JP2006039802A - Stock management system, stock management method, and stock management program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To predict orders with high precision on the basis of forecast information, thereby calculating the proper amount of orders. <P>SOLUTION: Errors between one or two previous consecutive pieces of forecast information and corresponding results of orders received are calculated for a plurality of combinations and the current forecast information is corrected, at least based on the average and the standard deviation of the plurality of errors obtained, so as to calculate the amount of errors. Since the current forecast information is not directly used but corrected using the average and the standard deviation of the errors, it is made possible to effectively suppress shortage and excess stock. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an inventory management system, an inventory management method, and an inventory management program, and in particular, an inventory management system and an inventory management method capable of calculating an appropriate order quantity based on forecast information given by a customer who is a consumer. And an inventory management program.

  In recent years, mainly in the manufacturing and distribution industries, a series of business activities ranging from material procurement and inventory management to product delivery have been integrated and managed across companies and organizations, thereby improving operational efficiency. The technique to do is attracting attention. Such an approach is generally called “Supply Chain Management (SCM)” and is said to have a great effect mainly on shortening delivery time, reducing shortage, and reducing inventory.

  In order to fully demonstrate the effects of supply chain management, close information exchange between suppliers and consumers is essential. For example, if the customer notifies the supplier of the prospective order details in the future, the supplier can make a future production plan based on this, and it is possible to suppress the occurrence of excess inventory and shortage. Become. Such information about the prospective order placement, that is, the prospective future order received from the supplier, is called “forecast information” and is essential information for the functioning of supply chain management. It can be said.

  Here, the types and characteristics of orders received from the supplier will be described. Since the consumer is “customer” from the viewpoint of the supplier, the customer is described as “customer” in the following description.

  The types of orders viewed from the supplier can be divided into two categories: “confirmed orders” and “reserved orders”. “Confirmed order” refers to an order whose delivery date and quantity are confirmed. Since the firm order is a firm order from the customer's point of view, after the supplier receives the firm order, the customer has an obligation to pick up the product. On the other hand, “reservation order” refers to an order whose quantity is fixed but whose delivery date is not yet determined. The reservation order is an order for the customer side to secure a predetermined amount of product while the delivery date is uncertain. Generally, after receiving the reservation order, the customer side is required to collect the product. In this respect, the reservation order has the same properties as the confirmed order.

On the other hand, the above-mentioned “forecast information” is not so-called orders but is expected information regarding future orders. Therefore, the delivery date and quantity are uncertain, and only the required amount for each scheduled delivery date is given. The forecast information is normally given from the customer side, but this does not cause the customer to take over the product, and is treated as prospect information for smoothly functioning the supply chain management.
JP 2003-312849 A JP 2003-316862 A JP 2004-75321 A JP-A-9-62941 JP 2000-207479 A Japanese Patent Laid-Open No. 7-239884 JP 2001-134463 A JP 11-345267 A JP 2002-12308 A Japanese Patent Laid-Open No. 10-76447

  Conventionally, various order forecasting methods have been proposed for the purpose of reducing shortage and reducing inventory, but without using forecast information, these are the demand forecast values and orders received ( In addition, since the future demand, the necessary inventory amount, and the like are predicted using a difference from the ordering result), there is a problem that the prediction accuracy is not necessarily high.

  Accordingly, an object of the present invention is to provide an inventory management system, an inventory management method, and an inventory management program capable of predicting orders with high accuracy based on forecast information, and thereby calculating an appropriate order quantity. It is.

  An inventory management system according to the present invention includes at least an input unit for inputting forecast information, a forecast storage unit for storing the forecast information, an order record storage unit for storing an order record, and the forecast storage. A processing unit that calculates the order quantity based on the forecast information stored in the section and the order record stored in the order record storage unit, and an output unit that outputs the order quantity, and the processing unit includes: The error between the past one or two or more forecast information in the past and the order record corresponding to this is calculated for a plurality of combinations, and at least the average value of the obtained errors and the standard of the obtained errors The order quantity is calculated by correcting current forecast information based on the deviation.

  In addition, the inventory management method according to the present invention includes a step of calculating an error between a past one or two or more consecutive forecast information and an order record corresponding thereto for a plurality of combinations, and at least a plurality of obtained errors. And calculating the order quantity by correcting the current forecast information based on the average value and the standard deviation of the obtained plurality of errors.

  Furthermore, the inventory management program according to the present invention is obtained at least in a step of calculating, for a plurality of combinations, an error between a past one or two or more consecutive forecast information and an order record corresponding thereto. And calculating the order quantity by correcting current forecast information based on an average value of a plurality of errors and a standard deviation of the plurality of errors obtained.

