat,
An order recommendation number calculation system, wherein the multiplication factor is determined from a combination of upward, downward, and flattening of the first sales trend and the second sales trend. Order recommendation comprising: first means for registering the status of goods in and out as data in and out of the data storage section; and second means for inputting the goods in / out information and calculating the number of orders for the goods. An order recommendation number calculation method in a number calculation system,
A first step of inputting the entry / exit information and calculating a first tendency value of sales by the method of least squares based on the number of sales in each week in a long-term period retroactive from the present time about the product;
A third tendency based on the second trend value based on the number of sales in each week in the first short-term period retroactive from the current entry / exit information and the number of sales in each week in the second short-term period retroactive from the present time A second step of calculating values by the least square method,
A comparison between the first trend value and the second trend value and a comparison between the first trend value and the third trend value are performed, and the number of orders of the product is calculated according to the comparison result. A third step,
The order recommendation number calculating method characterized by comprising. A first step of registering the status of goods in and out as data in and out of the data storage unit;
A second step of calculating the first tendency value of sales by the method of least squares based on the number of sales in each week in a long-term period that goes back from the present time for the goods, and inputting the entry / exit information;
A second trend value based on the number of sales in each week in the first short-term period retroactive from the entry / exit information and a third value based on the number of sales in each week in the second short-term period retroactive from the present time A third step of calculating a trend value of each by a least square method;
A comparison between the first trend value and the second trend value, and a comparison between the first trend value and the third trend value are performed, respectively, and the order number of the product is calculated according to the comparison result. 4 steps,
The order recommendation number calculating method characterized by comprising. On the computer,
A first step of registering the status of goods in / out of goods as in / out information in the data storage unit;
A second step of calculating the first tendency value of sales by the method of least squares based on the number of sales in each week of a long-term period that goes back from the present time for the product, by inputting the entry / exit information;
A second trend value based on the number of sales in each week in the first short-term period retroactive from the current entry / exit information and a third value based on the number of sales in each week in the second short-term period retroactive from the present time And a third step of calculating the trend value of each by the least square method,
A comparison between the first trend value and the second trend value, and a comparison between the first trend value and the third trend value are performed, respectively, and the order number of the product is calculated according to the comparison result. 4 steps,
A program that executes

Images