  According to the present invention, the current forecast information is not used as it is, but is corrected using the average value of errors and the standard deviation of errors. It becomes possible to suppress effectively.

  Specifically, the correction amount is calculated based on the average value of the error and the standard deviation of the error, the correction amount is added to the current forecast information, and the order quantity is calculated by subtracting the ordered quantity and the inventory quantity. It is preferable to do. In this way, if the order quantity is calculated in consideration of the ordered quantity and the inventory quantity, the order quantity can be automatically derived by repeating this operation every time new forecast information is supplied. Is possible.

  The calculation of the error may be performed by a plurality of combinations of a plurality of forecast information longer than the procurement lead time and the corresponding order results, or a plurality of combinations of one forecast information and the corresponding order results. You can go on. According to the former method, even when abnormal values are scattered in the forecast information or the like, it is possible to suppress the influence and calculate the order quantity with higher accuracy. On the other hand, according to the latter method, it is possible to calculate a highly accurate order quantity even in a case where the reception of forecast information is irregular.

In the former case, assuming that the average value of errors is A, the standard deviation of errors is σ, and the required safety factor is k, the correction amount S is
S = kσ-A
It may be calculated by On the other hand, in the latter case, assuming that the average value of error is A, the standard deviation of error is σ, the sum of procurement lead time and ordering interval is T, and the required safety factor is k, the correction amount S is

によって算出すればよい。

It may be calculated by

  If the correction amount becomes a negative value, this may be handled as zero. According to this, even when the actual demand greatly exceeds the forecast information, it is possible to effectively prevent the out of stock.

  As described above, in the present invention, since the current forecast information is not used as it is, but is corrected using the average value of errors and the standard deviation of errors, it is possible to perform highly accurate order prediction. Thus, it is possible to effectively suppress the occurrence of a shortage or an excessive inventory.

  Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

  FIG. 1 is a block diagram showing a configuration of an inventory management system according to a preferred embodiment of the present invention.

  As shown in FIG. 1, the inventory management system 100 according to the present embodiment includes a processing unit 110 that controls the operation of the entire inventory management system 100, a storage unit 120 that stores programs, data, and the like, which will be described later, and calculation of an order quantity. Are provided with an input unit 130 for inputting necessary data and an output unit 140 for outputting the calculation result of the order quantity.

  A controller IC such as a CPU (Central Processing Unit) can be used as the processing unit 110, and a hard disk device, a semiconductor memory, or the like can be used as the storage unit 120. As the input unit 130, a keyboard, a mouse, and various interfaces that can take in data sent online from a customer by electronic data exchange (EDI) can be used. For example, a display device such as a CRT or a printer can be used. Thus, the configuration of the inventory management system 100 is the same as that of a general personal computer. Therefore, the inventory management system 100 according to the present embodiment can be configured using a general personal computer.

  The storage unit 120 includes a program storage unit 121, a forecast storage unit 122, an order record storage unit 123, an error average value storage unit 124, an error standard deviation storage unit 125, an inventory amount storage unit 126, and an order remaining storage unit 127. And corresponding programs and data are stored. That is, the “stock management program” is stored in the program storage unit 121, and includes a forecast storage unit 122, an order record storage unit 123, an error average value storage unit 124, an error standard deviation storage unit 125, and an inventory amount storage unit 126. And the remaining order storage section 127 include “forecast information (F)”, “order record (D)”, “average value of errors (A)”, “standard deviation of errors ( σ) ”, the current“ inventory amount (Z) ”, and the“ in-order remaining (Y) ”currently in progress are stored. The plurality of forecast information stored in the forecast storage unit 122 constitutes a forecast transition table (see FIG. 3), which will be described later, and the plurality of order results stored in the order record storage unit 123 are the order results, which will be described later. A table (see FIG. 4) is formed. The plurality of remaining orders stored in the remaining order storage unit 127 constitutes an order remaining table (see FIG. 6) described later. Note that these storage units 121 to 127 do not have to be physically independent hardware resources, and can each be assigned a part of the storage area of one or more recording devices (such as a hard disk device and a semiconductor memory). It suffices that the allocated area may change dynamically.

  Hereinafter, the operation of the inventory management system 100 will be specifically described.

  FIG. 2 is a flowchart for explaining the operation of the inventory management system 100. The operation of the inventory management system 100 is performed by the processing unit 110 executing the “inventory management program” stored in the program storage unit 121.

  First, when the latest forecast information is input from the outside via the input unit 130 (step S10), this is stored in the forecast storage unit 122, and the forecast transition table is updated (step S11).

  FIG. 3 is a diagram showing the structure of the forecast transition table 10.

  As shown in FIG. 3, the forecast transition table 10 is assigned a scheduled order week and a week in which forecast information is received (referred to as “forecast reception week” in this specification). The forecast information is registered in each cell. In the example shown in FIG. 3, the interval for receiving forecast information and the ordering interval are both in units of one week. For example, a series of forecast information received in the 0th week includes the forecast information from the 1st week to the 3rd week. A series of forecast information composed of cast information and received in the first week is composed of forecast information from the second week to the fourth week.

Each forecast information is “a” for the forecast reception week and “b” for the planned order week.
F (a, b)
It is represented by For example, in the series of forecast information received in the second week, the order quantity for the fourth week is
F (2, 4)
It is written.

  In the example shown in FIG. 3, the ordering interval and the forecast receiving interval are both in units of one week, but these intervals may be 1 day, 3 days, etc., and these intervals are different from each other. It does not matter. Further, in the example shown in FIG. 3, an example is shown in which the period (in the present invention, “procurement lead time”) for procurement (or production) of the product is two weeks. It depends on the type.

In the present invention, a period obtained by adding the ordering interval to the procurement lead time is defined as an “inventory adjustment period”. In this example, the inventory adjustment period is 3 weeks. Therefore, the total value of the forecast information within the inventory adjustment period among the latest received forecast information can be regarded as the demand forecast value X. In other words, if the present is the end of the 14th week, the demand forecast value X is
X = F (14,15) + F (14,16) + F (14,17)
It is represented by However, since the forecast information is only expected information, if the order quantity is determined using the demand forecast value X as it is, there is a possibility that a shortage may occur or excessive inventory may occur. Therefore, in the present embodiment, as will be described later, the demand forecast value X is corrected and the order quantity is calculated based on this, as described later, in order to reduce the occurrence of shortage and excessive inventory as much as possible.

  FIG. 4 is a diagram illustrating the structure of the order record table 20 stored in the order record storage unit 123.

As shown in FIG. 4, the order record table 20 is composed of the order record for each week, and each order record has the order delivery week as “c”.
D (c)
It is represented by For example, the orders received in the fifth week are
D (5)
It is written. Here, “order delivery week” is a week in which all products ordered from customers are delivered.

  When the update of the forecast transition table 10 (step S11) is completed, the error between the forecast information and the order record corresponding thereto is calculated using values included in the forecast transition table 10 and the order record table 20. A plurality is calculated for each inventory adjustment period (step S12). In this example, since the inventory adjustment period is 3 weeks, it is only necessary to calculate the total value of the forecast for 3 consecutive weeks and the total value of the order results for 3 weeks corresponding to this for a plurality of combinations. .

More specifically, as shown in FIG. 5, the forecast information received in the first week is F (1,2), F (1,3), and F (1,4). F (1,2) + F (1,3) + F (1,4)
And the total value D (2) + D (3) + D (4) of the order results D (2), D (3) and D (4) corresponding to these
If the difference from is an error E (1),
E (1) = {F (1,2) + F (1,3) + F (1,4)}-{D (2) + D (3) + D (4)}
It can be expressed as. Similarly, when the difference between the total value of the forecast information received in the second week and the total value of the order results corresponding to these is the error E (2),
E (2) = {F (2,3) + F (2,4) + F (2,5)}-{D (3) + D (4) + D (5)}
It can be expressed as.

  Therefore, when the error E (n) is a positive value, it means that the forecast information is excessive, and when the error E is a negative value, the forecast information is too small. It means that there was.

  After calculating the error E (n) every forecast reception week, next, an average value A of these errors is obtained (step S13), and further, a standard deviation σ of these errors is obtained (step S14). ). The obtained error average value A is stored in the error average value storage unit 124 in the storage unit 120, and the obtained error standard deviation σ is stored in the error standard deviation storage unit 125 in the storage unit 120. Is done. The number of samples of the error E (n) used for calculation of the average value A of errors and the standard deviation σ is not particularly limited, but includes the most recent error (E (11) in this example) 5 to 5 About 20 samples may be used.

  FIG. 6 is a view showing the structure of the remaining order table 30 stored in the remaining order storage unit 127.

As shown in FIG. 6, the order backlog table 30 is configured by the order backlog for each week, and each order backlog is set to “d” as the order delivery date week.
Q (d)
It is represented by For example, the order backlog for week 15 is
Q (15)
It is written. Here, the “order delivery week” is a week in which all ordered products are received (or completed) at the beginning of the week. Therefore, the value of Q (15) indicates the quantity of products ordered three weeks before the inventory adjustment period, that is, at the end of the 12th week. This is because, in this example, the procurement lead time is 2 weeks, so it is necessary to determine the quantity of products to be completed every 3 weeks.

  Therefore, if the current is the end of the 14th week, the value of Q (15) represents the quantity of products that will be ordered at the end of the 12th week and will be received (or completed) at the beginning of the 15th week. Similarly, the value of Q (16) represents the amount of products scheduled to be placed at the end of the 13th week and received (or completed) at the beginning of the 16th week. That is, the sum of Q (15) and Q (16) indicates the amount of product in progress (the value of “Y” described later). Therefore, what should be calculated in the series of processes is the value of Q (17), that is, the amount of products scheduled to be placed at the end of the 14th week (current) and received (or completed) at the beginning of the 17th week.

When the necessary data is obtained in this way, the order quantity Q is calculated using these data (step S15). The order quantity Q is calculated as Y (Q (15) + Q (16) in this example), and Z is the current inventory quantity stored in the inventory quantity storage unit 126. Then
Q = X−Y−Z + S
This is done by calculating Here, “S” is a correction amount, and in this embodiment, S = kσ−A
Defined by “K” is a constant (safety factor) determined by the allowable shortage rate, and is obtained from a probability density function assuming that the variation follows a normal distribution. For example, if the allowable shortage rate is set to 1%, 2.5%, 5%, or 10%, k = 2.32,
k = 1.96,
k = 1.65,
k = 1.28
Should be set.

  That is, in the present embodiment, not only the product quantity Y and the inventory quantity Z in process are subtracted from the demand forecast value X based on the current forecast information, but also by correcting this, the order quantity Q with higher accuracy is obtained. Is calculated. As is apparent from the above equation, the correction amount S includes a component obtained by reversing the positive and negative values of the average value A of the error. Therefore, when the past forecast information is larger than the actual one, Therefore, the obtained order quantity Q decreases accordingly, and conversely, if the past forecast information is less than the actual one, the obtained order quantity Q increases accordingly. Thereby, it becomes possible to correct the demand forecast value X in consideration of the tendency of the past deviation amount (error).

  Further, the correction amount S includes a component (kσ) based on the standard deviation σ of the error. The component (kσ) based on the standard deviation σ of error is a value generally called “safety stock”, and the correction amount S increases as the standard deviation σ of error increases. If the standard deviation σ of the error is large, it means that the difference between the forecast information and the actual demand varies greatly, so it can be said that a shortage is likely to occur. For this reason, by increasing the correction amount S in accordance with the standard deviation σ of the error, it is possible to effectively suppress the shortage when the actual demand greatly exceeds the forecast information.

  On the other hand, a small standard deviation σ of error means that the difference between forecast information and actual demand is almost uniform. It is. For this reason, when the standard deviation σ of the error is small, the component (kσ) based on the standard deviation σ of the error is also reduced accordingly, and the correction amount S mainly depends on the average value A of the errors.

  When the order quantity Q is calculated in this way, the result (Q (17) in this example) is output via the output unit 140 (step S16), and based on this, the contents of the remaining order storage unit 127 ( The remaining ordering table 30) is updated (step S17), and the series of processing ends.

  Thus, according to the present embodiment, the in-process product quantity Y and the inventory quantity Z are subtracted from the demand forecast value X based on the current forecast information, and the error average value A and the error standard deviation σ. Since this is corrected based on this, it is possible to effectively suppress the occurrence of a shortage or an excessive inventory. In addition, in the present embodiment, since the error E (n) is calculated in units of the inventory adjustment period (in this example, 3 weeks), even if the abnormal values are scattered in the forecast information or the like, It is possible to suppress the influence and calculate the order quantity Q with higher accuracy.

In the present embodiment, the correction amount S is set to S = kσ−A
In this case, the correction amount S may be a negative value. Such a case occurs when the forecast information is almost uniformly larger than the actual demand, and it is possible to effectively suppress the occurrence of excessive inventory. On the other hand, if the correction amount S is a negative value, if the actual demand greatly exceeds the forecast information, there is a high possibility that a shortage will occur. This problem can be solved by treating the correction amount S as a negative value as zero.

  That is, as shown in the flowchart of FIG. 7, after obtaining the error average value A and the error standard deviation σ (steps S13 and S14), it is determined whether or not the correction amount S obtained based on these is zero or more. After the determination (step S20), if the correction amount S is zero or more, this is used as it is (step S21), and if the correction amount S is less than zero, that is, a negative value, it is converted to zero (step S22). The order quantity Q may be calculated (step S15). Such processing is also performed by the processing unit 110 executing the inventory management program stored in the program storage unit 121.

  According to this method, as described above, even if the actual demand greatly exceeds the forecast information, the possibility of the shortage occurring is reduced, but on the other hand, excessive inventory tends to occur. Therefore, it is preferable that the operator appropriately determines which method should be selected in accordance with the property of the customer or the product.

  The present invention is not limited to the embodiments described above, and various modifications are possible within the scope of the invention described in the claims, and these are also included in the scope of the present invention. Needless to say.

  For example, in the above embodiment, the error E (n) is calculated in units of the inventory adjustment period. However, in the present invention, it is not essential that the calculation unit of the error E (n) is the inventory adjustment period. You may use the difference | error of cast information and the order record corresponding to this. In this case, if the sum of the procurement lead time and the order interval (= inventory adjustment period) is T, the correction amount S is

によって算出することが好ましい。これによれば、分散の加法性定理により上記実施形態とほぼ同様の結果を得ることが可能となる。この方法は、フォーキャスト情報の受信が不定期であるケースにおいて特に有効である。

It is preferable to calculate by According to this, it is possible to obtain a result almost the same as that of the above embodiment by the additive additivity theorem. This method is particularly effective in a case where reception of forecast information is irregular.

  Even when this method is used, the correction amount S may be a negative value. When it is necessary to eliminate this, as described above, when the correction amount S becomes a negative value, it may be handled as zero.

1 is a block diagram showing a configuration of an inventory management system 100 according to a preferred embodiment of the present invention. 4 is a flowchart for explaining the operation of the inventory management system 100. It is a figure which shows the structure of the forecast transition table. It is a figure which shows the structure of the order record table 20. FIG. It is a figure for demonstrating the calculation method of error E (n). It is a figure which shows the structure of the order remaining table. It is a flowchart for making the correction amount S always a positive value.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Forecast transition table 20 Order reception result table 30 Order remaining table 100 Inventory management system 110 Processing part 120 Storage part 121 Program storage part 122 Forecast storage part 123 Order reception result storage part 124 Average value storage part 125 Error standard deviation storage Unit 126 inventory amount storage unit 127 remaining order storage unit 130 input unit 140 output unit

Claims (9)

At least an input part for inputting forecast information;
A forecast storage unit for storing the forecast information;
An order record storage unit for storing an order record;
A processing unit that calculates the order quantity based on the forecast information stored in the forecast storage unit and the order record stored in the order record storage unit;
An output unit for outputting the order quantity,
The processing unit calculates an error of the past forecast information from one or two or more consecutive forecast information and an order record corresponding thereto, and an average value of the errors and a standard deviation of the errors And the order quantity is calculated by correcting current forecast information based on at least the average value and the standard deviation.   The processing unit calculates a correction amount based on the average value and the standard deviation, adds the correction amount to the current forecast information, and calculates the order quantity by subtracting the ordered quantity and the inventory quantity. The inventory management system according to claim 1, wherein:   The inventory management system according to claim 1, wherein the processing unit calculates the error from a plurality of forecast information longer than a procurement lead time and an order record corresponding thereto.   The inventory management system according to claim 1, wherein the processing unit calculates the error from one forecast information and an order record corresponding to the forecast information. When the average value is A, the standard deviation is σ, and the required safety factor is k, the processing unit calculates the correction amount S,
S = kσ-A
The inventory management system according to claim 3, wherein the inventory management system is calculated by: When the average value is A, the standard deviation is σ, the sum of procurement lead time and ordering interval is T, and the required safety factor is k, the correction amount S is

The inventory management system according to claim 4, wherein the inventory management system is calculated by:   The inventory management system according to any one of claims 2 to 6, wherein when the correction amount becomes a negative value, the processing unit treats the correction amount as zero.   A step of calculating an error between a past one or two or more consecutive pieces of forecast information and a corresponding order record for a plurality of combinations, and at least an average value of the plurality of errors obtained and a plurality of errors obtained And a step of calculating the order quantity by correcting current forecast information based on the standard deviation of the inventory management method. On the computer,
Calculating an error between a past one or two or more consecutive forecast information and an order record corresponding thereto for a plurality of combinations;
Calculating the order quantity by correcting the current forecast information based on at least the average value of the plurality of errors obtained and the standard deviation of the plurality of errors obtained. Management program.

Images